Congestion Control Protocols in Wireless Sensor Networks
Congestion Control Protocols in Wireless Sensor
Networks:A Survey
Mohamed Amine Ka?,Djamel Djenouri,Jalel Ben-Othman,and Nadjib Badache
Abstract—The performance of wireless sensor networks(WSN) is affected by the lossy communication medium,application diversity,dense deployment,limited processing power and stor-age capacity,frequent topology change.All these limitations provide signi?cant and unique design challenges to data transport control in wireless sensor networks.An effective transport pro-tocol should consider reliable message delivery,energy-ef?ciency, quality of service and congestion control.The latter is vital for achieving a high throughput and a long network lifetime. Despite the huge number of protocols proposed in the literature, congestion control in WSN remains challenging.A review and taxonomy of the state-of-the-art protocols from the literature up to2013is provided in this paper.First,depending on the control policy,the protocols are divided into resource control vs.traf?c control.Traf?c control protocols are either reactive or preventive (avoiding).Reactive solutions are classi?ed following the reaction scale,while preventive solutions are split up into buffer limitation vs.interference control.Resource control protocols are classi?ed according to the type of resource to be tuned.
Index Terms—Wireless sensor networks,transport protocols, congestion control,contention,resource control,traf?c control.
I.I NTRODUCTION
A WIRELESS sensor network(WSN)is a set of tiny nodes
that are equipped with embedded computing devices interfacing with sensors/actuators.They generally use short-range wireless transmitters and they act autonomously-but cooperatively-to route data,hop-by-hop towards a central node called sink,or base station.
A WSN comprises a large set of distributed nodes over a wide geographical(indoor or outdoor)area to monitor a physical or environmental event[1].With the emergence of IoT(Internet of Things),WSN becomes more and more attractive by their integration in a real world of interconnected objects through internet[2].As IoT consists of the perception and transmission of information for everything in many forms [3]–[5],sensing is the axis of concepts related to this paradigm like M2M(Machine To Machine)and CPS(Cyber Physical Systems)[6]–[11].
CPS tries to assist the interaction between the physical world and the virtual one through the integration of sensing, communication,computing and control,while the interfacing Manuscript received April23,2013;revised September19,2013,December 18,2013and January13,2014.
M.Ka?, D.Djenouri,and N.Badache are with the DTISI,CERIST Research Center,Algiers,ALGERIA(e-mail: {ka?,ddjenouri,badache}@mail.cerist.dz).M.Ka?and N.Badache are also with the USTHB,University Houari Boumediane,Algiers,ALGERIA. J.Ben-Othman is with the Laboratoire L2TI.Universit′e de Paris13,Paris, FRANCE(e-mail:jalel.ben-othman@univ-paris13.fr).
Digital Object Identi?er10.1109/SURV.2014.021714.00123of M2M systems and WSN permit to take decisions with limited human intervention by emphasizing on the commu-nications among machines and the practical applications to make appropriate actions[6]–[11].
Some typical applications of WSN includes telemedicine monitoring,intelligent transportation,home automation,fac-tory monitoring,energy conservation,target tracking and environmental monitoring,etc[1],[12]–[15].Traf?c patterns in sensor networks can be derived from the monitored physical processes.These applications might be interested in different sensory data and therefore create different requirements in terms of QoS(Quality of Service)and reliability. Further,depending on speci?c applications,the delivery of upstream traf?c can be event-driven,continuous,query-driven, and hybrid.These types of applications are presented in the following.
Event-based applications:in this category,the network load is light but it unpredictably becomes active in response to a detected event.Depending on the application,the gen-erated data may be large.For example,in the battle?eld surveillance application,each node senses its surrounding in a continuous manner.When an event is detected(a tank entry), every node sends its samples to the base-station which can result in congestion[16].Even the information generated at event happening causes congestion,its importance is vital for application?delity.In practice,different combinations of traf?c density derive from event-based applications.Some ap-plications generate light occasional traf?c from small regions, while others generate large frequent traf?c across the covered sensing area[17].
Continuous sensing applications(Time-driven):some critical applications require continuous sending of sensing values to get real time values,e.g.,nuclear stations monitoring. If the load of the network does not allow for continuous transmissions,periodic sensing can be used,but with an adequate periodicity that satis?es the application requirements. Query-driven applications:contrary to event driven appli-cations where the sensing nodes trigger the sending after the event detection,in query-driven applications it is the sink that invokes and queries sensing nodes to answer.
Hybrid applications:This kind of applications will be common in the future.In such applications,often bulk data is generated in addition to the constantly sensed data.For example,in structural health monitoring,each sensor measures structural vibration continuously at a certain rate.When the sensors detect a signi?cant anomaly,they generate and send out data at a much higher rate[18],which will lead certainly to congestion happening.
1553-877X/14/$31.00c 2014IEEE
Congestion occurs when the traf?c load exceeds the avail-able capacity on node level(buffer over?ow)or link level (interference or contention)[19].The delivery of traf?c,even being well regulated,is hindered by the poor and time-varying channel quality,asymmetric communication channels, the need of multi-hop forwarding,and the hidden terminal problem[20],which make the congestion being severe.In the case of traf?c load?uctuation,a high degree of unfairness is remarked at remote nodes,besides the previous problems. These circumstances lead that congestion causes the waste of the nodes’energy[21]–[23],but the most serious is that it degrades the event detection reliability[24].
Many papers dealing with transport layer issues have been published,e.g.transport protocols[16]–[18],[20],[24]–[77], comparative studies[3],[78],[79],cross-layer design that include the transport layer[80]–[82].
However,no standard transport layer protocol for WSN exists despite many efforts conducted by the IETF[5],[83]–[90]to adapt protocols of different layers in the context of 6LoWPAN to be suitable for WSN environment.
In[4],[19],[91]–[100],authors present surveys for trans-port layer protocols or congestion control based protocols. None of these surveys provide a deep and comprehensive taxonomy,neither cover a large category of protocols.This paper reviews various existing techniques for detecting and controlling congestion.The rest of the paper is organised as follows,in Section II,congestion control paradigms are pre-sented with a discussion on their strategies.Section III presents some evaluation parameters to evaluate congestion control protocols.Some state-of-the-art protocols are presented in Section IV and V following our classi?cation presented in section II.Section VI concludes the paper and presents some future works.
II.C ONGESTION C ONTROL P ARADIGM
Beside the application type,the?ow type is of high importance to guide a real congestion control.Flow types may include a single packet,few packets,a large number of packets,which require light control,medium level control, and tight control,respectively.When a large number of nodes transmit information,their?ows will cross at intermediate nodes.This high number of sources increases the congestion but helps improving the reliability.For example,in tree architectures,every intermediate node can suffer from con-gestion causing packet loss,which in turn decreases network performance and throughput and cause energy waste.It is very dif?cult to predict the intersection points due to network dynamics(addition or removal of sensors or a change in the report rate),variability in radio channel quality over time. All these can transform uncongested parts of the network to under-provisioned and congested regions[20].The area around the intersection will become a hotspot and there is a possibility of congestion(buffer over?ow)and contention (links interference).For these reasons,a congestion control algorithm for data packet transmission is necessary. Contention-based Congestion:when many nodes within range of one another attempt to transmit simultaneously,losses occur due to interference and packet loss is engendered.This reduces the throughput of all nodes in the area[101].If the packet generation rate is suf?ciently small,simultaneous transmission becomes independent of the rate.Rather,it depends on the exact time generation of the packet.Explicit local synchronization(or also named phase shifting)among neighbors can reduce this type of loss[35],but it cannot eliminate the problem as non-neighboring nodes can still interfere(hidden nodes).The contention may happen between different?ows in the same area,and between different packets of the same?ow,especially in the case of high density networks.Consequently,the nodes’channel capacity becomes time-variant.
Buffer-based Congestion:each node uses a buffer for the packets waiting to be sent.The over?ow of this buffer causes congestion and packets loss.This is due to high reporting rate that varies in time due to dynamic channel conditions. The many-to-one nature(or converge cast)of WSNs causes congestion,in addition to the other causes shared with general wireless networks.
In[80],it is shown that when using large buffer sizes,the network load increase dramatically harms the event reliability, due to the limited capacity of the shared wireless medium. When buffer size is reduced,event reliability can be improved to some extent.For low buffer size values,buffer over?ows lead to a larger number of packet losses but result in lower channel contention and lower end-to-end packet latency values compared to those values of higher buffer sizes.This result is opposite to the conventional thought that limited storage always leads to performance degradation.This property is advantageous for real time applications.
The[80]study also shows the effect of maximum retrans-mission limit.Although moderate increase in this limit has a signi?cant difference with low retransmission values,the excess in retransmission does not have positive impact on the overall network reliability.
The congestion control functionality follows,in general, three steps starting by its detection,which will be noti?ed to the concerned node,so that an appropriate control will be taken.The following subsections treat in details these functionalities.
A.Congestion Detection Strategies
In literature[16]–[18],[20],[24]–[78],many congestion detection mechanisms are used and tested.The most used are:packet loss,queue length,packet service time,the ratio between packet service time and packet inter-arrival time, delay.In many cases,a single parameter cannot indicate congestion accurately.
Packet loss:It can be measured at the sender if ACKs (Acknowledgements)are used;this suggests reliability to be ensured by the protocol[28].It can also be measured at the receiver with sequence numbers use.Further,CTS(Clear To Send)packet loss can be used as congestion indication,as in [67].
Not overhearing the parent’s forwarding on the upstream link,by a child node over the downstream link,can be used as an indication for packet loss[35],as well.The time to repair losses(if reliability ensured)can be used as a congestion indication[51].Loss ratio is also used in some protocols[52],
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY3
[60].However,the losses can be caused by wireless errors rather than packets collision.
Queue length:as each node has a buffer;its length can serve a simple and good indication of congestion.In[20], [25],[29],[33],[36],[39],[42],[45],[46],[49],[50],[53], [56],[58],[59],[63],[69],a?xed threshold is used and the congestion is signalled as soon as the buffer length exceeds this threshold;while in[32],the remaining buffer length from the overall size is used.
In[37],[43],the difference between the remaining buffer and the traf?c rate is used as congestion indication.The traf?c rate represents the excess rate,which is the difference between the output rate and the sum of sourced and forwarded rates. In[40],[62],the buffer length is used in addition to the difference of output and input time,which is quite similar to output and input rate.In[37],[61],buffer length and capacity of the node are used together.
The number of non-empty queues can indicate congestion level[65].When there is a congestion,this number is larger than0.This number increases with network load.If the link layer applies retransmissions,link contention will be re?ected through buffer length[49].
Queue length and Channel load:In case of increase in packets collision,and after several unsuccessful MAC (Medium Access Control)retransmissions,packets are re-moved.Consequently,the decrease in buffer occupancy due to these drops may mean the absence of congestion when only buffer state is used for congestion detection.
Therefore,for accurate congestion detection,a hybrid ap-proach is required using queue length and channel loading as a congestion indication[16],[17],[27],[34],[72].Channel busyness ratio or channel load is the ratio of time intervals when the channel is busy(successful transmission or collision) to the total time.
In[16],the authors use the busyness channel ratio,similarly to channel load,but apply it to a subset of nodes,and queue length for another set of nodes.The node activates channel monitoring only when it receives a packet to forward. Therefore,there is no overhead to measure channel loading [17].
DST[54]uses node delay and buffer length as an indication of congestion.It depends on the used rate and channel load. Channel busyness ratio and throughput measurement: In[66],the authors use throughput in addition to channel busyness to take into account the effects of hidden nodes problem in multi-hop environment.The throughput quanti?es the number of successful transmissions.
Packet service time:as the inverse of packet service rate, it is the interval between packet arrival at the MAC layer and its successful transmission.It covers packet waiting,collision resolution,and packet transmission time at the MAC layer [18].
This value changes regarding to the queue length and channel load,so it is just another measure of them.It also represents the one hop node delay,as in[102].In[47],the end-to-end delay is calculated in a similar way.
But using only the service time may be wrong when the incoming traf?c is equal or less than the outgoing one through the overloaded channel[70].
Packet service time and queue length:in[18],service time is used to continuously adjust the rate at which children send their packets.The diminution is performed based on the queue length.
Ratio of packet service time and packet inter-arrival time(scheduling time):A scheduler between network and MAC layer gives packets from network queues to the MAC layer.The scheduling time quanti?es the number of packets scheduled per time unit.This ratio indicates both node level and link level congestion[24].In[38],[44],[55],the ratio of rates instead of times is used and authors named it packet service ratio.In[73],the difference between service and scheduling rates is used instead of the ratio.In[31],in addition to the precedent ratio,buffer length is also used to detect congestion.
Delay:In general,it quanti?es the necessary time since the packet generation,at the sender,until its successful reception at the next hop receiver[54],[102],or end point receiver[47]. It can also be calculated as a part of the total delay,as in ATP [30](queuing delay).However,the use of delay as a measure of congestion may be misleading.The largest amount of delay is caused by the sleep latency due to the use of duty-cycling at the MAC layer[103].
B.Congestion Noti?cation
When congestion is detected,the information must be propagated to allow taking an appropriate decision.This infor-mation can be as small as a single bit(congestion noti?cation bit)[20],[25],[27]–[29],[54],[72],as rich as a new data rate information[47],other values helping on the calculation of the new rate[18],[24],[30]–[33],[38],[41],or even the actual congestion level[31],[37],[39],[43].
Congestion information can be transmitted in data packets header(implicit noti?cation)[18],[20],[24],[25],[27],[29], [30],[33],[37]–[39],[42],[44],[45],[49],or in separate control messages(explicit noti?cation)[16],[17],[28],[31],
[34],[36],[41],[47],[52],[62],[63].
C.Congestion Control
For some applications,applying the same type of congestion control at all nodes would not ameliorate the throughput.For example,in event-based applications with limited messages per event,congestion control by traf?c regulation at the sources does not apply.Phase shifting may serve as the appropriate alternative in this case[35].However,intermediate nodes can and have to regulate the rate at which they forward the event packets to the sink when a bottleneck happens,and the rate control will take place at intermediate nodes. When the event is reported in several messages(e.g.multi-media applications),congestion control extend to rate control at the sources.In this case,phase shifting is useful.
The congestion control cannot be decoupled from the MAC protocol,and adequate protocol should?rst be used to avoid congestion.In applications where the event cannot be known a priory,random access contention based MAC protocols are necessary(CSMA”Carrier Sense Multiple Access”-based). In continuous and periodic applications with high rate a TDMA”Time Division Multiple Access”-like scheme is more appropriate.
4IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
Fig.1.Traf?c control classes.
Congestion control and fairness are two different but related aspects[69].Fairness is the ability to ensure all data sources have equal access to the network bandwidth.
In a WSN monitoring and control applications,events may have different priorities and need to be reported at different rates.In this case,it is subject to weighted fairness instead of equality fairness.This paradigm is realized in different ways. In[20],[69],a token bucket scheme is used and each node can transmit only if it has a token.In[16],[18],[24],[29], [31],[35],[37]–[39],[42],[47],[51],[53]–[55],[58],[72], [73],the exact rate partitioning is used for both equal and weighted division,while in[40],[41],[49],[60]–[62],[68], [70],[76]scheduling is applied in addition to rate partitioning. Different metrics can be used for priority de?nition,depending upon application needs,e.g.,event,node,region,or time[62]. In[24],[37],[47],[58],[69]the priority is de?ned at the node level according to the importance of its data.Further,routed packets(packets at intermediate nodes)are prioritized over sourced packets.In[38],[39],[42],[44],[53],[55],[62],the priority is de?ned at the data or event level.At the same node, different sensed events have different priorities.RAP[26]that has been proposed for query and event applications,gives more priority for packets originated from remote nodes from the sink over near sources,using packet Velocity Monotonic Scheduling(VMS).It chooses the forwarding order according to the distance and the end-to-end deadline.The priority is the ratio of the distance to the destination and the deadline value. RAP uses prioritized queues at each node.As packets from different prioritized senders may interfere on the same radio, RAP applies prioritized MAC to avoid collisions between different senders.Nonetheless,RAP does not present any rate control.It also requires localization,which comes at additional overhead.DST[54]uses the remaining deadline time as the packet priority.A packet gets higher scheduling priority with a decreasing value.In[74],a system rules is used to map data type to a transmission rate and a traf?c class scheduling using phenomena’s priority and its location.
In end-to-end congestion control protocols,it is the end sink responsibility to detect the congestion[17],[25],[28],[30], [47],[66].The sink may just receive the congestion indication and applies the control through an exact rate adjustment for each source[29],[54],[68],[76].It can also be responsible for both the detection and the control[51],[52].End-to-end control has a long latency,as at least one Round-Trip-Time (RTT)is needed to detect congestion.If congestion is transient and feedback latency is important,the noti?cation may be much later than the congestion period.Thus,the solution may be inappropriate to WSN showing transient congestion[18]. The hop-by-hop back-pressure protocols[16]–[18],[20],[24], [27],[31]–[46],[49],[50],[53],[56]–[64],[67],[69]–[74], [102]react immediately to the congestion at the intermediate node,but they need more control at these nodes.
Many congestion control algorithms for WSNs are designed across the transport and MAC layers(and even the network layer)for ef?cient congestion detection and control.The cross layer design,by the interaction between different layers, that helps in enhancing sensor networks protocols has been investigated in[81].[75]Shows through a case study how the cross layer helps in minimizing end-to-end delay,where [80],[82]investigate the usefulness of the cross layer design to congestion control.
Upon congestion detection,and depending on the appli-cation strategy,either traf?c control is applied by throttling the node rates,or resource control is used by exploiting idle resources.
1)Traf?c Control:The regulation or rate change of packets sending after the Congestion Noti?cation(CN)can be assured in different ways.
The AIMD(Additive Increase Multiplicative Decrease) scheme or its variants are usually applied[17],[35],[49], [66],[67],especially when using a single CN bit.In[20], [28],[32],[56],[69],[74],temporarily halt of packet sending is used to permit the congested nodes to empty their queues. The no embodying of event reporting nodes number,when calculating the increment/decrement factor of rate change in AIMD schemes,leads to inappropriate values[62].
On the other hand,if detailed congestion information is available,exact and accurate rate adjustment can be imple-mented[16],[18],[24],[29]–[31],[37]–[39],[42],[43],[47], [50]–[55],[58],[60],[61],[64],[72],[73].For adjusting the reporting rates,either a sink-based or in-network based solutions are used.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY5
In[16],[46],packets are dropped not to propagate con-gestion.However,no congestion noti?cation is used,which causes sources to waste their resources by continuing submis-sion of traf?c that will not achieve its?nal destination.
The traf?c control can be performed in an avoiding or reacting manner,both of them can be based on interference or buffer over?ow control.Avoiding interference is through scheduling the transmissions to avoid collisions[60],[68], [88],or partitioning the rate to prevent exceeding the inter-fering nodes capacity[49],[50],[61],[76].
The schemes that avoid the buffer over?owing are based upon limiting the sending[32],[56],[69].
With the reacting-based traf?c control,both interference and buffer over?ow are mitigated.The mitigation is based on either an organized hierarchic rate control,or individual control,so that only the rate of the concerned node is adjusted.
The hierarchic-based organized rate control is applied through a rate-based scheduling[40],[62]or through an equal (or weighted)rate partitioning,without any schedule[16], [18],[20],[24],[29],[35],[37],[38],[42],[43],[47],[53]–[55],[58],[64],[70],[72],[73].When applying individual traf?c control,it is assured with either exact rate control[30], [39],[51],[52],[57],[66]or with a coarse grained control [17],[28],[31],[46],[67],[74].Figure1illustrates explicitly our traf?c control classi?cation.
2)Resource Control:Resource provisioning techniques could be used when rate control methods cannot meet ap-plication’s requirements,since reducing source traf?c during a critical situation may violate application requirements.It is better to increase the capacity by turning-on more resources in order to face the high resulted traf?c[34].In presence of congestion,routing methods that employ alternative routes can be used to send data around the congested area[22],[33], [102],[104].Load balancing the traf?c between congested and uncongested routes upon congestion reaction has been used in[34],[44],[71],where preventive load balancing with an interference avoiding-based scheduling was used in[34], [63].
Resource control can be assured using clustering and mul-tiple radios[27].The cluster-heads are equipped with two radios;one is used to exchange packets with member nodes (short distance),where the other one is used to communicate with other cluster-heads and the sink(long distance).Some protocols adapt transmission power to ensure long distance sending[45].Other protocols assure resource control by adapting duty cycling parameters,to balance between energy-ef?ciency(in low traf?c scenarios)and traf?c?delity[41]. Figure2highlights resource control classes.
Some protocols neither apply traf?c nor resource control upon congestion detection,but rather they apply aggregation strategies.Prioritized MAC schemes can also be used to give the congested nodes a prioritized channel access(making back-off length dependent on local congestion)permitting draining their buffer.
The choice of the control to be applied must answer appli-cations requirements but at the same time has its consequences on the network lifetime.In[79],a comparative study between traf?c control(SenTCP[31])and resource control(HTAP [33])protocols in event-based networks has been presented.
Fig.2.Resource control classes.
Parameters of comparison include:average node energy con-sumption,network lifetime,number of packets drops and source data rate.The results show that alternative path creation algorithms assure a high network lifetime while keeping the data rate stable,whereas data rate reduction algorithms present less power consumption per node and minimal packet drops. While spreading the traf?c through different paths reduces the congestion,it increases the contention because of the crossing of these multiple routes toward the sinks.
In section IV,a resume of traf?c control will be presented regarding the above classi?cation,while in section V resource control protocols will be discussed.Through the literature, these protocols’performance is highlighted using simulations, experimentation or also by modelling their behaviour.This is done choosing speci?c metrics as evaluations parameters.The following section gives more details on theses parameters.
III.E VALUATION M ETRICS
After the conception of a congestion control protocol,it must be evaluated in the purpose to show its ef?ciency in the presence of overload traf?c.Through the literature,measures to evaluate the sensor networks performance under congestion are numerous.The measurement parameters allow comparing control strategies in speci?c cases.The commonest metrics used by the proposed protocols are:network ef?ciency,energy ef?ciency,sink received throughput,network fairness,and packet latency.
Network ef?ciency:quanti?es the energy wasted on trans-missions that do not deliver packets.The packets dropping cost varies depending on the distance from sink.
Energy ef?ciency:it is measured in joule(J).It includes energy spent in channel listening and packets transmissions and forwarding in the whole network.It is also measured per unit of successful communication or received packets[25], [34],[35],[45],[52],[55],[59],[67],[72].In[44],[53], [62],residual energy is used as the ratio of?nal energy to initial energy.In[16],[55],[69],energy ef?ciency has
6IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
been presented by the delivery ratio,which is de?ned in the following.
Energy Tax:the ratio between the total number of packets dropped in the sensor network and the total number of packets received at the sinks[17],[27],[33].
Packet Loss Ratio or Delivery Ratio:It is the ratio of the number of packets lost due to both buffer over?ow and bit-error[18],[28],[31],[32],[37],[38],[40],[47],[55], [61],[62],[102](respectively received[36]),to the number of packets generated.In[16],[33],[45],the number of packet drops is used,where in[73],the number of retransmissions per node is used.
Fairness:quanti?es the variation degree in sending rates.A fair allocation of bandwidth delivered to the base-station from each node over multiple hops is desirable[16],[18],[20],[30], [35],[46],[49],[56],[60],[66],[69].The weighted fairness regarding data priority is introduced in[37],[39],[70].In[62], node throughput is used as fairness guaranty.
End-to-end delay(packet latency):It is measured as the time taken by a packet to reach the base-station from the time it was generated[20],[25],[28],[32],[33],[39],[46],[47], [57],[61],[64],[66],[67].In[53],a per hop delay is used, where a weighted delay is used in[42],[70].
Control packet overhead:It quanti?es the number of packets used by the protocol[102],or the ratio to total packets [43],[59].
The total throughput at the sink:It is the number of successfully received packets during time unit[16],[24],[30], [31],[36],[37],[39],[46],[49],[50],[55],[66],[67],[70], [72],[73].In[51],[60],[64],network good-put is de?ned as the lowest observed packet reception rate at the base-station from any node in the network.
In[38],[42],the throughput is weighted in respect to data priorities.In[53],[59],the total number of packets received by the sink during simulation time is used. Instantaneous queue size:It shows the stability or?uctu-ation of queues[38],[44],[49],[50],[53],[61],[70].In[55], the weighted queues notion is used,where the weight of a queue is determined by the importance of the events associated to it.
Memory requirements:It is generally based on buffer length,the code length,and the number of the considered sensing units[55].
Fidelity index:It is the fraction of the number of packets targeted to be received by the application,to that properly received[34].
Fidelity Penalty:It is the delivery of the required number of data event packets within a certain time limit[17]. Generated rate(or source rate):It is the total number of data packets generated by the sources per second[16],[43], [73].
In[78],a comparative study has been carried out using some metrics and different WSNs topologies.Different congestion control policies have been used(traf?c control vs.resource control).
IV.T RAFFIC C ONTROL P ROTOCOLS
Existing transport protocols designed for Internet cannot be directly applied to WSNs as they either lack of reliability or ?ow control(UDP”User Datagram Protocol”),or have high control overhead and inappropriate reactions to wireless losses (TCP”Transmission Control Protocol”).These causes added to the speci?c WSNs characteristics are behind the motivation for new transport protocols.In TCP,ACK reception causes transmission window size increase.With a low-rate stream, this window in?ation is arti?cial and does not mean that the indicated capacity is actually available.When an event occurs causing a sequence sending of packets,TCP supposes that the large window is usable,which is misleading and causes packet loss[74].
In this section many traf?c control protocols are presented following the classi?cations presented in section II.
A.Equal or Weighted Rate Partitioning
In this category of protocols,congestion control is applied in a reactive manner,where nodes decrease their rates in response to congestion detection.This decrease is not applied independently to the detector node but in relation to its entire sub-tree,or even to the whole network.The decrease or the increase(when the congestion is eliminated)is performed by taking into account node priority,which results to an equal or weighted rate decreasing(respectively increasing).However, this control is completely decoupled with the scheduling.Table I,at the end of this section,summarizes some protocols of this class.
1)ARC[35]:Adaptive Rate Control treats contention in event and periodic applications by introducing a random delay (back off)at the application layer before transmitting packets (phase shifting).This way,it eliminates the hidden node problem without explicit control.
ARC uses packet loss as collision or congestion indication at each hop to adjust transmission rate of periodic applications. If the packet is successfully injected(overhearing parent for-warding),the node increases its transmission rate.Otherwise, it decreases its sourced rate and backs off for a phase change. An AIMD control is performed in a fair manner between sourced and routed traf?c,using the sub-tree deep.Prioritized fairness has not been envisaged with ARC.
2)ESRT[29]:In Event to Sink Reliable Transport,sensors change their sending rate using the sinks feedback regarding the reliability level or congestion detection.Every node sets a CN congestion noti?cation bit in the packets as soon as its buffer reaches a threshold.The sink periodically computes a new reporting rate for all sources based on reliability measurement,the received CN,and the old reporting rate.It broadcasts it with a high-powered ampli?cation.
ESRT running includes in?ve states,No Congestion Low Reliability(NCLR),No Congestion,High Reliability(NCHR), Congestion High Reliability(CHR),Congestion Low Reliabil-ity(CLR),and Optimal Operating Region(OOR).
In NCLR,the reporting rate is increased to reach an acceptable reliability,while in both NCHR and CHR,the reporting rate is decreased.In CLR,the reporting rate decrease is sharper.In OOR the reporting rate is unchanged for the next decision interval.
Treating different characterized regions(event priority,node density)in the same way will decrease the system throughput. Moreover,ESRT does not give attention to interference.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY7
3)EECC[72]:Energy Ef?cient Congestion Control pro-tocol designs a source rate congestion control.Each node adds its current weight,which is de?ned as the product of channel busyness ratio and the buffer occupancy,to the packet received from its children,and passes the packet to its parent. The sum of such weights is then used.When buffer size and channel busyness ratio reach their higher threshold,the node sets the congestion noti?cation bit in every data packet sent. By receiving this noti?cation,the parent calculates the new rate and informs its children nodes.
After the sink collects enough data,it uses a clustering algorithm to partition nodes according to the sending rates and data similarity.Nodes within the same cluster work alternatively following an established schedule to save energy. However,cumulating weights does not directly re?ect node channel busyness.The rate adjustment has not been clearly elaborated,as well as the rate sharing between children nodes.
4)FUSION[20]:uses three congestion control techniques: hop-by-by?ow control,source limiting scheme,and prior-itized MAC.It also uses buffer occupancy as a congestion indication level.When a node overhears a packet from its parent with the congestion bit set,it stops forwarding data to allow its parent emptying its queue.If the congestion persists, the hop-by-hop back-pressure reaches the source,which will consequently decrease its rate.By overhearing,each node determines the number of children nodes of its parent(N). The sending rate is regulated fairly between descendants using a token bucket scheme,where the node gains one token for N packet forwarded by its parent.In addition,FUSION uses a prioritized MAC scheme to give the congested nodes a prioritized channel access.
5)CCF[18]:Congestion Control and Fairness for Many to One Routing protocol modi?es its rate using the packet service time which is the period between sending the packet at the transport layer to the network layer and the reception of successful transmission noti?cation.Each node measures the average sending rate which is then divided by the number of children in its sub-tree.Each child compares this last with the rate sent from the parent to use and propagate the smaller one downstream by piggybacking in data https://www.wendangku.net/doc/7515472058.html,F also uses the queue length for requesting child nodes to reduce their rates if the de?ned threshold is reached.Maintaining a separate queue for every child is not memory ef?cient in dense networks.Further,in case of idle nodes or nodes with little traf?c,the remaining bandwidth is not allocated to active nodes.
6)PCCP:Priority-based Congestion Control Protocol[24] considers that sensor nodes may have different importance and need different throughput.It uses then a weighted fairness de?ned with a node priority index.PCCP uses a scheduler between network and MAC layers,as well as two queues at each node;one for sourced traf?c and another for transit traf-?c.It periodically detects congestion,using the ratio between packet service time and packet inter-arrival time at MAC layer.This ratio is used to achieve exact rate adjustment with priority-based fairness.PRA(Priority-based Rate Adjustment) algorithm is used to guarantee fairness between source and the sub-tree transit traf?c.PRA adjusts the scheduling and the source rate using the priority weight of the sourced data and the global priority(sum of sub-tree weights)of the node to control link and node level congestion.In low congestion scenarios,PCCP increases scheduling and source rate of all traf?c sources without priority index,and it implicitly uses the information about active nodes.While in case of high congestion,it decreases the sending rate of all traf?c sources based on their priority index.
7)MCCH,APRC,PHTCCP:In these protocols,it is sup-posed that nodes sense different events and the sink assigns different priorities to data according to its importance.Multi-path routing is used in MCCH[38](Multipath Congestion Control for Heterogeneous traf?c)[44].Both APRC[53] (Application Priority-based Rate Control)and PHTCCP[55] (Prioritized Heterogeneous Traf?c-oriented Congestion Con-trol Protocol)use single path routing.
Nodes dispose queues with different priorities for the dif-ferent events.The queues are scheduled according to the inter queue priority and each node adapts its scheduling and output rates.In MCCH,where each node has multiple parents;the sum of parents scheduling rates gives the total scheduling rate. MCCH and PHTCCP use the ratio of average packet ser-vice rate and packet scheduling rate as congestion indication (named as packet service ratio),similarly to PCCP,while APRC uses the average of queue lengths.
MCCH,PHTCCP and APRC apply a scheduler rate control when receiving a parent congestion indication or observing local measuring.They piggyback in the packet headers the scheduler rate and the number of child nodes.In addition to these parameters,MCCH piggybacks also the packet service rate,while APRC and PHTCCP adds the number of active child nodes,the average queue length of a node and its child nodes.PHTCCP uses traf?c priority based MAC protocol (differentiating inter-frame-spacing and back-off mechanisms) by assigning short IFS and back-off to the higher priority traf?c.
As MCCH and APRC do not apply any priority between children nodes for rate controlling adjustments,it is possible for a node with higher priority packets to have the same rate as other nodes having less priority packets.Further,queues scheduling at nodes are not elaborated.
In APRC,rate control de?nes priority depending on the actual queue length and not on the data.PHTCCP uses the same principle with application’s weighting.
Even routed packets have higher priority than sourced packets;their location should also be considered.Further,this priority will penalize shallow nodes on the tree compared to deep nodes.Piggybacking control information in every packet causes a huge overhead.
8)QCCP-PS,NCC:Yaghmaee et al.propose QCCP-PS [58](Queue based Congestion Control Protocol with Prior-ity Support)for multimedia sensor networks and NCC[42] (Novel Congestion Control Protocol)for vital real-time signs monitoring in biomedical sensor networks.A separate queue for each child and another for local traf?c are used,and the queue length is used as congestion indication.In NCC,packet class priority is applied within every queue.
When the queue length exceeds the de?ned max threshold, the rate is decreased,and it is increased once the length is bellowing another minimum threshold.Between the two
8IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
thresholds,the congestion index is related to queue length linearly.Periodically,each node calculates the sending rate of its child sources and its local traf?c source with considering the priority and the current congestion degree of the child nodes queues.This new rate is divided and sent to the child nodes,according to their total priority,i.e.the sum of the priorities of the sub tree rooted at the node,which is shared only between active nodes.The sending rate is the minimum of the parent assigned value and the local service rate.We think queue length alone is not suf?cient for calculating congestion and service time may be useful for determining contention level.
9)CCF2[73]:Congestion Control and Fairness protocol is a distributed congestion control algorithm proposed for tree based communications,using contention-based MAC that targets fair sending rate assignment.It is supposed that every node may have in?nite data to send.It periodically monitors its output and input traf?c rates.Based on the difference in these rates and the queued packets,it decides to increase or decrease the allocable bandwidth to sourced and forwarded traf?c.It is similar to PCCP[24]with regards to congestion detection,where the service rate is substituted with the output rate,and the scheduling rate with the input rate,but instead of calculating the ratio,CCF2calculates the difference between the https://www.wendangku.net/doc/7515472058.html,F2uses an AIMD-like scheme and shares the increase/decrease between children nodes.
10)UHCC and HCCP:Both UHCC[37](upstream hop-by-hop congestion control)and HCCP[43](Hybrid Conges-tion Control Protocol),use packet delivery rate and buffer size to detect congestion.Each node uses the difference between the remaining buffer size and the net?ow size to calculate and exchange its congestion degree.The net?ow size is the difference between the sum of sourced and upstream neighbors ?ows rates,and the?ows forwarded to downstream neighbors. Congestion happens when the sum of children and sourced rates is larger than the forwarding rate.It is controlled by reducing children and source rates.
When the congestion degree is high,HCCP regulates rates by giving more incoming rate to upstream nodes with more data to send.If the rate is not suf?cient,the congestion will be extended to other upstream nodes.
This rate sharing manner can be misleading as nodes with more data to send are not necessarily more important.In addition to this drawback,the protocol does not take the inter-ference problem into account.In UHCC,the rate adjustment is assured using traf?c priority ratio,which is the fraction between traf?c priority at source or child node and the total traf?c priority.It also uses the node congestion index to share the rate between the concerned nodes.
11)DST[54]:Delay Sensitive Transport protocol targets critical delay event applications where the late event noti-?cation at the sink leads to application failure.The event delay is the time between the event detection and the sink noti?cation.DST uses a Time Critical Event First(TCEF) scheduling with prioritized MAC to ensure delay bounds.It measures the elapsed time to update the remaining time to the deadline at each node and piggybacks it in event packets. With decreasing values,the packets get higher priority.It also de?nes event reliability as the number of received packets in a decision interval.If the packets number is below a threshold,the reporting frequency is increased.DST detects congestion using average node packets delay and buffer level. Average node delay measures the contention around the node which varies depending on the used rate and channel load. Congested node having delay or buffer values higher than a threshold informs the sink using the noti?cation(CN)bit in packet https://www.wendangku.net/doc/7515472058.html,ing reliability indicator and current network condition,the sink adjusts sensors reporting frequency,as in ESRT[29].
Neither DST nor ESRT try to avoid collision-based con-gestion,but they just decrease the source rates.No details for deadline attribution and TCEF scheduling have been given.
12)Prioritizing Information for QoS Control:This proto-col[47]prioritizes the sensed information based on its nature. It uses end-to-end packet delay to evaluate congestion and update the rate at the sink.The new rate and the congestion level are forwarded to the sources including the congestion level.Intermediate nodes store data for a prede?ned time related to the estimated total packet drop probability.However, the congestion index is not well used,and no detail is given on how to use priority in rate https://www.wendangku.net/doc/7515472058.html,ing end-to-end control may slow the appropriate control and west more time and energy,and it is prone to packet loss.
13)FACC[16]:Fairness-Aware Congestion Control pro-tocol is a rate-based protocol dividing intermediate nodes into near-source and near-sink based on the application and QoS requirements.Near-source nodes record a per-?ow state and allocate a weighted fair rate to passing?ows based on available bandwidth.While near-sink nodes use probabilis-tic removing algorithm based on queue occupancy and hit frequency.Near-sink nodes send warning messages to the near-source nodes once a packet is dropped.Consequently, near-source nodes compare the incoming rate of each?ow and the shared bandwidth to allocate a fair rate for passing ?ows,and notify the concerned sources to update their rate in an AIMD-like manner.FACC uses channel busyness ratio as congestion indication in near source nodes.The differentiation between near source and near sink nodes is not necessary and engender additional load.The use of busyness ratio and buffer occupancy together may potentially achieve a better performance.
14)WRCP:[64](Wireless Rate Control Protocol)tries improving the convergence time of a rate control using explicit capacity information in the purpose to overcome the long convergence time to the achievable rate,and frequent over capacity use that characterize AIMD schemes especially when the set of active?ows in the system is continuously changing, which occur specially with short?ows.WRCP applies a receiver capacity model in a tree like network by associating constant node capacities(one hop CSMA capacity),instead of links capacities,by assuming that the receiver capacity de-pends on the neighbors number rather than their transmission rates.WRCP models(by a linear equation)the relationship between the receiver capacity and passing?ows’rates.Each node divides its allowable rate between traversing?ows. 15)DPCC[70]:decentralized,predictive congestion con-trol protocol controls congestion in a hop-by-hop manner through an adaptive?ow and adaptive back-off interval se-
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY9
TABLE I
E QUAL OR W EIGHTED R ATE P ARTITIONING PROTOCOLS.
Protocol Congestion Detection Congestion
Noti?cation Congestion Con-
trol
H-
by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation Parameters Compared
with
ARC[35]Packet loss-Phase Shifting+
AIMD Control H-by-H Event and
Periodic
No Implementation Fairness,Energy Ef?ciency Different
CSMA
Schemes
ESRT [29]Queue Length Bit in the
header
Rate adjustment E-to-E Event No Simulation Normalized Reliability,Average
Power Consumption
Alone
EECC [72]Cumulated Channel
Busyness*Buffer
length
Information
in header
Rate control H-by-H Continuous,
Event
no Simulation
NS2
Throughput,Energy Consump-
tion,Packet Delivery Ratio
CODA,
ESRT
FUSION [20]Queue Length Bit in the
header
Stop Sending,pri-
oritized MAC
H-by-H Hybrid No Experimentation Network Ef?ciency,Node Im-
balance,Aggregate Sink Re-
ceived Throughput,Fairness,
Packet Latency
NCC,Rate
Limiting
CCF[18]Packet Service Time
+Queue Length Information
in header
Rate adjustment H-by-H Event Yes Simulation,
Implementa-
tion
Fairness,Number of Retrans-
missions Per Packet,Packet
Generation Rate
Different
components
of the
protocol
PCCP [24]Packet Service Time/
Packet Inter-arrival
Time
Information
in header
Exact Rate Con-
trol
H-by-H Event,
Continu-
ous
No Simulation Normalized System Throughput CCF
MCCH [38]Packet Service Ratio Information
in header
Traf?c Control H-by-H Continuous no Simulation Packet Drop,Queue Length,pri-
ority based Throughput
alone
APRC [53]Queue Length Information
in header
Rate adjustment H-by-H Continuous No Simulation Scheduling Rate,Queue Length,
Node Delay,Packets Received,
Residual Energy,Throughput
CCF,No
Congestion
Control
PHTCCP [55]packet service ratio Information
in header
Rate adjustment H-by-H Periodic,
event
no Simulation
NS2
Packet Drops,Weighted Queue
Length,Memory Requirements,
Throughput,Energy Ef?ciency
CCF,No
Congestion
Control
QCCP-PS,NCC [42],[58]Queue Length Information
in header
Rate Adjustment H-by-H Multimedia No Simulation Throughput,Achieved Priority,
Packet Loss Probability
PCCP,CCF
CCF2 [73]Service rate-
scheduling rate-buffer
length
Information
in header
Periodic Rate
Control
H-by-H continuous no Simulation Goodput,Fairness,Data Gener-
ation Rate,Link Layer Retrans-
missions
alone
HCCP [43]Remaining Queue
Length
Feedback
msg,
Information
in header
Rate Control H-by-H continuous No Simulation,
NS2
Source Rate,Control Overhead AFA,BB
UHCC [37]Remaining Queue
Length-Excess
Traf?c Rate
Information
in header
Rate Control H-by-H Periodic No Simulation Throughput,Fairness,Loss Ra-
tio
PCCP,CCF
DST[54]Node Delay+Queue
Length Bit in the
header
Rate Adjustment E-to-E Event No Simulation
NS
Convergence Time,Energy
Consumption
ESRT
Prioritizing for QoS [47]E-to-E delay+queue
length
Feedback
msg
Rate Control E-to-E Event,
Periodic,
Continu-
ous
Yes Simulation
NS2
E-2-E Latency,E-2-E
Throughput,Data Loss,
Priority Achieved
Different
components
of the
protocol
FACC [16]Channel busyness+
queue length
Feedback
msg
Rate Control H-by-H continuous No Simulation
NS2
Packet Loss,Source Rate,Fair-
ness,Throughput
CODA,
NCC
WRCP [64]Information
in header
Rate Control H-by-H No Implementation E-to-E Delay,Goodput IFRC
DPCC [70]Queue Utilization and
Channel Quality
Information
in ACK
header
Rate Control,
Adaptive back off
H-by-H Simulation
NS2
Queue Utilisation,Throughput
Network Ef?ciency
CODA
lection schemes.It detects congestion using queue utilization and channel quality.The receiver regulates the rate in a weighted manner using the current and predicted congestion level,by estimating the outgoing traf?c?ow.The adaptive back-off interval is performed fairly regarding the number of neighbor nodes and fading channels in order to schedule adaptively the packet transmissions.DPCC uses MAC ACKs with piggybacking the envisaged rate.
B.Rate Control-based Scheduling
Like protocols of the previous category,those presented here after use a hierarchical reactive control,applied to the source and forwarder nodes.They differ,however,from the previous ones with respect to the applied method.The communications are re-scheduled by considering the rate change,without using any interference calculation.Table II summarizes the rate control-based scheduling protocols.
The operation of the proposed solution in[40]and MCCP [62]is based on successive data and schedule intervals.During data interval,nodes send events using the schedule they received from their next hop nodes.A single packet is sent per slot,whose length determines the reporting rate.Short slot length allows to forward more traf?c per time unit.During the schedule intervals,nodes generate the schedule for the next data interval to obtain maximum throughput and avoid congestion.The nodes use an initial event request where they indicate to the next hop nodes the initial event reporting rate and the size of their sub-trees to permit the calculation of the schedule at the beginning of the schedule interval,which contains the slot length,the total number of slots,and the allocated number of slots.During the schedule interval,nodes at one hop from the sink send the schedule packets to the nodes at the previous hop.Then every node compares the allocated slot length with the one it calculates and forwards
10IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
the greater one.This schedule manner does not need a wide synchronization.
Time slots are dynamically assigned,depending on the per-hop average packet delivery time,and the buffer size.The average packet delivery time observed during the data interval is used as the slot length for the next data interval.Every node measures change in buffer occupancy between two consecutive data intervals and the predicted buffer occupancy(the actual interval buffer occupancy value+difference from the previous interval value).If the predicted value is not in the optimal range,then nodes adjust their slot length for the next interval by adding or subtracting a deviation factor.MCCP(Multi event Congestion Control Protocol)[62]accepts the following event reporting modes:general event reporting,per-node fair event reporting,and prioritized multiple event-reporting. Slot attribution in this scheme does not show how to avoid contention,since no interference set establishment is used.On the other hand,if a purely sequential tree-based scheduling is used,performance will be very low.
C.Coarse Grained Rate Control Protocols
Unlike the previous two classes belonging to reactive con-trol protocols,those presented in the following,limit the reaction to the concerned node,i.e.only those nodes causing the congestion(generally within one hop from the detector) to decrease their rates.
In this class,the AIMD paradigm is used to react to congestion.As the information contained in the congestion noti?cation is limited,approximate rate adjustment is applied to the concerned nodes.Table III summarizes protocols of this class.
1)CODA[17]:Congestion Detection and Avoidance pro-tocol uses a the present and past channel load conditions, and the current buffer state at each receiver,as congestion indication.CODA listens and measures channel load only at transmission moment,as carrier sensing is required be-fore transmission.Once congestion is detected,the receiver broadcasts an explicit congestion noti?cation back-pressure to its neighbors and adjusts locally the rate in order to avoid congestion spread.The neighbors consequently diminish their sending rates.The back-pressure upstream propagation is decided according to the local network conditions.If the congestion persists,the back-pressure is propagated up to the sources.
The source asks for constant feedback(ACK)from the sink, through setting”regulate bit”in event packets,to preserve its rate.If the source does not receive ACKs,it reduces its rate. Also,low event packets rate reception at the sink is explained as a congestion indication that forces the sink to stop sending ACKs.
CODA does not ensure fairness,and it does not detail how to change the rate after the congestion.The bandwidth may be badly used if the traf?c control is not well designed.
2)SENTCP[31]:is a hop-by-hop congestion control pro-tocol with three principles,computing congestion degree, sending feedback,and processing this feedback.It uses the ratio between the average packet service time and the average packet inter-arrival time,as well as the buffer occupancy ratio to estimate congestion degree at each node.If packet length is variable,it uses bit ratio rather than packet ratio.The feedbacks are sent to the neighboring nodes to adjust their sending rate.The later adjust their local rate,and they may relay the feedback to the next-hop.Every node substitutes the received congestion values by the ones it calculates.SenTCP may send feedback periodically,or when the buffer ratio exceeds a?xed threshold.
3)ART[28]:Asymmetric and Reliable Transport Mecha-nism provides event and query reliability,combined with con-gestion control.It classi?es nodes to Essential nodes(E)and Non-essential ones(N).Higher energy level nodes are chosen as E nodes and form a topology toward the sink,ensuring end-to-end event and query reliability by recovering lost messages.
E nodes send NACK(Negative ACK)to the sink when the query is lost,by using message’s sequence numbers as loss indication.For ensuring reliable event messages transfer,E nodes send an event alarm message to the sink and wait for the ACK.They retransmit this alarm if it is lost.
ART uses distributed congestion control handled by the E-nodes.It regulates the traf?c by decreasing the active non-essential nodes.If an ACK is not received by E nodes during a time-out period,traf?c of N nodes is reduced by sending them a Congestion Alarm message CA to stop their sending.If the congestion is not removed(ACK not received),the E node resends the CA by increasing the hop-count.When receiving the ACK,E nodes send Congestion Safe CS message to N nodes(with the hop-count value of the latest CA)to resume their normal sending.
Choosing E-nodes by only considering energy may be ineffective in some scenarios,where other parameters like the communication and event coverage assured by the nodes would be of high importance.Further,the fairness aspect between E-nodes has been completely ignored.
4)XLM[67]:is a cross-layer protocol fusing communica-tion layers into a single protocol to minimize energy consump-tion,adapt communication decisions,and avoid congestion. XLM applies a receiver-based contention using routing level location,hop-by-hop congestion control,and distributed duty cycle.A node initiates a transmission by the broadcast of an RTS(Request To Send)with its location and that of the sink. By the reception of the RTS,each neighbor that is closer to the sink decides upon its participation according to the RTS Signal To Noise ratio(SNR),the remaining energy and available buffer space.If no CTS are received because of network congestion,the node multiplicatively decreases its generated rate.Otherwise,the generated rate is linearly increased for each received ACK.The overhead caused by this approach is heavy as each transmission at every hop must be preceded by a handshake message exchange.Also,the interpretation of CTS loss as a congestion is not accurate.
5)Bandwidth Management Architecture Protocol:This protocol[74]develops a rule system to specify how the generated traf?c should be treated.It contains three compo-nents for bandwidth management:a rule system with priority queuing,a hop-by-hop?ow control scheme,and a routing prot ocol.Each rule maps the data type and the generated value to a transmission rate,and a traf?c class scheduling using phenomena’s priority and its location.Nodes queue packets
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY11
TABLE II
R ATE CONTROL-BASED SCHEDULING PROTOCOLS.
Protocol Congestion Detec-
tion Congestion
Noti?ca-
tion
Congestion
Control
H-
by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation Parameters Compared with
TDMA like, MCCP [40],[62]Queue Length+
Packet Delivery
Time
Scheduling
msg
Slot Length
Change
H-by-H event No Simulation
NS2
Packet Receive Ratio,
Energy Consumption,
Throughput
NCC,No Scheduled jittered
Forwarding,Source based
Congestion Control
using the traf?c class,and each node implements a rate-control mechanism.The packets are forwarded from the highest-priority not empty queue.When the queue size exceeds a threshold,the receiving node sends a synchronous NACK to slowing down the transmitter by momentary stopping sending. The transmitter waits by overhearing the congested nodes transmissions,and it resumes transmissions after hearing at least two packet transmissions from this node,which are as an indication of free queue space.However,sending from higher priority queues until empting them may penalize others.
6)PCC[46]:Priority-Based Coverage-Aware Congestion control protocol is a hop-by-hop mechanism at the network and MAC layers.Nodes generate periodic packets at a constant rate until event happening,where nodes generate event packets (indicated in the header)with higher rate and priority.Inter-mediate nodes forward packets with different priority using this indication.PCC uses queue scheduling with two queue thresholds to drop event and non-event packets at the network layer.When the queue length is less than the low threshold all packets are saved.When the queue length is between the two thresholds,non-event packets are stochastically dropped.If the queue value is greater than the high threshold,all non event packets are dropped,as well as some event packets.PCC de-?nes packets cumulative survival probability and transmission failure probability in MAC/PHY(Physical)layer to quantify link quality and ensure fairness to remote nodes by cumulating their values.
PCC does not use any rate control and it performs static packets rate.It does not perform collision control,despite hav-ing information about channel state that is obtained through estimating link quality.
D.Exact Rate Control Protocols
This category uses tunable decrease,where the information contained in the congestion noti?cation permit to the con-cerned node to decrease its rate in a precise manner depend-ing on the degree of the congestion.Table IV summarizes protocols of this category.
1)ATP[30]:Ad-hoc Transport Protocol uses feedback for three purposes,i)initial rate feedback for start-up rate estima-tion,ii)progressive rate feedback for congestion detection, congestion avoidance,and congestion control,iii)and path failure noti?cation feedback.
The intermediate nodes calculates available rate and pig-gyback it on the forwarded data packets.The receiver then collects and sends it periodically.Every node maintains two parameters.The average queuing delay of traversing packets, Qt,and the average packet transmission delay at that node, Tt.Tt depends on the contention between nodes in the same vicinity,and Qt depends on packets congestion of different ?ows at the same node.The node stamps the sum Qt+Tt if the previous value on the packet is smaller.Qt and Tt measures queue length and channel load,similarly to CODA[17],but in an accurate end-to-end manner.
The sender uses the feedback to increase,decrease,or maintain its rate.If the rate’s feedback is lost(path failures), ATP performs a multiplicative decrease of the sending rate up to a maximum of two epochs.If no feedback is received for the third epoch,the sender moves to the connection initiation phase.ATP does not consider energy issue and it does not provide enough details with regards to fairness.
2)RCRT[51]:Rate-Controlled Reliable Transport is a centralized sink initiated transport protocol for loss-intolerant concurrent WSN applications.Each source initiates a?ow by establishing an end-to-end connection with the sink,using initial round-trip time(RTT)estimation and source desired rate.RCRT detects congestion and adapts the rate at the sink if the time to repair a loss is much more than RTT. It maintains a per-?ow list of the out of order received packets.The list’s length indicates the number of received packets after the?rst no recovered loss,which re?ects the loss elapsed time.The number of RTTs elapse after the loss is a congestion indication.RCRT uses AIMD rate scheme with a time-dependent multiplicative decrease,based on loss rate. It uses a NACK scheme loss recovery,but it tolerate moderate end-to-end losses that may be caused by transient congestion or poor wireless links.So,sources transmit at a higher rate even with few losses.The new rate is piggybacked in NACK or sent in a separate packet.The use of an end-to-end scheme has slow reaction,and it causes high energy consumption. 3)PORT[52]:Price Oriented Reliable Transport protocol employs the termed node’s price,which measures the commu-nication cost from a node to the sink,in terms of path loss rate. This metric is increased when congestion happens.The sink continuously reports to the sources the desired reporting rate according to their price,the?delity needed by the sink,and their contribution for enhancing this?delity.PORT requests packets from less congested nodes to save energy,while maintaining the necessary level of reliability.
Each node dynamically chooses its forwarder node,using the loss rates relating the node to its neighbors,and the prices of the latter.Sending end-to-end control information to every node may be dif?cult to achieve in multi-hop networks. Dynamic maintenance of a list of neighbors at every node with continuous update of loss rates has a signi?cant memory footprint and communication overhead.
4)LATP[66]:Link Adaptive Transport Protocol is a trans-port layer end-to-end rate control scheme based on MAC layer feedback of the bottleneck node.It controls the offered load
12IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
TABLE III
R ATE CONTROL-BASED SCHEDULING PROTOCOLS.
Protocol Congestion Detection Congestion No-
ti?cation Congestion
Control
H-by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation Parameters Compared with
CODA [17]Queue Length+Channel
Load
Back-pressure
msg
AIMD H-by-H,
E-to-E
Event No Simulation,Ex-
perimentation
Energy Tax,Fidelity
Penalty
No congestion
control,open
loop control
SENTCP [31]Packet inter-arrival Time/
Service Time,Buffer Oc-
cupancy Ratio
Feedback msg Rate Control H-by-H Event,
Periodic
No Simulation Throughput,Packet Loss
Ratio
TCP
ART [28]Ack Loss Feedback msg Stop
Sending
H-by-H Event,
Query
Yes Simulation NS2Residual Energy,Net-
work Lifetime,E-to-E
Delay,Loss Ratio
Alone
XLM [67]CTS Packet Loss AIMD H-by-H Event Yes Analytical,
Simulation
Goodput,Consumed En-
ergy,Latency
ESRT,CBR,
RMST
PCC [46]Queue Length Locally Packet Drop H-by-H Periodic,
Event
No Simulation Throughput,E-to-E De-
lay,Fairness
FIFO
based on the contention degree.As the link capacity is time variant,the feedback provides the available path capacity for the sender to improve https://www.wendangku.net/doc/7515472058.html,TP uses the channel busyness ratio and throughput value(successful transmission)to predict the source’s link contention.As link busyness ignores the hidden node problem,adding throughput estimation provides an accurate contention state.The throughput time includes RTS-CTS-DATA-ACK time.The sender controls the rate using periodical?nal-receiver feedback,as intermediate nodes piggyback contention information in outgoing packets.The receiver estimates contention degree on the path and informs the sender.The sender?rst uses a small rate until the reception of the?rst feedback,and then it controls the rate in an AIMD like manner.
5)ECODA:In ECODA[39](Enhanced COngestion De-tection and Avoidance protocol),packets are dynamically prioritized,using their initial static packet priority,delay and hop-count.The delay is the time from the packet generation to current time.ECODA de?nes the buffer weighted priority as the sum of terms,each of which is the number of packets in the appropriate priority class multiplied by the class priority.The weighted buffer difference of a node is the difference between its weighted buffer value and the maximum weighted buffer of its neighbors.ECODA uses buffer length and weighted buffer difference to deduce congestion status,which are piggybacked in packets.The packet delay value is also piggybacked on packets so that continuous neighbors delay estimation on a path gives the path delay to be used by the source as the rating send.When receiving a back-pressure message,the source node decreases its rate,or adjusts the rate for different paths if multiple paths exist.This is done by using the maximum delay.ECODA uses two queues,for sourced and forwarded traf?c respectively.The scheduler selects the next packet in a round-robin way,and the forwarding queue contains separated priority ordered sources packets.It uses an AIMD scheme,but it differentiates forwarding and sourced rates.The weighted buffer concept can be ineffective when a node has few packets but with a higher priority than another one with more packets. Similar problem can be viewed in forwarding queue containing packets from the same source with higher priority than others, which must wait for the round-robin cycle.
6)CONSISE[57]:provides downstream Congestion con-trol from the sink to the sensors.There are two categories of nodes:-Receiver nodes that are concerned with the messages, i.e.,buffering packets does not represent any overhead or additional cost,which simpli?es the buffering of different packets,and-No-receiver nodes that just act as forwarders, and for which buffer occupancy represents an overhead. CONSISE periodically adjusts the downstream sending rate of sensors to minimize downstream congestion in?uenced by traf?c,from the sensors to the sink.It also adjusts the con-tention caused by broadcast.CONSISE controls the receiving and sending rate for a receiver,the sending and receiving rate for a non-receiver.The non-receiver nodes that relay receiver nodes form a chain acting as a single virtual link,with the same sending and receiving rate.Every node maintains the maximum sending rate based on local channel conditions, and the current sending rate based on downstream channel conditions.A receiver determines the fastest route from the sink by choosing at each epoch the upstream node from which the maximum number of packets was received,and it noti?es this node that it is selected as the preferred upstream receiver. The preferred node sets its sending rate using the receiving rate(s)of its downstream receiver(s).A node that does not get noti?cation to be a preferred receiver gradually decreases its sending rate.Every node piggybacks in every packet it forwards the current and the maximum sending rates,as well as the Id of bottleneck downstream receiver.The sending rate is determined by the explicit feedbacks received from the downstream nodes.Every node maintains a separate list structure of the upstream receivers,including their sending rates,downstream dependency,and their required receiving rates.
This protocol does not give importance to collision mit-igation,as it just tries controlling sending rate without any scheduling.It also causes control overhead through piggyback-ing control information.
E.Interference-based Rate Partitioning Protocols
Contrary to all previous solutions that are reactive,the once presented in the following are preventive and try to avoid the congestion caused by the interference.This is assured by exploiting the knowledge of the interfering nodes,which is used for a capacity sharing between the nodes.This category of protocols is summarized in table V.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY13
TABLE IV
E XACT RATE CONTROL PROTOCOLS.
Protocol Congestion Detec-
tion Congestion
Noti?cation
Congestion Con-
trol
H-
by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation Parameters Compared
with
ATP[30]Queuing Delay+
Transmission Delay Information in
header
Rate adjustment E-to-E Yes Simulation Throughput,Fairness TCP default,
TCP ELFN
PORT [52]Packet loss rate Feedback msg H-by-H Resource
control+E-to-E
Traf?c Control
E-to-E continuous No Simulation
NS2
Energy Consumption Directed
Diffusion+
ESRT
RCRT [51]High Time to Re-
pair Losses
New Rate in
NACK header,
or Feedback
Rate msg
AIMD Rate Con-
trol
E-to-E All types Yes Implementation Goodput,Rate,Packet
Reception
IFRC
LATP [66]Channel Busyness
Ratio+Throughput
Measurement
Information in
header
Rate control E-to-E Multimedia
streaming
(Continuous)
No Simulation
NS2
Delay,Jitter,Through-
put,Packet Loss Rate
TFRC,TCP
NewReno
ECODA [39]Weighted Queue
length
Information in
header
Rate adaptation H-by-H Periodic No Simulation
NS2
Throughput,E-to-E De-
lay,Weighted Fairness
CODA
CONSISE [57]Periodic Rate Con-
trol
Information in
header
Receivers Rate H-by-H Sink to
Sensors
Information
Yes Simulation
NS2
Latency,Number
of Retransmissions,
Number of requests Sent
NCC,NACK
based loss re-
covery
1)IFRC[49]:Interference Aware Fair Rate Control pro-tocol is a rate allocation scheme for tree-based wireless sensor networks.It uses a CSMA-like protocol with link layer retransmissions.Each node maintains a queue for both generated and routed packets and detects congestion using the queue length threshold.It shares this information and its source rate with potential interferers using overhearing.Every node adapts its rate in an AIMD manner not to exceed the channel capacity.A node n1is a potential interferer of node n2if a?ow originating from n1uses a link interfering with the link between n2and its parent.The Potential interferers set of a node covers the node’s sub-tree,its neighbors sub-trees,and also includes nodes in its parent’s neighbors sub-trees.Each node allocates to its potential interferes a fair and ef?cient share of the nominal bandwidth.Finally,each node uses the minimum of all attributed values.When a node,i,is congested, all its descendants are noti?ed of its congestion and reduce their rates.The node’s neighbors,including its parent,set their originating rates to the originating rate of i;this provided that the latter is lower than their rates.The process is repeated at all the neighbors of is parent.Recursively,descendants of the parent’s neighbors reduce their rate to i’s rate.
2)FLUSH[50]:is conceived for applications handling large data.The sink schedules data transfers sequentially in a round-robin fashion,to avoid inter-?ows interfering.After a sink request,Flush has four phases,i)topology query,ii) data transfer,iii)acknowledgement,and iv)integrity check.It uses end-to-end ACKs,implicit control,and hop-by-hop rate control.In the transfer phase,Flush dynamically chooses a sending rate for a path using a combination of local bandwidth measurements and interference estimation algorithm.It com-municates this rate by piggybacking to every node between the bottleneck and the source.This is to avoid intra-path interference.The maximum sending rate without collisions and loss depends on the interference range at each node, as well as the path length(for short paths).Every node continually estimates and updates its interference range and its necessary sending time by using information it acquires by overhearing the channel.It then piggybacks them later on data packets.Flush rate control uses two rules:
-A node transmits when its successor is free from interfer-ence.i.e.,each node waits the forwarding of its successor and all nodes that have interfering transmissions with the successor’s reception.
-A node sending rate should not exceed its successor sending rate.As a result,the source does not send faster than the slowest node along the path.If a nodes queue reaches a threshold,it temporarily increases the advertised delay to avoid congestion.
The sink saves missing packets sequence numbers for recovery at the ACK phase.Flush uses end-to-end selective negative ACKs,but it relies on link layer retransmission.When the data is recovered,the sink veri?es its integrity.
Flush divides time into slots and one packet can be sent per slot,and nodes cannot send and receive in the same slot.The maximum tolerated transmission rate of a node(that does not lead to collision)located N hops from the sink with i hops interference range is:r(N,I)=1/Min(N,2+I),supposing that the maximum rate is1Pkt/s.The best rate requires every node to determine the smallest safe inter-packet delay. Flush is very restrictive as only one source at a time can transmit,which breaks intermediate nodes with source traf?c from transmitting during the whole allocated transmission. Flush has been limited to capacity sharing,but with a high abstraction of slot scheduling.A duty cycle based mechanism cannot be used with Flush,as it bases on continuous measuring of interference.
3)CADT[61]:Capacity Aware Data Transport protocol is similar to FLUSH[50],but it permits to handling several?ows at the same time.It tries using the maximum link capacity by performing rate control to the congested links,using link inter-ference and buffer occupancy.Every node piggybacks control information(transmission rate,transmission interval,current buffer size)in its data packets.It overhears the transmission slot of its neighbors in the interference set.The capacity of bottleneck link reveals the sub-network capacity.The transmis-sion interval for a link,l(i,j),is called transmission interval of node,i.It includes the total transmission time from node,i,to node,j,reception at node j,transmission/reception time of all
14IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
TABLE V
I NTERFERENCE-BASED R ATE P ARTITIONING P ROTOCOLS.
Protocol Congestion
Detection Congestion
Noti?ca-
tion
Congestion
control
H-
by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation parameters Compared
with
IFRC [49]Queue length Information
in header
Rate adjust-
ment
H-by-H Continuous No implementation Throughput,Packet Reception,
Rate Adaptation,Instantaneous
Queue size,Max/Min Goodput
alone
FLUSH [50]Queue Length Information
in header
Rate control H-by-H Query Yes Implementation Overall Throughput,Transfer
Phase Throughput
Different?xed
sending rate
CADT [61]Queue length+
Link Capacity
Information
in header
Rate adjust-
ment
H-by-H Continuous No Simulation Packet Delivery Ratio,Packet
Delivery Latency,Queue Length
Alone
the nodes in the interference set.This value varies over time and it is continually updated.
The minimum transmission interval of node,i,on the link,l(i,j),is the maximum transmission interval of all the links in the interference set,except l(i,j).The sum of interfering set rates must not exceed the link capacity to avoid interference. CADT estimates link capacity using aggregated transmission interval(the sum of all links transmission interval in the interference set).
However,the calculation of the capacity has not been justi?ed.The rate control uses AIMD,by exploiting the buffer of immediate downstream node,and the link state of the concerning node as well.
4)Mesh Interference Protocol:This protocol[76]takes as input the topology of the network,the?ows routing paths, and their desired sending rate.It captures the network’s interference dependencies as an approximate con?ict graph, and it uses an iterative process to estimate the(max-min fair) safe sending rate for each?ow in the purpose to re?ect the total network throughput.
F.Interference Aware Scheduling
Randomized access schemes are energy inef?cient and wit-ness reduced throughput due to the increased contention.This can be avoided by structured communication with bandwidth allocation and access scheduling.Protocols of this class share with the previous one the aspect of congestion avoidance. However,instead of merely calculating the rate sharing be-tween interferers,a schedule taking into account interference dependencies is applied.Table VI summarizes protocols of this class.
1)Max-Min Fair Collision-Free Scheduling:This protocol [68]tries ensuring fairness for predictable,stable,large sized and high data rates?ows with tree-based WSN scheme. It presents a linear programming formulation algorithm for establishing max-min fair bandwidth allocation,and a collision free distributed scheduling algorithm for time slot allocation using BFS(breadth?rst search).
No details on how to detect interfering nodes has been provided.The bandwidth capacity is changing during time, which requires continuous allocation update.
2)QCRA[60]:Quasi-static Centralized Rate Allocation protocol aims at determining(by the sink)optimal and fair sources transmission rates using information about topology, link loss rates,and communication pattern,to be piggybacked on data packets.Its algorithm is based on rate assignment heuristic,with nodes using CSMA.
It de?nes the network goodput as the minimum packet reception rate at the base station from any node in the network. Its heuristic computes a coarse-grained TDMA schedule be-tween the node’s neighbors to determine neighborhood traf?c rates.A node and its neighbors are split into independent sets of nodes able to transmit simultaneously.From each set, the node that transmits more?ows(sourced and forwarded ones)determines the number of?ows related to the set. The sum of number of?ows in all the sets de?nes the total bandwidth requirement at a node.The heuristic uses this sum to fairly divide the bandwidth among?ows at a node,assuming an implicit coarse-grained time-division.The operation is repeated recursively.As it is based on a CSMA like protocol,losses will certainly happen and their rate is used to assign nodes sending rates.
QCRA allocation decisions are periodic and link loss rates are used to perform rate allocation decisions for the next epoch.Each epoch lasts for tens of minutes.QCRA measures the channel capacity by sending packets from one node to another,rather than using theoretical channel capacity.But the centralized approach at the sink does not bring an added gain.The bandwidth capacity is changing and needs to be recalculated.
3)TSCH[86]–[89]:TSCH(Time slotted Channel Hop-ping)mode of the IEEE802.15.4e is a medium access scheme used with LLNs(Low power and Lossy Networks).LLNs result in large mesh networks composed of resource con-strained devices generally related to internet and serving in many industrial applications such as process control and home automation,characterized by its multipoint-to-point(MP2P) traf?c[88].In TSCH,schedule based approach with channel hopping is used,which requires time synchronization between nodes.The communication is resumed on the repetition of the frame time,which contains slots reserved to each node to send or shared by many nodes.6TSCH is the entity responsible to run this mechanism,and may be seen as an adaptation layer.6TSCH is responsible for controlling topology links resources through the schedule.It also controls the mech-anisms that de?ne how nodes join the network to ensure its good performance by avoiding interferences and ensuring synchronization between neighboring nodes.6TSCH ensures also?ow control by the administration of queues policies for arrived and sent packets,in order to inform TSCH to decline new ones.As TSCH guaranties frames authenticity,6TSCH applies the necessary mechanism to ensure key monitoring at joining event and securing data transfer and control.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY15
TABLE VI
I NTERFERENCE A WARE S CHEDULING
Protocol Congestion Detection Congestion
Noti?ca-
tion Congestion
Control
H-
by-H/
E-to-E
Application
Type
Loss
Recov-
ery
Evaluation
Type
Evaluation Parameters Compared
with
QCRA [60]Packet Loss Rate Information
in header
Periodic
Rate
Adjustment
H-by-H Continuous No Implementation Goodput,Rate IFRC
G.Buffer Over?ow Avoiding
This category of protocols has the preventive(avoiding) feature.But instead to avoid the interference,it is the buffer over?ow that causes the congestion that is avoided.Table VII summarizes protocols of this class.
1)Congestion Avoidance Based on Lightweight Buffer Management,AFA:In[56]a tree based scheme is used, where AFA[69](Aggregate Fairness Algorithm)generalizes the behavior to multipath networks.The authors behavior is avoiding congestion by permitting the parent to send at a rate matching the combined children rates.The sender transmits only when the receiver has enough buffer space.In[56],each nodes packet header piggybacks the buffer state.Before a child sends a packet,it checks parent’s buffer.If full,it does not send until perceiving a non-full buffer state.In AFA,the weighted fairness notion is added.The node uses a token-bucket scheme similarly to Fusion[20]when the parent buffer is not full.As the child may still lose parent’s buffer state due to the hidden terminal problem,the[56]protocol proposes a 1/6-buffer solution(advertising one sixth of reel buffer size). It also proposes an adaptive1/k-buffer solution,where k is dynamically modi?ed by the node advertising its buffer.Each node starts with k=6and reduces it in the absence of buffer over?ow.After a buffer over?ow,it increases k.In practice, it dynamically adjusts k without buffer over?ow. However,both AFA and[56]suppose that the collision problem is resolved by the MAC protocol using exponential back-off,which is not always effective in practice.
2)RBC[32]:Reliable bursty converge-cast protocol offers end-to-end packet based reliability in forwarding direction by scheduling retransmissions to reduce contention with newly generated packets.It implicitly detects losses,in a hop-by-hop way by hearing parent’s packets header.
The base-station uses the bloc ACK method,to enhance link utilization and adapts retransmission timer regarding network state.RBC ranks nodes by their queue size and the number of queued packets transmission attempts.Therefore,new packets are sent immediately to enable continuous packet forwarding. To reduce interference of the same rank packets and balance the network queuing and channel contention,inter-node packet scheduling uses packets number of a certain rank,permitting to nodes with more packets transmit earlier.
RBC implements a simple hop-by-hop?ow control.Nodes piggyback their free queue size in packets and the sender detecting this number below a threshold stops sending for a certain period.Nevertheless,reducing interference by enabling the higher rank nodes is not ef?cient,as these nodes may interfere.Further,the fairness is not well ensured.
V.R ESOURCE C ONTROL P ROTOCOLS
A.Alternative Path Use
In these protocols,the application?delity is considered,and the enhancement of resource use in the critical situation is adopted.Multi-path is used as a solution to augment resources. However,only a single path is used at the same time,and the other paths are used as alternatives in case of congestion. 1)SPEED[102]:tries maintaining a desired uniform deliv-ery speed of real-time applications by diverting traf?c through multiple routes and regulating sending rate,so,the end-to-end packet delay becomes proportional to the distance between the source and destination.It uses single hop sender delay estimation as a congestion indication.The sender timestamps queued packets and calculates the round trip single hop delay when receiving their ACK.SPEED searches for the next hop candidates that can support the desired delivery speed.If no candidate is found,the packet is buffered momentary or dropped not to propagate the congestion.It uses back-pressure re-routing to divert traf?c.SPEED supposes the existence of location information,which is resource consuming.Further, no traf?c control is used,which may result in excess capacity use.In[105],[106]enhancements of SPEED were proposed.
2)HTAP[33]:Hierarchical Tree Alternative Path algo-rithm is proposed for event-based sensor applications.It tries ensuring application reliability during overload periods with-out reducing the sources rate when sending critical events. HTAP combines two algorithms,Alternative Path Creation (APC)and Hierarchical Tree Creation(HTC),and it uses the network density to choose between them.When congestion takes place or a node’s battery is about to draining,APC and HTC form alternative paths to the sink by unused nodes. APC uses these nodes by randomly exploiting neighboring table,while in HTC these nodes are placed in a hierarchical levelled tree starting from0for the leaves nodes.Every node piggybacks its buffer occupancy,re?ecting its congestion state, when sending packets and the neighbors refresh their neigh-boring state tables when overhearing packets.A congested receiver sends a back-pressure packet to the sender in the purpose to remove congestion.The sender stops transmitting to this node and searches for a less congested receiver which leads to alternative paths creation.
B.Congestion Reacting Multipath
Unlike the alternative path protocols,the ones presented here use concurrent paths at the same time to enhance resource usage.Multiple-path rate balancing is used in response to congestion.
16IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
TABLE VII
B UFFER O VERFLOW AVOIDANCE PROTOCOLS.
Protocol Congestion Detection Congestion
Noti?ca-
tion Congestion
Control
H-
by-H/
E-to-E
Application
Type
Loss
Re-
covery
Evaluation
Type
Evaluation Parameters Compared with
Lightweight Buffer Man-agement, AFA[56], [69]Queue Length Information
in header
Stop
Sending
H-by-H Event,
Periodic
No Analysis,
Simulation
Packet Loss,Source
Rate,Energy
Expenditure
E-to-E Congestion
Control,CODA,No
Congestion Control
(NCC)
RBC[32]Remaining Queue
Length Information
in header
Stop
Sending
H-by-H Event Yes Experimentation Packet Delivery Delay,
Loss Ratio
SEA,SWIA
1)TADR[59]:Traf?c-Aware Dynamic Routing protocol uses the idle or under-loaded nodes to remove congestion and enhance the throughput.It routes packets around the congestion areas and distributes them on multiple paths.It uses the depth to?nd the shortest paths and queue length to detect congestion.When there is no congestion,TADR chooses the shortest paths.In congestion case,TADR dynamically picks out multiple paths,so that the uncongested areas record or forward the excess of packets.
2)CAR[71]:Congestion-Aware Routing protocol discov-ers the congested zone between sources and the sink,that it preserves to forward high-priority traf?c.CAR separates High Priority(HP)traf?c from Low Priority(LP)traf?c and uses multipath forwarding.HP traf?c is the only routed through the shortest path nodes;while LP traf?c is forwarded by uncongested nodes through longer paths.CAR follows three phases,starting by the HP network formation,congestion zone discovery,and differentiated https://www.wendangku.net/doc/7515472058.html,bining these functions divides the network into congested zones and non-congested zones,where only the HP traf?c is routed through the congested zone.
3)QOS-ACC:In QoS-ACC(QoS adaptive cross-layer pro-tocol)[44],the authors suppose that nodes sense different events and the sink assigns different priorities to data accord-ing to its importance.They use multipath routing.QAC-ACC nodes send data to the appropriate next hop measuring QoS requirement of the packet(minimum delay,minimum service rate,reliability level),by using a distributed MAC manager. It considers a primary route and at least one alternative route. Nodes dispose priority queues for different events.A classi?er at the network layer puts sourced and forwarded packets of ”the same priority events”in the same queue with a higher priority to forwarded data as their loss results in more resource wastage.The scheduler plans the queues according to the inter queue priority.Therefore,adjusting the scheduling rate of a node adjusts its output rate.At the beginning,each node starts with a lower scheduling rate,and the originating rate depends on the scheduling rate and the data priority.QOS-ACC has two active queues,QRT and QNRT,for real time and non-real time applications,respectively.A higher priority is assigned to real time applications and one back-up queue(QBACK-UP)is used for unacknowledged non-real time data.QOS-ACC uses the ratio of average packet service rate and packet scheduling rate as congestion indication,(named as packet service ratio). It uses implicit congestion noti?cation by overhearing.QAC-ACC applies a resource control method when receiving the parent congestion indication by splitting the real time-traf?c to an alternate route.No detail concerning how to choose the scheduling rate is given,as despite of a resource control method use,a rate control has to be done in failure of resource control.Moreover,the use of the backup queue is not detailed.
C.Interference Avoiding Scheduling
In addition to the simultaneous use of multi-path,scheduling is also used in this category of protocols.Table VIII summa-rizes the protocols of this category.
1)TARA[34]:Topology-Aware Resource Adaptation strat-egy proposes using resource control for the aim of ensuring application?delity.It uses a minimum number of nodes along the routing path during idle periods,to minimize energy consumption,and activates appropriate nodes forming new paths with suf?cient capacity,to handle increasing traf?c without congestion.
The authors study the in?uence of multiple paths on the end-to-end channel capacity and provide some guidelines for resource control schemes.In absence of interference,the capacity of a topology is the throughput of one-hop capacity. It becomes much smaller in the presence of link interference and a topology’s throughput is limited by the bottleneck link(s) throughput.TARA calculates thus the capacity fraction for this portion.TARA de?nes the link congestion sum as the sum of link’s traf?c and interfering links’traf?c.The chosen bottleneck has the largest congestion sum value.TARA uses graph-coloring approach for capacity estimation.It de?nes the topology interference degree and constructs the spatial interference graph,where each vertex is a wireless trans-mission,and the edges indicate that two transmissions are in the same interference range.Two links with an edge cannot transmit concurrently under the optimal schedule. Calculating the capacity fraction is to assign a coloring to the spatial interference graph,where each color is a time frame corresponding to transmission over a link.It represents the maximum throughput,or the delivery rate observed by the sink.When a hotspot node becomes congested(buffer occupancy and channel loading upper than the threshold),it chooses two nodes for constructing the detour path between the distributor and the merger nodes.If the creation of a detour path is not possible,traf?c control mechanism is used by sending a back-pressure message to upstream neighbors.
2)I2MR[63]:Interference-Minimized Multipath Routing protocol is proposed for high rate streaming.It tries increas-ing the throughput by discovering disjoint paths for load balancing.It applies congestion control to load the paths at the highest possible rate.It uses con?ict graphs to indicate interfering link groups that cannot be simultaneously active.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY17
TABLE VIII
R ESOURCE CONTROL PROTOCOLS.
Protocol Congestion
Detection Congestion
Noti?ca-
tion
Congestion Con-
trol
H-
by-H/
E-to-E
Application
Type
Loss
Re-
cov-
ery
Evaluation
Type
Evaluation Parameters Compared with
HTAP [33]Queue length Information
in header
Resource Control
(Alternative Path)
H-by-H Event No Simulation-
MATLAB
Network power,Packet
drops,E-to-E Delay
alone
SPEED [102]Single hop Packet
Delay
MAC
layer
feedback
Resource+Traf?c
control
H-by-H Real time no Simulation
(GloMossim),
Experimenta-
tion
E-to-E Miss Ratio,
Control Packet
Overhead,Energy
Consumption
AODV,DSR
QOS-ACC [44]Packet Service
Ratio
Information
in header
Resource control
(Alternative
Routes)
H-by-H Continuous Yes Simulation Queue Length,Residual
Energy
CCF,No Congestion Con-
trol
TADR [59]Queue length Resource Control H-by-H Event Simulation
TOSSIM
Receiving Packets Rate,
Throughput Ratio,En-
ergy Consumption per
Received packet
MintRoute
TARA [34]Queue length+
Channel load
Feedback
msg
Resource Control
(Detouring Path)
H-by-H Continuous No Simulation Fidelity Index,Total En-
ergy Consumption,Bit
Energy Consumption
NCC,Traf?c Control ideal
resource control,topology-
unaware resource control
I2MR [63]Queue length Feedback
msg
Rate Control H-by-H Continuous No Simulation
Glomosim
Throughput,Energy
Consumption
AODV,NDMR
SIPHON [27]Queue length+
Channel load
Bit in the
header
Traf?c
Redirection
H-by-H Event No Simulation,
Experimenta-
tion
Energy tax,Energy tax
Savings,Fidelity ratio,
Residual energy
CODA
Talonet [45]Queue length Information
in header
Rate control+De-
touring paths
H-by-H Continuous No Simulation
NS2
Dropped packets,Power
Consumption
TARA,Back-pressure,NCC
ADCC [41]Active period-
Required Service
Time
Feedback
msg
Resource+Rate
Control
H-by-H Periodic No Simulation
NS2,Imple-
mentation
Packet Reception Rate,
Loss Rates
NCC,Traf?c Control
STCP [25]Queue Length Bit in the
header
Rate control
or Traf?c
Redirection
E-to-E Event,
Continu-
ous
Yes Simulation Packet Latency,Energy
Spent
alone
PLR [36]Queue Length Feedback
msg
Traf?c control+
Resource control
(Alternative Path)
H-by-H Continuous No Simulation
NS2
E-to-E Throughput,
Packet delivery ratio
PCCP,LWBM
Total Interference Correlation Factor(TICF)for a set of disjoint paths,derived from the con?ict graph and de?ned as the number of links in the two paths that can interfere, describes the degree of interferences for all the paths in the set,and it is used to evaluate the quality of a path for multipath load balancing.I2MR records the interference zone of the?rst discovered path to avoid discovering another path within this interference zone.Each source node sends data concurrently using the primary and secondary path pair.It switches to the backup path only when either of the active paths fails. Intermediate nodes detect long-term congestion using buffer length and notify the source to reduce its rate to the next prede?ned one.If the smallest prede?ned rate is reached and the congestion still happens,the source suspends packet loading(besides reducing the loading rate)for a prede?ned time,and sets a?ag that is cleared later.If the congestion persists even with these two mechanisms,the source starts discovering new paths.
D.Dual Emission-based Protocols
In this class of protocols,the resource enhancement is envisaged either by using more than one radio,or different power emissions.
1)Siphon[27]:proposes distributing wireless dual radio virtual sinks to avoid congestion when it is persistently de-tected.Virtual Sinks(VS)form dynamically a secondary ad-hoc network,and route congested traf?c to the physical sink to maintain the events rate,and avoid throttling or deleting packets.Siphon follows three phases starting by discovering the virtual sinks that will be selected.After the detection of congestion,the concerned node enables a redirection bit(in the network layer header)that permits to divert traf?c out of the neighborhood,utilizing the VSs.Upon a VS reception of a redirected packet,it sends it to the near VS toward the sink. Every VS checks for its congestion level on the primary and secondary radios and does not advertise its existence if its radios is(are)overloaded.SIPHON adopts CODA congestion detection strategy in node-initiated detection.
2)TALONet[45]:uses traf?c and resource control to avoid congestion.It uses two different transmission levels to alleviate link-level congestion,and buffer management to alleviate node-level congestion.It uses a multi-path detouring to increase the channel capacity(so can be classi?ed in the previous class,too).TALONet works on three phases,network formation phase where it creates a virtual grid framework where each node is normal or talon.A talon node is the nearest to a certain grid point,which collects and relays the sensing data during the data transmission phase.The second phase is data dissemination,where the normal node transmits its sensing data to a neighboring talon approaching to the sink. This process is repeated until the data reaches the sink.The normal node uses the low power level,while the talon uses the higher one.TALONET applies the congestion avoiding 1/k node buffer management approach as proposed by[56]. Talon nodes piggyback congestion information in data packets. It uses detouring paths to distribute the traf?c in case of congestion and decreases the source rate,in addition to its two level transmitting.If all receiving candidate nodes closer to the sink have no buffer spaces,the source considers the nodes located one-hop from the sink as receiving candidates.
18IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
Since talon nodes use higher power level to relay data,they exhaust their energy faster than normal nodes.To prevent this TALONet uses conditional or periodic topology update.How-ever,using two levels for transmitting does not automatically avoid link level contention,as two normal neighboring nodes may interfere.The1/k buffer method is restrictive.
E.Duty Cycling-based Protocols
Resource enhancement for critical event handling is envis-aged in these protocols by adapting the duty cycle.This is to balance energy consumption and application?delity.Table VIII summarizes the protocols of using Ressource control strategy.
ADCC[41](Adaptive Duty-cycle based Congestion Con-trol)protocol controls congestion using MAC adjustment.It uses resource control by increasing reception rate and traf?c control by decreasing transmission rate.ADCC periodically calculates the required service time using children’s packets inter arrival time to detect congestion.This is using the difference between the required service time and the duration of the active state in the duty-cycle.When incoming traf?c is low,the active time of the receiving node is reduced to save energy.If congestion degree is below the higher threshold,it adjusts its duty cycle to reduce congestion by only applying resource control.If the degree is above the threshold and the active time reached the high limit,it noti?es the children to reduce their rate by a calculated ratio.This duty cycle scheme is not appropriate for event-based applications when the events are unpredictable.The inter arrival time does not give the required service time but the scheduling time,this affects the correctness of the service time equation.
VI.C ONCLUSION AND F UTURE WORKS
The wireless nature of WSNs expose their utilisation to harsh environment conditions like contention(links interfer-ence)and congestion(buffer over?ows)which impact the overall system performance.Transport protocols play a pivotal role in improving the network reliability and throughput.
In this survey,these protocols are analysed in terms of their suitability to detect congestion and notify the concerned nodes so that an appropriate control will be taken.Also,many evaluation parameters to show protocols ef?ciency in overload traf?c circumstances are presented.
Our study on congestion control protocols has shown that the application and?ow types–characterized by the many-to-one nature communications–in?uence and guide the control applied to the traf?c.
Depending on the application types,different mechanisms are used to handle the congestion.Either traf?c control by throttling the node rates or resource control by exploiting idle resources are used to meet the application requirements. From the?ow type point of view,applying the same type of congestion control at all nodes may be a wrong decision to ameliorate the throughput.For example,phase shifting the source nodes is of best control solution for small packet event-based applications,while rate regulation is an adequate solution at intermediate nodes when a bottleneck happens. With voluminous packet event-based applications,the rate control extends to the sources.These applications rely on an contention based MAC protocols.
In periodic and continuous applications,reacting to the congestion can lead to performance degradation due to the elevated frequency of packets sending.So,an avoiding con-gestion control strategy through scheduling the transmissions to eliminate collisions and partitioning the rate to prevent ex-ceeding the interfering nodes capacity seems to be a promising solution.
As sensor networks are energy constrained,upper bounding the sending rates of sources by only congestion limits may re-sult in reducing the network lifetime.Applying an upper bound related to the application?delity is thus of high importance. ESRT[29]and PORT[52]are examples of solutions using this principle.
A scheme similar to CTP[48](A Con?gurable and Extensi-ble Transport Protocol)can be proposed to handle congestion and other transport layer properties in WSN.In CTP,each transport protocol property is implemented as a separate micro-protocol and can be chosen based on the requirements of the application and executed when a speci?c event occurs. The use of the standard IEEE802.15.4e[107]and different IETF proposals[5]in the context of LLNs based IPv6protocol stack,starting by TSCH at the MAC layer,6TSCH[86],[87], [89]and6LoWPAN[108]as adaptation layer,RPL[90], [109],[110]at the network layer,and COAP[83]at the application layer,may enhance the performance of the whole application and can be viewed as the IoT communication based solution meeting industrial requirements and needs.
In the LLN[88]paradigm,subnets of mesh networks are composed with constrained resource devices(such as WSN used in industrial and home automation)and attached to specialized routers.This brings the isolated WSN based applications and protocols towards the IoT based systems[2], which became more attractive notably by the use of IETF proposals cited previously.
A deep comprehension of the mechanisms above and their ef?ciency permit the design of a comprehensive transport protocol that also deals with reliability.
Reliability is habitually dealt with at the transport layer and it is essential to assure effective and dependable applications. Reviewing transport protocols that control the reliability aspect represents a perspective to this work.
R EFERENCES
[1]I.Akyildiz,W.Su,Y.Sankarasubramaniam,and E.Cayirci,“Wireless
sensor networks:A survey,”Computer Networks,vol.38,pp.393–422, 2002.
[2] A.Dunkels and J.Vasseur,“IP for Smart Objects,”Internet Protocol
for Smart Objects(IPSO)Alliance.White paper#1.Version1.1,July 2010.
[3]V.Michopoulos,L.Guan,G.Oikonomou,and I.Phillips,“A Compar-
ative Study of Congestion Control Algorithms in IPv6Wireless Sensor Networks,”in Proc.IEEE Int.Conf.Distributed Computing in Sensor Systems and Workshops(DCOSS),June2011,pp.1–6.
[4] D.Ekaterina and G.Andrei,“Survey on Congestion Control Mecha-
nisms for Wireless Sensor Networks,”Internet of Things,Smart Spaces, and Next Generation Networking,vol.7469,pp.75–85,Aug.2012.
[5]M.Palattella,N.Accettura,X.Vilajosana,T.Watteyne,L.Grieco,
G.Boggia,and M.Dohler,“Standardized Protocol Stack for the
Internet of(Important)Things,”IEEE Commun.Surveys Tutorials, vol.15,no.3,pp.1389–1406,2013.
KAFI et al.:CONGESTION CONTROL PROTOCOLS IN WIRELESS SENSOR NETWORKS:A SURVEY19
[6]K.Zheng, F.Hu,W.Wang,W.Xiang,and M.Dohler,“Radio
resource allocation in LTE-advanced cellular networks with M2M communications,”IEEE Commun.Mag.,vol.50,no.7,pp.184–192, 2012.
[7]J.Kim,J.Lee,J.Kim,and J.Yun,“M2M Service Platforms:Sur-
vey,Issues,and Enabling Technologies,”IEEE Commun.Surveys& Tutorials,vol.PP,no.99,pp.1–16,2013.
[8]M.Chen,J.Wan,S.Gonzalez,X.Liao,and V.Leung,“A Survey
of Recent Developments in Home M2M Networks,”IEEE Commun.
Surveys&Tutorials,vol.PP,no.99,pp.1–17,2013.
[9]J.Shi,J.Wan,H.Yan,and H.Suo,“A survey of cyber-physical
systems,”in Proc.Int.Conf.on Wireless Communications and Signal Processing,November2011,pp.1–6.
[10]J.Wan,H.Yan,Q.Liu,K.Zhou,R.Lu,and D.Li,“Enabling cyber-
physical systems with machine-to-machine technologies,”Int.J.of Ad Hoc and Ubiquitous Computing,vol.13,no.3/4,pp.187–196,Nov 2013.
[11]J.Wan,M.Chen,F.Xia,D.Li,and K.Zhou,“From machine-to-
machine communications towards cyber-physical systems,”Computer Science and Information Systems,vol.10,no.3,pp.1105–1128,Jun 2013.
[12]N.Xu,“A Survey of Sensor Network Applications,”IEEE Commun.
Mag.,vol.40,2002.
[13]I.F.Akyildiz,T.Melodia,and K.R.Chowdhury,“A survey on wireless
multimedia sensor networks,”Computer Networks(Elsevier),vol.51, no.4,pp.921–960,March2007.
[14]I. F.Akyildiz and I.H.Kasimoglu,“Wireless sensor and actor
networks:research challenges,”Ad-hoc Networks,vol.2,no.4l,pp.
351–367,October2004.
[15]P.Spachos,L.Song,and D.Hatzinakos,“Prototypes of opportunistic
Wireless Sensor Networks supporting indoor air quality monitoring,”
in Proc.IEEE Consumer Commun.and Netw.Conf.(CCNC),jan2013, pp.851–852.
[16]X.Yin,X.Zhou,R.Huang,Y.Fang,and S.Li,“A Fairness-Aware
Congestion Control Scheme in Wireless Sensor Networks,”IEEE Trans.Veh.Technol.,vol.58,no.9,pp.5225–5234,November2009.
[17] C.-Y.Wan,S.B.Eisenman,and A.T.Campbell,“Energy-ef?cient
congestion detection and avoidance in sensor networks,”ACM Trans.
https://www.wendangku.net/doc/7515472058.html,w.,vol.7,no.4,pp.1–31,feb2011.
[18] C.T.Ee and R.Bajcsy,“Congestion control and fairness for many-to-
one routing in sensor networks,”in ACM,SenSys,2004,pp.148–161.
[19] D.J.Flora,V.Kavitha,and M.Muthuselvi,“A Survey on Congestion
Control Techniques in Wireless Sensor Networks,”in Proc.Int.Conf.
on Emerging Trends in Electrical and Computer Technology(ICE-TECT),2011.
[20] B.Hull,K.Jamieson,and H.Balakrishnan,“Mitigating congestion
in wireless sensor networks,”in Proc.2nd int.conf.on Embedded networked sensor systems,SenSys,November3–52004,pp.134–147.
[21]P.Spachos,P.Chatzimisios,and D.Hatzinakos,“Energy Aware
Opportunistic Routing in Wireless Sensor Networks,”in Proc.IEEE Globecom Workshops(GC Wkshps),dec2012,pp.405–409.
[22]X.Mao,S.Tang,X.Xu,X.-Y.Li,and H.Ma,“Energy Ef?cient
Opportunistic Routing in Wireless Sensor Networks,”IEEE Tran.
Parallel Distrib.Syst.,vol.22,no.11,pp.1934–1942,Nov2011.
[23]T.Lu and J.Zhu,“Genetic Algorithm for Energy-Ef?cient QoS
MulticastRouting,”IEEE Commun.Lett.,vol.17,no.1,Nov2013. [24] C.Wang,B.Li,K.Sohraby,M.Daneshmand,and Y.Hu,“Upstream
congestion control in wireless sensor networks through cross-layer optimization,”IEEE J.Sel.Areas Commun.,vol.25,no.4,pp.786–795,2007.
[25]Y.G.Iyer,S.Gandham,and S.Venkatesan,“STCP:A Generic
Transport Layer Protocol for Wireless Sensor Networks,”in Proc.IEEE Int.Conf.on Computer Commun.and Netw.(ICCCN),2005,pp.449–454.
[26] C.Lu,B.Blum,T.Abdelzaher,J.Stankovic,and T.He,“RAP:A Real-
Time Communication Architecture for Large-Scale Wireless Sensor Networks,”in Proc.IEEE RTAS,2002.
[27] C.-Y.Wan,S. B.Eisenman, A.T.Campbell,and J.Crowcroft,
“Overload traf?c management for sensor networks,”ACM Trans.Sen.
Netw.,vol.3,no.4,oct2007.
[28]N.Tezcan and W.Wang,“ART:an asymmetric and reliable transport
mechanism for wireless sensor networks,”https://www.wendangku.net/doc/7515472058.html,w.,vol.2,no.
3-4,pp.188–200,april2007.
[29]Y.S.Zgr,Y.Sankarasubramaniam,¨O.B.Akan,and I.F.Akyildiz,
“ESRT:Event-to-Sink Reliable Transport in Wireless Sensor Net-works,”in Proc.4th ACM Int.Symp.Mobile ad hoc networking and computing,MobiHoc,2003,pp.177–188.[30]K.Sundaresan,V.Anantharaman,H.-Y.Hsieh,and R.Sivakumar,
“ATP:A Reliable Transport Protocol for Ad Hoc Networks,”IEEE Trans.Mobile Computing,vol.4,no.6,pp.588–603,2005.
[31] C.Wang,K.Sohraby,and B.Li,“SenTCP:A Hop-by-Hop Conges-
tion Control Protocol for Wireless Sensor Networks,”in Proc.IEEE INFOCOM(Poster Paper),2005.
[32]H.Zhang,A.Arora,Y.-r.Choi,and M.G.Gouda,“Reliable Bursty
Convergecast in Wireless Sensor Networks,”Computer Communica-tions,vol.30,pp.2560–2576,2007.
[33] C.Sergiou,V.Vassiliou,and A.Pitsillides,“Reliable data transmission
in event-based sensor networks during overload situation,”in Proc.3rd Int.Conf.on Wireless internet,WICON,2007,pp.1–:8.
[34]J.Kang,Y.Zhang,and B.Nath,“TARA:Topology-Aware Resource
Adaptation to Alleviate Congestion in Sensor Networks,”IEEE Trans.
Parallel Distrib.Syst.,vol.18,no.7,pp.919–931,jul2007.
[35] A.Woo and D.E.Culler,“A Transmission Control Scheme for Media
Access in Sensor Networks,”in Proc.ACM Mobicom,Jul.16-212001, pp.221–235.
[36]M.Razzaque and C.-s.Hong,“Congestion detection and control
algorithms for multipath data forwarding in sensor networks,”in Proc.
11th Int.Conf.Advanced Commun.Technology,ICACT,Feb.15-18 2009,pp.651–653.
[37]G.Wang and K.Liu,“Upstream hop-by-hop congestion control in
wireless sensor networks,”in Proc.IEEE20th Int.Symp.on Personal, Indoor and Mobile Radio Communications,13-16Sept2009,pp.
1406–1410.
[38]M.Monowar,M.Rahman,and C.-s.Hong,“Multipath Congestion
Control for Heterogeneous Traf?c in Wireless Sensor Network,”in Proc.10th Int.Conf.on Advanced Communication Technology,ICACT, vol.3,Feb.17-202008,pp.1711–1715.
[39]L.Tao and F.Yu,“ECODA:Enhanced congestion detection and
avoidance for multiple class of traf?c in sensor networks,”in Proc.
IEEE15th Asia-Paci?c Conf.on Commun.,APCC,2009,pp.726–730.
[40] F.Hussain,G.Seckin,and Y.Cebi,“Many-to-one congestion control
scheme for densely populated WSNs,”in Proc.3rd IEEE/IFIP Int.
Conf.in Central Asia on Internet,ICI,2007,pp.1–6.
[41] D.Lee and K.Chung,“Adaptive duty-cycle based congestion control
for home automation networks,”IEEE Trans.Consum.Electron., vol.56,no.1,pp.42–47,February2010.
[42]M.Moghaddam and D.Adjeroh,“A Novel Congestion Control
Protocol for Vital Signs Monitoring in Wireless Biomedical Sensor Networks,”in Proc.IEEE Wireless Commun.and Netw.Conf.(WCNC), 2010,pp.1–6.
[43]J.-P.Sheu and W.-K.Hu,“Hybrid Congestion Control Protocol in
Wireless Sensor Networks,”in Proc.IEEE Veh.Technol.Conf.,VTC, 2008,pp.213–217.
[44]M.Rahman,M.Monowar,and C.-s.Hong,“A QoS Adaptive Con-
gestion Control in Wireless Sensor Network,”in Proc.10th Int.Conf.
Advanced Commun.Technol.,ICACT,Feb2008,pp.941–946. [45]J.-M.Huang,C.-Y.Li,and K.-H.Chen,“TALONet:A power-ef?cient
grid-based congestion avoidance scheme using multi-detouring tech-nique in Wireless Sensor Networks,”in Proc.IEEE Wireless Telecom-munications Symp.,WTS,2009,pp.1–6.
[46]X.Qiu,D.Ghosal,B.Mukherjee,J.Yick,and D.Li,“Priority-Based
Coverage-Aware Congestion Control for Multihop Wireless Sensor Networks,”in Proc.IEEE28th https://www.wendangku.net/doc/7515472058.html,puting Systems Workshops,ICDCS,2008,pp.285–290.
[47] A.Sharif,V.Potdar,and A.J.D.Rathnayaka,“Prioritizing Information
for Achieving QoS Control in WSN,”in Proc.24th IEEE Int.Conf.
Advanced Information Networking and Applications(AINA),April 2010,pp.835–842.
[48]P.G.Bridges,G.T.Wong,M.Hiltunen,R. D.Schlichting,and
M.J.Barrick,“A Con?gurable and Extensible Transport Protocol,”in Proc.20th Annu.Joint Conf.IEEE Computer and Commun.Societies.
INFOCOM,2001.
[49]S.Rangwala,R.Gummadi,https://www.wendangku.net/doc/7515472058.html,indan,and K.Psounis,
“Interference-Aware Fair Rate Control in Wireless Sensor Networks,”
in Proc.of the ACM SIGCOMM,September11–152006,pp.63–74.
[50]S.Kim,R.Fonseca,P.Dutta, A.Tavakoli, D.Culler,P.Levis,
S.Shenker,and I.Stoica,“Flush:a reliable bulk transport protocol for multihop wireless networks,”in Proc.5th Int.Conf.on Embedded networked sensor systems,SenSys,2007,pp.351–365.
[51]J.Paek and https://www.wendangku.net/doc/7515472058.html,indan,“RCRT:Rate-controlled reliable transport
protocol for wireless sensor networks,”ACM https://www.wendangku.net/doc/7515472058.html,w.,vol.7, no.3,pp.20:1–20:45,oct2010.
[52]Y.Zhou,M.Lyu,J.Liu,and H.Wang,“PORT:a price-oriented reliable
transport protocol for wireless sensor networks,”in Proc.16th IEEE
20IEEE COMMUNICATIONS SURVEYS&TUTORIALS,ACCEPTED FOR PUBLICATION
Int.Symp.on Software Reliability Engineering,ISSRE,2005,pp.10 pp.–126.
[53]M.O.Rahman,M.M.Monowar, B.G.Choi,and C.S.Hong,
“An approach for congestion control in sensor network using priority based application,”in Proc.2nd Int.Conf.on Ubiquitous information management and communication,ICUIMC,2008,pp.430–435. [54]V.Gungor and O.Akan,“DST:delay sensitive transport in wireless
sensor networks,”in Proc.7th IEEE Int.Symp.on Computer Networks, ISCN,2006,pp.116–122.
[55]M.M.Monowar,M.O.Rahman,A.-S.K.Pathan,and C.S.Hong,
“Congestion control protocol for wireless sensor networks handling prioritized heterogeneous traf?c,”in Proc.5th Annu.Int.Conf.on Mobile and Ubiquitous Systems:Computing,Networking,and Services, Mobiquitous,2008,pp.17:1–17:8.
[56]S.Chen and N.Yang,“Congestion Avoidance Based on Lightweight
Buffer Management in Sensor Networks,”IEEE Trans.Parallel Distrib.
Syst.,vol.17,no.9,pp.934–946,september2006.
[57]R.Vedantham,R.Sivakumar,and S.-J.Park,“Sink-to-sensors conges-
tion control,”Ad Hoc Networks,vol.5,no.4,pp.462–485,2007. [58]M.Yaghmaee and D.Adjeroh,“A new priority based congestion
control protocol for Wireless Multimedia Sensor Networks,”in Proc.
Int.Symp.on a World of Wireless,Mobile and Multimedia Networks, WoWMoM,2008,pp.1–8.
[59]T.He,F.Ren,C.Lin,and S.Das,“Alleviating Congestion Using
Traf?c-Aware Dynamic Routing in Wireless Sensor Networks,”in Proc.5th Annu.IEEE Commun.Society Conf.on Sensor,Mesh and Ad Hoc Commun.and Networks,SECON,2008,pp.233–241. [60] F.Bian,S.Rangwala,and https://www.wendangku.net/doc/7515472058.html,indan,“Quasi-static Centralized Rate
Allocation for Sensor Networks,”in Proc.4th Annu.IEEE Commun.
Society Conf.on Sensor,Mesh and Ad Hoc Commun.and Networks, SECON,2007,pp.361–370.
[61]M.Rahman,M.Monowar,and C.Hong,“A Capacity Aware Data
Transport Protocol for Wireless Sensor Network,”in Proc.Int.Conf.
on Computational Science and Its Applications,ICCSA,ser.Lecture Notes in Computer Science,2009,pp.491–502.
[62] F.B.Hussain,Y.Cebi,and G.A.Shah,“A multievent congestion
control protocol for wireless sensor networks,”EURASIP J.Wirel.
https://www.wendangku.net/doc/7515472058.html,w.,pp.44:1–44:12,jan2008.
[63]J.-Y.Teo,Y.Ha,and C.-K.Tham,“Interference-Minimized Multipath
Routing with Congestion Control in Wireless Sensor Network for High-Rate Streaming,”IEEE Trans.Mobile Computing,vol.7,no.9,pp.
1124–1137,September2008.
[64] A.Sridharan and B.Krishnamachari,“Explicit and precise rate control
for wireless sensor networks,”in Proc.7th ACM Conf.Embedded Networked Sensor Systems,SenSys,November4–62009,pp.29–42.
[65]Y.Xiong,“MFACCS:A Congestion Control Scheme
for Wireless Sensor Networks,”Master’s thesis, Texas A&M University,2005.[Online].Available: https://www.wendangku.net/doc/7515472058.html,/viewdoc/summary?doi=10.1.1.133.6354 [66]P.Navaratnam,H.Cruickshank,and R.Tafazolli,“A link adaptive
transport protocol for multimedia streaming applications in multi hop wireless networks,”https://www.wendangku.net/doc/7515472058.html,w.Appl.,vol.13,no.3-4,pp.246–258, aug2008.
[67]M.C.Vuran,O.B.Akan.,and I.F.Akyildiz,“XLM:A Cross-Layer
Protocol for Wireless Sensor Networks,”in Proc.IEEE40th Annu.
https://www.wendangku.net/doc/7515472058.html,rmation Sciences and Systems,2006.
[68] A.Sridharan and B.Krishnamachari,“Max-min fair collision-free
scheduling for wireless sensor networks,”in Proc.IEEE Int.Conf.
on Performance,Computing,and Commun.,2004,pp.585–590. [69]S.Chen and Z.Zhang,“Localized algorithm for aggregate fairness in
wireless sensor networks,”in Proc.12th Annu.Int.Conf.on Mobile computing and networking,MobiCom,September23–262006,pp.
274–285.
[70]M.Zawodniok and S.Jagannathan,“Predictive Congestion Control
Protocol for Wireless Sensor Networks,”IEEE Trans.Wireless Com-mun.,vol.6,no.11,pp.3955–3963,November2007.
[71]R.Kumar,R.Crepaldi,H.Rowaihy,A.F.,G.Cao,M.Zorzi,and T.F.,
“Mitigating Performance Degradation in Congested Sensor Networks,”
IEEE Trans.Mobile Computing,vol.7,no.6,pp.682–697,jun2008.
[72]V.Enigo and V.Ramachandran,“An Energy Ef?cient Congestion
Control Protocol for Wireless Sensor Networks,”in IEEE Int.Advance Computing Conf.,IACC,6-7March2009,pp.1573–1578.
[73]S.Brahma,M.Chatterjee,and K.Kwiat,“Congestion control and
fairness in wireless sensor networks,”in Proc.8th IEEE Int.Conf.
on Pervasive Computing and Communications Workshops(PERCOM Workshops),2010,pp.413–418.
[74] B.Hull,K.Jamieson,and H.Balakrishnan,“Bandwidth Management
in Wireless Sensor Networks,”in Proc.1st Int.Conf.on Embedded networked sensor systems,Sensys,2003,pp.306–307.
[75]P.Hurni,T.Braun,B.Bhargava,and Y.Zhang,“Multi-hop Cross-
Layer Design in Wireless Sensor Networks:A Case Study,”in Proc.
IEEE Int.Conf.on Wireless and Mobile Computing,Networking and Commun.,WIMOB,2008,pp.291–296.
[76]Y.Li,L.Qiu,Y.Zhang,R.Mahajan,Z.Zhong,G.Deshpande,
and E.Rozner,“Effects of Interference on Wireless Mesh Networks: Pathologies and a Preliminary Solution,”in HotNets,2007.
[77]J.Kang,Y.Zhang,and B.Nath,“End-to-End Channel Capacity Mea-
surement for Congestion Control in Sensor Networks,”in Proc.2nd Int.
Workshop on Sensor and Actor Network Protocols and Applications, SANPA,2004.
[78]V.Vassiliou and C.Sergiou,“Performance Study of Node Placement
for Congestion Control in Wireless Sensor Networks,”in Proc.3rd Int.
Conf.on New Technologies,Mobility and Security(NTMS),Dec2009, pp.1–8.
[79] C.Sergiou and V.Vassiliou,“Alternative path creation vs data rate
reduction for congestion mitigation in wireless sensor networks,”in Proc.9th ACM/IEEE https://www.wendangku.net/doc/7515472058.html,rmation Processing in Sensor Networks,IPSN,April12–162010,pp.394–395.
[80]M.C.Vuran and V.C.Gungor,“On the interdependency of congestion
and contention in wireless sensor networks,”in Proc.SENMETRICS ’05,july2005.
[81]T.Melodia,M.C.Vuran,and D.Pompili,“The State of the Art in
Cross-layer Design for Wireless Sensor Networks,”in Proc.Eurongi Workshops On Wireless and Mobility,Springer Lecture Notes on Computer Science,LNCS388,july2005.
[82]V.C.Gungor,M.C.Vuran,and O.B.Akan,“On the cross-layer
interactions between congestion and contention in wireless sensor and actor networks,”Ad Hoc Netw.,vol.5,no.6,pp.897–909,Aug.2007.
[83]L.Eggert,“Congestion Control for the Application Protocol(CoAP),”
Internet Draft draft-eggert-core-congestion-control-01Expires July31, January272011.
[84]Z.Shelby,K.Hartke,and C.Bormann,“Constrained Application
Protocol(CoAP),”Internet-Draft will expire on December30,2013.
[85] E.Kohler,M.Handley,and S.Floyd,“Datagram Congestion Control
Protocol(DCCP),”RFC4340,March2006.
[86] E.T.Watteyne,M.Palattella,and L.Grieco,“Using IEEE802.15.4e
TSCH in an LLN context:overview,Problem Statement and Goals,”in draft-watteyne-6tsch-tsch-lln-context-01(work in progress),Nov.2013.
[87]P.E.Thubert,R.Assimiti,and T.Watteyne,“An Architecture for
IPv6over the TSCH mode of IEEE802.15.4e.”in draft-ietf-6tisch-architecture-00(work in progress),Nov.2013.
[88]P.Thubert,T.Watteyne,M.R.Palattella,X.Vilajosana,and Q.Wang,
“IETF6TSCH:Combining IPv6Connectivity with Industrial Perfor-mance,”in Proc.IEEE7th Int.Conf.Innovative Mobile and Internet Services in Ubiquitous Computing.,2013.
[89] E.X.Vilajosana and K.Pister,“Minimal6TiSCH Con?guration,”in
draft-ietf-6tisch-minimal-00.(work in progress),Nov.2013.
[90]T.Watteyne,A.Molinaro,M.Richichi,and M.Dohler,“From MANET
To IETF ROLL Standardization:A Paradigm Shift in WSN Routing Protocols,”IEEE Commun.Surveys Tutorials,vol.13,no.4,pp.688–707,2011.
[91] C.Wang,K.Sohraby,B.Li,M.Daneshmand,and Y.Hu,“A survey
of transport protocols for wireless sensor networks,”IEEE Network, vol.20,no.3,pp.34–40,2006.
[92] C.Wang,K.Sohraby,Y.Hu,B.Li,and W.Tang,“Issues of transport
control protocols for wireless sensor networks,”in Proc.Int.Conf.on Communications,Circuits and Systems,vol.1,2005,pp.422–426. [93]R.Chakravarthi, C.Gomathy,S.K.Sebastian,K.Pushparaj.,and
V.Mon,“A Survey on Congestion Control in Wireless Sensor Net-works,”Int.J.of Computer Science&Commun.,vol.1,no.1,pp.
161–164,January-June2010.
[94]J.Zhao,L.Wang,S.Li,X.Liu,Z.Yuan,and Z.Gao,“A Survey
of Congestion Control Mechanisms in Wireless Sensor Networks,”in Proc.6th Int.Conf.on Intelligent Information Hiding and Multimedia Signal Processing(IIH-MSP),2010,pp.719–722.
[95]M.Rahman,A.E.Saddik,and W.Gueaieb,“Wireless Sensor Network
Transport Layer:State of the Art,”Sensors,Lecture Notes Electrical Engineering,Springer Berlin Heidelberg,vol.21,pp.221–245,2008.
[96]H.Zhang and V.Naik,“Data Transport Control in Wireless Sensor
Networks,”Guide to Wireless Sensor Networks,Computer Communi-cations and Networks,Springer London,pp.239–259,2009.
[97] F.Shaikh, A.Khelil,and N.Suri,“A comparative study of data
transport protocols in wireless sensor networks,”in Proc.IEEE Int.
Symp.on a World of Wireless,Mobile and Multimedia Networks, WoWMoM,2008,pp.1–9.
家装用尺寸一览表
家装用尺寸一览表 Revised by Hanlin on 10 January 2021
家装用尺寸一览表 ▌标准入户门洞0.9m*2m， ▌房间门洞0.9m*2m， ▌厨房门洞0.8m*2m， ▌卫生间门洞0.7m*2m ▌客厅：长沙发：240*90*75cm长方形茶几：130*70*45cm电视柜：200*50*180cm电视离沙发：3m电视高度与电视柜高差：40到120cm走道宽度：100至120cm ▌厨房：橱柜操作台：台面高80cm左右面积90*46（最小20最大60）cm吊柜：离台面60cm左右高度在145cm到150cm餐桌：餐桌高：750—790mm。餐椅高；450—500mm。圆桌直径：二人500mm．二人800mm，四人900mm，五人1100mm，六人1100-1250mm，八人1300mm，十人l500mm，十二人1800mm。方餐桌尺寸：二人700×850(mm)，四人1350×850(mm)，八人2250×850(mm) ▌卫生间：浴缸长度：一般有三种1220、1520、1680mm；宽：720mm，高：450mm。坐便：750×350(mm)。冲洗器：690×350(mm)。盟洗盆：550×410(mm)。淋浴器高：2100mm。化妆台：长：1350mm；宽450mm。 ▌卧室：标准双人床尺寸：150*190、150*200厘米，被套的尺寸应配180*215和200*230之间的。加大双人床尺寸：180*200厘米，被套一般为200*230或220*240。床头柜宽：400毫米-600毫米，深：350毫米-450毫米高：500毫米-700毫米。衣柜：柜门尺寸，单
2017最完整家装尺寸大全
家具设计的基本尺寸（单位：cm） 衣橱：深度：一般60~65；推拉门：70，衣橱门宽度：40~65 推拉门：75~150，高度：190~240 矮柜：深度：35~45，柜门宽度：30-60 电视柜：深度：45-60，高度：60-70 单人床：宽度：90，105，120；长度：180，186，200，210 双人床：宽度：135，150，180；长度180，186，200，210 圆床：直径：186，212.5，242.4（常用） 室内门：宽度：80-95，医院120；高度：190，200，210，220，240 厕所、厨房门：宽度：80，90；高度：190，200，210 窗帘盒：高度：12-18；深度：单层布12；双层布16-18（实际尺寸） 沙发：单人式：长度：80-95，深度：85-90；坐垫高：35-42；背高：70-90 双人式：长度：126-150；深度：80-90 三人式：长度：175-196；深度：80-90 四人式：长度：232-252；深度80-90 茶几：小型，长方形：长度60-75，宽度45-60，高度38-50（38最佳） 中型，长方形：长度120-135；宽度38-50或者60-75 正方形：长度75-90，高度43-50 大型，长方形：长度150-180，宽度60-80，高度33-42（33最佳） 圆形：直径75，90，105，120；高度：33-42 方形：宽度90，105，120，135，150；高度33-42 书桌：固定式：深度45-70（60最佳），高度75 活动式：深度65-80，高度75-78 书桌下缘离地至少58；长度：最少90（150-180最佳） 餐桌：高度75-78（一般），西式高度68-72，一般方桌宽度120，90，75；长方桌宽度80，90，105，120；长度150，165，180，210，240 圆桌：直径90，120，135，150，180 书架：深度25-40（每一格），长度：60-120；下大上小型下方深度35-45，高度80-90活动未及顶高柜：深度45，高度180-200 木隔间墙厚：6-10；内角材排距：长度（45-60）*90
家装基本尺寸大全
家具设计的基本尺寸（单位：厘米） 衣橱：深度：一般60~65；推拉门：70，衣橱门宽度：40~65 推拉门：75~150，高度：190~240 矮柜：?深度：35~45，柜门宽度：30-60 电视柜：深度：45-60，高度：60-70 单人床：宽度：90，105，120；长度：180，186，200，210 双人床：宽度：135，150，180；长度180，186，200，210 圆床：?直径：186，，（常用） 室内门：宽度：80-95，医院120；高度：190，200，210，220，240 厕所、厨房门：宽度：80，90；高度：190，200，210 窗帘盒：高度：12-18；深度：单层布12；双层布16-18（实际尺寸） 沙发：单人式：长度：80-95，深度：85-90；坐垫高：35-42；背高：70-90双人式：长度：126-150；深度：80-90 三人式：长度：175-196；深度：80-90 四人式：长度：232-252；深度80-90 茶几：小型，长方形：长度60-75，宽度45-60，高度38-50（38最佳） 中型，长方形：长度120-135；宽度38-50或者60-75 正方形：?长度75-90，高度43-50 大型，长方形：长度150-180，宽度60-80，高度33-42（33最佳）
圆形：直径75，90，105，120；高度：33-42 方形：宽度90，105，120，135，150；高度33-42 书桌：固定式：深度45-70（60最佳），高度75 活动式：深度65-80，高度75-78 书桌下缘离地至少58；长度：最少90（150-180最佳） 餐桌：高度75-78（一般），西式高度68-72，一般方桌宽度120，90，75； 长方桌宽度80，90，105，120；长度150，165，180，210，240 圆桌：直径90，120，135，150，180 书架：深度25-40（每一格），长度：60-120；下大上小型下方深度35-45，高度80-90 活动未及顶高柜：深度45，高度180-200 木隔间墙厚：6-10；内角材排距：长度（45-60）*90 桌类家具高度尺寸：７００ｍｍ、７２０ｍｍ、７４０ｍｍ、７６０ｍｍ四个规格； 椅凳类家具的座面高度：４００ｍｍ、４２０ｍｍ、４４０ｍｍ三个规格。 桌椅高度差应控制在２８０至３２０ｍｍ范围内。
家装各种最佳尺寸标准大全!
提供全方位装修指南，装修设计知识、丰富设计案例！ 家装各种最佳尺寸标准大全! 家装最实际的规格尺寸 标准红砖24*11.5*53； 标准入户门洞0.9米*2米， 房间门洞0.9米*2米， 厨房门洞0.8米*2米， 卫生间门洞0.7米*2米， 标准水泥50kg/袋。 厨房 1.吊柜和操作台之间的距离应该是多少？ 60厘米。 从操作台到吊柜的底部，您应该确保这个距离。这样，在您可以方便烹饪的同时，还可以在吊柜里放一些小型家用电器。 2.在厨房两面相对的墙边都摆放各种家具和电器的情况下，中间应该留多大的距离才不会影响在厨房里做家务？ 120厘米。 为了能方便地打开两边家具的柜门，就一定要保证至少留出这样的距离。 150厘米。 这样的距离就可以保证在两边柜门都打开的情况下，中间再站一个人。 3.要想舒服地坐在早餐桌的周围，凳子的合适高度应该是多少？ 80厘米。 对于一张高110厘米的早餐桌来说，这是摆在它周围凳子的理想高度。因为在桌面和凳子之间还需要30厘米的空间来容下双腿。 4.吊柜应该装在多高的地方？ 145至150厘米。
提供全方位装修指南，装修设计知识、丰富设计案例！ 餐厅 1. 一个供六个人使用的餐桌有多大？ 2. 120厘米。 这是对圆形餐桌的直径要求。 140*70厘米。 这是对长方形和椭圆形捉制的尺寸要求。 2.餐桌离墙应该有多远？ 80厘米。 这个距离是包括把椅子拉出来，以及能使就餐的人方便活动的最小距离。 3.一张以对角线对墙的正方形桌子所占的面积要有多大？ 180*180平方厘米。 这是一张边长90厘米，桌角离墙面最近距离为40厘米的正方形桌子所占的最小面积。 4.桌子的标准高度应是多少？ 72厘米。 这是桌子的中等高度，而椅子是通常高度为45厘米。 5.一张供六个人使用的桌子摆起居室里要占多少面积？ 300*300厘米。 需要为直径120厘米的桌子留出空地，同时还要为在桌子四周就餐的人留出活动空间。这个方案适合于那种大客厅，面积至少达到600*350厘米。 6.吊灯和桌面之间最合适的距离应该是多少？ 70厘米。 这是能使桌面得到完整的、均匀照射的理想距离。 卫生间 1．卫生间里的用具要占多大地方？ 马桶所占的一般面积： 37厘米×60厘米。
装修预留的尺寸标准
【精华】室内装修，必须预留的最佳尺寸标准大全 2014-08-29筑龙房地产筑龙房地产 阅读引语 强烈推荐大家存的一份装修预留尺寸标准！！非常实用！！ 现在新房子的设计一般都会交给专门的设计师来做，但哪怕是专业设计师制作的设计图稿，没有实地接触的设计师可能还会存在一些设计尺寸上的死角。另 外，落实图稿的是施工队的工人，工人往往疏忽大意就会犯错。于是房子装修完了，总是小错误不断。因此小哥觉得大家有必要存一份尺寸标准，监工时要用起来 哦！且看且分享吧！ PART1：【客 厅】 【面积：20平方米～40平方米】 客厅是居室的门面，可以说对家具尺寸的要求是最严格的，多大的沙发配多大的茶几，多远的距离适合摆放电视等等，别看都是一些小数字，却足以令你的客厅成为一个舒适协调的地方。
电视组合柜的最小尺寸？ 【200×50×180厘米】 对于小户型的客厅，电视组合柜是非常实用的，这种类型的家具一般都是由大小不同的方格组成，上部比较适合摆放一些工艺品，柜体厚度至少要保持30厘米；而下部摆放电视的柜体厚度则至少要保持50厘米，同时在选购电视柜时也要考虑组合柜整体的高度和横宽与墙壁的面宽是否协调。 长沙发或是扶手沙发的椅背应该有多高？ 【85至90厘米】 沙发是用来满足人们的放松与休息的需求，所以舒适度是最重要的，这样的高度可以将头完全放在*背上，让颈部得到充分放松。如果沙发的*背和扶手过低，建议增加一个*垫来获得舒适度，如果空间不是特别宽敞，沙发应该尽量靠墙摆放。 扶手沙发与电视机之间应该预留多大的距离？
【3米左右】 这里所指的是在一个29英寸的电视与扶手沙发或和长沙发之间最短的距离，此外，摆放电视机的柜面高度应该在40厘米到120厘米之间，这样才能让看者非常舒适。 与容纳三个人的沙发搭配，多大的茶几合适呢？ 【120×70×45厘米或100×100×45厘米】 在沙发的体积很大或是两个长沙发摆在一起的情况下，矮茶几就是很好的选择，茶几的高度最好和沙发坐垫的位置持平。 目前市场上较为流行的是一种低矮的方几，材质多为实木或实木贴皮的，质感较好。 细节补充： 照明灯具距桌面的高度，白炽灯泡60瓦为100厘米，40瓦为65厘米，25瓦为50厘米，15瓦为30厘米；日光灯距桌面高度，40瓦为150厘米，30瓦为140厘米，20瓦为110厘米，8瓦为55厘米。 PART2：【餐 厅】 【面积：10平方米～20平方米】 用餐的地方是一家人团聚最多的地方，通常也是居室中较为拥挤的一个空间，因为有较多的餐椅需要放置，也是家人同时集中的地方，所以它的尺寸就更要精打细算才能使餐厅成为一个温馨的地方。
完整家装尺寸大全
家具设计地基本尺寸（单位：） 衣橱：深度：一般；推拉门：，衣橱门宽度： 推拉门：，高度： 矮柜：深度：，柜门宽度： 电视柜：深度：，高度： 单人床：宽度：，，；长度：，，， 双人床：宽度：，，；长度，，， 圆床：直径：，，（常用） 室内门：宽度：，医院；高度：，，，， 厕所、厨房门：宽度：，；高度：，， 窗帘盒：高度：；深度：单层布；双层布（实际尺寸） 沙发：单人式：长度：，深度：；坐垫高：；背高： 双人式：长度：；深度： 三人式：长度：；深度： 四人式：长度：；深度 茶几：小型，长方形：长度，宽度，高度（最佳） 中型，长方形：长度；宽度或者 正方形：长度，高度 大型，长方形：长度，宽度，高度（最佳） 圆形：直径，，，；高度： 方形：宽度，，，，；高度 书桌：固定式：深度（最佳），高度 活动式：深度，高度 书桌下缘离地至少；长度：最少（最佳） 餐桌：高度（一般），西式高度，一般方桌宽度，，；长方桌宽度，，，；长度，，，，圆桌：直径，，，， 书架：深度（每一格），长度：；下大上小型下方深度，高度 活动未及顶高柜：深度，高度 木隔间墙厚：；内角材排距：长度（）* 室内常用尺寸 、墙面尺寸 ()踢脚板高；—. ()墙裙高：—. ()挂镜线高：—(画中心距地面高度). ．餐厅
() 餐桌高：—. () 餐椅高；—. () 圆桌直径：二人．二人，四人，五人，六人，八人，十人，十二人. () 方餐桌尺寸：二人×()，四人×()，八人×()， () 餐桌转盘直径；—. 餐桌间距：(其中座椅占)应大于. () 主通道宽：—. 内部工作道宽：—. () 酒吧台高：—，宽. () 酒吧凳高；一. 在客厅 .长沙发与摆在它面前地茶几之间地正确距离是多少？ 厘米 在一个（**高厘米）地长沙发面前摆放一个（**高厘米）地长方形茶几是非常舒适地.两者之间地理想距离应该是能允许你一个人通过地同时又便于使用，也就是说不用站起来就可以方便地拿到桌上地杯子或者杂志. b5E2R。 .一个能摆放电视机地大型组合柜地最小尺寸应该是多少？ **高厘米 这种类型地家具一般都是由大小不同地方格组成，高处部分比较适合用来摆放书籍，柜体厚度至少保持厘米；而低处用于摆放电视地柜体厚度至少保持厘米.同时组合柜整体地高度和横宽还要考虑与墙壁地面积相协调..如果摆放可容纳三、四个人地沙发，那么应该选择多大地茶几来搭配呢？ **高厘米 在沙发地体积很大或是两个长沙发摆在一起地情况下，矮茶几就是很好地选择，高度最好和沙发坐垫地位置持平. .在扶手沙发和电视机之间应该预留多大地距离？ 米 这里所指地是在一个英寸地电视与扶手沙发或长沙发之间最短地距离.此外，摆放电视机地柜面高度应该在厘米到厘米之间，这样才能使观众保持正确地坐姿. .摆在沙发边上茶几地理想尺寸是多少？ 方形：**高厘米. 椭圆形：*高厘米. 放在沙发边上地咖啡桌应该有一个不是特别大地桌面，但要选那种较高地类型，这样即使坐着地时候也能方便舒适地取到桌上地东西. p1Ean。 .两个面对面放着地沙发和摆放在中间地茶几一共需要占据多大地空间？ 两个双人沙发（规格 **高厘米）和茶几（规格**高厘米）之间应相距厘米. .长沙发或是扶手沙发地地靠背应该有多高？
装修常用家具尺寸表
装修常用家具尺寸 在工地 1、标准红砖23*11*6；标准入户门洞0.9米*2米，房间门洞0.9米*2米，厨房门洞0.8米*2米，卫生间门洞0.7米*2米，标准水泥50kg/袋。 在厨房 1.吊柜和操作台之间的距离应该是多少？ 60厘米。 从操作台到吊柜的底部，您应该确保这个距离。这样，在您可以方便烹饪的同时，还可以在吊柜里放一些小型家用电器。 2.在厨房两面相对的墙边都摆放各种家具和电器的情况下，中间应该留多大的距离才不会影响在厨房里做家务？ 120厘米。 为了能方便地打开两边家具的柜门，就一定要保证至少留出这样的距离。 150厘米。 这样的距离就可以保证在两边柜门都打开的情况下，中间再站一个人。 3.要想舒服地坐在早餐桌的周围，凳子的合适高度应该是多少？ 80厘米。 对于一张高110厘米的早餐桌来说，这是摆在它周围凳子的理想高度。因为在桌面和凳子之间还需要30厘米的空间来容下双腿。
4.吊柜应该装在多高的地方？ 145至150厘米。 这个高度可以使您不用垫起脚尖就能打开吊柜的门。 在餐厅 1.一个供六个人使用的餐桌有多大？ 120厘米。 这是对圆形餐桌的直径要求。 140*70厘米。 这是对长方形和椭圆形捉制的尺寸要求。 2.餐桌离墙应该有多远？ 80厘米。 这个距离是包括把椅子拉出来，以及能使就餐的人方便活动的最小距离。 3.一张以对角线对墙的正方形桌子所占的面积要有多大？ 180*180平方厘米 这是一张边长90厘米，桌角离墙面最近距离为40厘米的正方形桌子所占的最小面积。 4.桌子的标准高度应是多少？ 72厘米。
这是桌子的中等高度，而椅子是通常高度为45厘米。 5.一张供六个人使用的桌子摆起居室里要占多少面积？ 300*300厘米。 需要为直径120厘米的桌子留出空地，同时还要为在桌子四周就餐的人留出活动空间。这个方案适合于那种大客厅，面积至少达到600*350厘米。 6.吊灯和桌面之间最合适的距离应该是多少？ 70厘米。 这是能使桌面得到完整的、均匀照射的理想距离。 在卫生间 1．卫生间里的用具要占多大地方？ 马桶所占的一般面积：37厘米×60厘米 悬挂式或圆柱式盥洗池可能占用的面积：70厘米×60厘米 正方形淋浴间的面积：80厘米×80厘米 浴缸的标准面积：160厘米×70厘米 2．浴缸与对面的墙之间的距离要有多远？ 100厘米。想要在周围活动的话这是个合理的距离。即使浴室很窄，也要在安装浴缸时留出走动的空间。总之浴缸和其他墙面或物品之间至少要有60厘米的距离。
家装尺寸数据大全
干货│家装尺寸数据大全，大家快掏 出小本本记好了！ 一、那些在工地的数据 (3) 二、那些在客厅涉及的家装数据 (4) 三、那些在厨房涉及到的家装数据 (8) 四、那些在餐厅涉及到的家装数据 (9) 五、那些在卫生间涉及到的家装数据 (11)
装修从来不是一件一蹴而就的事 它是一项关乎未来几十年生活质量的细活儿 可以精确到一丝一毫 因此了解一些家具尺寸的数据是非常必要的常识 为了有效避免以下惨烈装修车祸现场 比如心爱的沙发多出一块经常绊倒人 又比如一眼看中的床卧室竟然放不下······
下面各位装修的宝宝赶紧来围观一起涨姿势 一、那些在工地的数据 1、标准红砖23*11*6; 2、标准入户门洞0.9米*2米， 3、房间门洞0.9米*2米， 4、厨房门洞0.8米*2米， 5、卫生间门洞0.7米*2米， 6、标准水泥50kg/袋。
二、那些在客厅涉及的家装数据 1.长沙发与摆放在它面前的茶几之间的正确距离是多少？ 30厘米在一个（240*90*75高厘米）的长沙发面前摆放一个（130*70*45高厘米）的长方形茶几是非常舒适的。两者之间的理想距离应该是能允许你一个人通过的同时又便于使用，也就是说不用站起来就可以方便地拿到桌上的杯子或者杂志。 2.一个能摆放电视机的大型组合柜的最小尺寸应该是多少？ 200*50*180厘米这种类型的家具一般都是由大小不同的方格组成，高处部分比较适合用来摆放书籍，柜体厚度至少保持30厘米；而低处用于摆放电视的柜体
厚度至少保持50厘米。同时组合柜整体的高度和横宽还要考虑与墙壁的面积相协调。 3.如果摆放可容纳三、四个人的沙发，那么应该选择多大的茶几来搭配呢？140*70*45高厘米。在沙发的体积很大或是两个长沙发摆在一起的情况下，矮茶几就是很好的选择，高度最好和沙发坐垫的位置持平。 4.在扶手沙发和电视机之间应该预留多大的距离？ 3米。这里所指的是在一个25英寸的电视与扶手沙发或长沙发之间最短的距离。此外，摆放电视机的柜面高度应该在40厘米到120厘米之间，这样才能使观众保持正确的坐姿。
淘宝店铺装修尺寸大全(终极版)
1.商品图片的尺寸：宽500*高500像素，大小在120KB以内，要求JPG或GIF格式，到发布宝贝页面上上传图片。最好大于312*310px 2.店标图片的尺寸：宽100*高100像素，大小在80K以内，支持JPG或GIF格式，动态或静态的图片均可。上传步骤：“管理我的店铺”-“基本设置”-“店标”-“浏览”-“确定” 3.宝贝描述图片的尺寸：没有特殊要求，可根据需要宽500*高500像素，大小在100K以内，这样图片的打开速度较快。要求JPG或GIF格式，静态或动态均可。将图片上传到电子相册，再复制到商品页面中去。 4.公告栏图片的尺寸：宽不超过480像素，长度不限制，大小在120KB以内GIF或JPG格式，动态或者静态均可。上传“管理我的店铺”-“基本设置”-“公告栏”-“确定”。 5.宝贝分类图片尺寸：宽不超过165,长度不限制，大小在50KB以内，要求GIF或JPG格式，动态或者静态均可，先将图片上传到电子相册得到一个缩短网址后进入“管理我的店铺”-“基本设置”-“宝贝分类” 6.旺旺头像图片尺寸：宽120*高120像素，大小在100KB以内，格式为JPG或GIF，动态或者静态均可。 7.论坛头像图片尺寸：最大为宽120*高120像素，大小在100KB以内，GIF或者JPG格式，动态或者静态图片均可。上传方法“我的淘宝”-“个人空间”-“修改资料”-“上传新头像”。 8.论坛签名档图片尺寸：宽468*高60像素，大小在100KB以内，JPG或者GIF格式，动态或者静态均可，上传“我的淘宝”-“个人空间 淘宝店铺装修最佳尺寸 普通店铺 1.店标 大小：100*100px <=80k 代码：无（图片做好后直接上传） 格式：jpg、gif 设置：管理我的店铺—基本设置—店标—浏览—选择本地做好店标文件 2. 店铺公告尺寸：320*400 3.宝贝分类尺寸:88*88和88*30（宝贝分类含3个) 4.店铺介绍尺寸：600*450 5.计数器尺寸:137*94 6.论坛签名尺寸:468*60
2019最完整家装尺寸大全!
家里装修，最重要的是什么？ 不是缤纷夺目的软装搭配设计，也不是酷炫十足的多功能变化装置，而是严格把控每一个细节尺寸，保证在装修完毕之后，根本的硬件设施合乎人体工程学的基本要求，让家里每个人住着舒适开心，这才是最重要的。这里，不仅整理出了完善的室内常见尺寸，更有一些独具风格的创意设计尺寸，让家装不再是难事。 室内常见家具的基本尺寸（单位：cm） 客厅篇 沙发： 单人式：长度：80-95，深度：85-90；坐垫高：35-42；背高：70-90 双人式：长度：126-150；深度：80-90 三人式：长度：175-196；深度：80-90 四人式：长度：232-252；深度80-90 茶几： 小型，长方形：长度60-75，宽度45-60，高度38-50（38最佳） 中型，长方形：长度120-135；宽度38-50或者60-75 正方形：长度75-90，高度43-50 大型，长方形：长度150-180，宽度60-80，高度33-42（33最佳） 圆形：直径75，90，105，120；高度：33-42 方形：宽度90，105，120，135，150；高度33-42 墙面尺寸： (1)踢脚板高：8—20 (2)墙裙高：80—150
(3)挂镜线高：160—180(画中心距地面高度) 厕所、厨房门：宽度：80，90；高度：190，200，210 窗帘盒：高度：12-18；深度：单层布12；双层布16-18（实际尺寸） 厨房餐厅篇 餐桌高：75—79 餐椅高：45—50 圆桌直径：二人50，二人80，四人90，五人110，六人110-125，八人130，十人l50，十二人180。 方餐桌尺寸：二人70×85，四人135×85，八人225×85 餐桌转盘直径：70—80 餐桌间距：(其中座椅占50)应大于50 主通道宽：120—130 内部工作道宽：60—90 酒吧台高：90—l05，宽50 酒吧凳高：60一75 卧室篇 衣橱：深度：一般60~65；推拉门：70，衣橱门宽度：40~65 推拉门：75~150，高度：190~240 矮柜：深度：35~45，柜门宽度：30-60 电视柜：深度：45-60，高度：60-70 单人床：宽度：90，105，120；长度：180，186，200，210 双人床：宽度：135，150，180；长度：180，186，200，210 圆床：直径：186，212.5，242.4（常用） 室内门：宽度：80-95，医院120；高度：190，200，210，220，240 书桌：固定式：深度45-70（60最佳），高度75 活动式：深度65-80，高度75-78 书桌下缘离地至少58；长度：最少90（150-180最佳） 圆桌：直径90，120，135，150，180 书架：深度25-40（每一格），长度：60-120；下大上小型下方深度35-45，高度80-90 木隔间墙厚：6-10；内角材排距：长度（45-60）*90 室内常用尺寸（单位：cm） 客厅篇 1.长沙发与茶几之间的距离 =30cm 在一个（240*90*75）的长沙发面前摆放一个（130*70*45）的长方形茶几是非常舒适的。是能允许一人通过的同时又便于使用的理想距离。 2.一个能摆放电视机的大型组合柜的最小尺寸=200*50*180 这种类型的家具一般都是由大小不同的方格组成，高处部分比较适合用来摆放书籍，柜体厚度至少保持30厘米；而低处用于摆放电视的柜体厚度至少保持50厘米。同时组合柜整体的高度和横宽还要考虑与墙壁的面积相协调。 3.如果摆放可容纳三、四个人的沙发，应该选择搭配的茶几大小=140*70*45
家装设计尺寸标准
家具设计标准尺寸 家具设计的基本尺寸（单位：厘米 衣橱：深度：一般60~65；推拉门：70，衣橱门宽度：40~65 推拉门：75~150，高度：190~240 矮柜：深度：35~45，柜门宽度：30-60 电视柜：深度：45-60，高度：60-70 单人床：宽度：90，105，120；长度：180，186，200，210 双人床：宽度：135，150，180；长度180，186，200，210 圆床：直径：186，212.5，242.4（常用） 室内门：宽度：80-95，医院120；高度：190，200，210，220，240 厕所、厨房门：宽度：80，90；高度：190，200，210 窗帘盒：高度：12-18；深度：单层布12；双层布16-18（实际尺寸） 沙发：单人式：长度：80-95，深度：85-90；坐垫高：35-42；背高：70-90 双人式：长度：126-150；深度：80-90 三人式：长度：175-196；深度：80-90 四人式：长度：232-252；深度80-90 茶几：小型，长方形：长度60-75，宽度45-60，高度38-50（38最佳 中型，长方形：长度120-135；宽度38-50或者60-75 正方形：长度75-90，高度43-50 大型，长方形：长度150-180，宽度60-80，高度33-42（33最佳） 圆形：直径75，90，105，120；高度：33-42 方形：宽度90，105，120，135，150；高度33-42 书桌：固定式：深度45-70（60最佳），高度75 活动式：深度65-80，高度75-78 书桌下缘离地至少58；长度：最少90（150-180最佳） 餐桌：高度75-78（一般），西式高度68-72，一般方桌宽度120，90，75； 长方桌宽度80，90，105，120；长度150，165，180，210，240 圆桌：直径90，120，135，150，180 书架：深度25-40（每一格），长度：60-120；下大上小型下方深度35-45，高度80-90 活动未及顶高柜：深度45，高度180-200 木隔间墙厚：6-10；内角材排距：长度（45-60）*90 室内家具尺寸标准大全 ●电视柜尺寸： 电视组合柜最小尺寸：2000×500×1800毫米。 电视组合柜厚度：上部至少要300毫米，下部摆放电视的柜体至少要500毫米。电视柜面高度：在400—1200毫米，另一说在400－520毫米，又一说600—700毫米。电视柜：深度450—600毫米，高度600-700毫米。●沙发尺寸：
家装应该知道的尺寸讲解
客厅尺寸篇（单位：mm） 01、沙发尺寸：一般深度800~900、坐位高350~420、背高700~900 单人式：长度：800-950，深度：850-900坐垫高：350-420；背高：70-90 双人式：长度：1260-1500；深度：800-900 三人式：长度：1750-1960；深度：800-900 四人式：长度：2320-2520；深度：800-900 02、茶几尺寸：茶几高度一般在330~420，但边角茶几有时稍高一些，为430~500 03、沙发和茶几之间的距离一般控制在300比较合适 04、一般电视机和沙发之间最短距离控制在3000 05、放置台式电视机的柜台高度，一般控制在400到1200之间 06、液晶电视机壁挂高度一般控制在电视机屏幕的中心点和观看电视时的视线平行，一般在1100，常规控制在1000-1500 餐厅尺寸篇（单位：mm） 一、餐桌尺寸 圆桌直径：二人500、三人800、四人900、五人1100、六人1200 （前几种规格圆桌人均占有弧长为600－800，以满负荷使用计算，一般固定其尺寸来使用） 八人1300-1400，十人1500-l600，十二人1800-2000 （此类推下去规格，人均占弧长控制在500-550，考虑非满负荷使用状况（餐桌转盘直径；700—800） 方桌尺寸: 此只探讨长条方桌，因正方方桌可通过长条方桌来推算其所需尺寸：一般短边控制在800－850， 长边则按人均占有计算：控制在550-700，接近700为佳。 二、餐桌一般高：750—790，餐椅一般高；450—500mm 三、酒吧台高一般：900—l050，宽500，酒吧凳高；600一750
家装尺寸常识
家居装修设计常规尺寸大全【人体工程学尺寸】 一、人体工程学尺寸参考【单位：cm】 1、体重：（男：68.9 女：56.7） 2、身高：（男：173.5 女：159.8） 3、座直臀至头顶的高度：（男： 90.7 女：84.8） 4、两肘间的宽度：（男：41.9 女：38.4） 5、肘下支撑物的高度：（男：24.1 女：23.4） 6、座姿大腿的高度：（男：14.5 女：13.7） 7、座姿膝盖至地面的高度：（男：54.4 女：49.8） 8、坐姿臀部至腿弯的长度：（男：49.0 女：48.0） 9、坐姿臀宽：（男：35.6 女：36.3） 10、活动空间（可蹲空间）男：1220～1470 女：1160～1320 【家装】 一、常用室内基本尺寸【单位：mm】 1、支撑墙体：厚度2400 2、室内隔墙断墙体：厚度1200 3、木隔间墙厚：60～100——内角材排距：长度（45～60）*90 4、窗帘盒：高度：120～180——深度：单层布120—双层布160～180 5、玄关：宽1000——墙厚2400 6、阳台：宽1400～1600——长3000～4000（一般与客厅的长度相同） 7、踏步：高1500～1600——长990～1150——宽250 扶手宽100——扶手间距200——中间的休息平台宽1000 8、踢脚板高：80～200 9、墙裙高：800～1500 10、挂镜线高：1600～1800(画中心距地面高度) 11、楼梯:850～1000 12、栏杆:高度：800～1100 13、房间内通道: 宽度：650（最小） 14、餐桌后通道：宽度：750 (其中座椅占500mm) 15、人肩宽520（400～450不能通过），可通行距离760～910 16、过道:宽度：900～1200
家装用尺寸一览表
家装用尺寸一览表标准化管理处编码[BBX968T-XBB8968-NNJ668-MM9N]
家装用尺寸一览表 ▌标准入户门洞*2m， ▌房间门洞*2m， ▌厨房门洞*2m， ▌卫生间门洞*2m ▌客厅：长沙发：240*90*75cm 长方形茶几：130*70*45cm电视柜：200*50*180cm 电视离沙发：3m 电视高度与电视柜高差：40到120cm 走道宽度：100至120cm ▌厨房：橱柜操作台：台面高80cm左右面积90*46（最小20最大60）cm 吊柜：离台面60cm左右高度在145cm到150cm餐桌：餐桌高：750—790mm。餐椅高；450— 500mm。圆桌直径：二人500mm．二人800mm，四人900mm，五人1100mm，六人1100-1250mm，八人1300mm，十人l500mm，十二人1800mm。方餐桌尺寸：二人700× 850(mm)，四人1350×850(mm)，八人2250×850(mm) ▌卫生间：浴缸长度：一般有三种1220、1520、1680mm；宽：720mm，高：450mm。坐便：750×350(mm)。冲洗器：690×350(mm)。盟洗盆：550×410(mm)。淋浴器高：2100mm。化妆台：长：1350mm；宽450 mm。 ▌卧室：标准双人床尺寸：150*190、150*200厘米，被套的尺寸应配180*215和200*230之间的。加大双人床尺寸：180*200厘米，被套一般为200*230或220*240。床头柜宽：400毫米-600毫米，深：350毫米-450毫米高：500毫米-700毫米。衣柜：柜门尺寸，单扇一门宽度不超过1200mm，高度不超过2400mm。挂衣区尺寸，上衣区高度在100cm-120cm，不低于90cm，宽度在40cm；长衣区是140cm-150cm指间，不低于130cm，宽度在40cm。裤架尺寸。柜子的深度一般在600-650mm之间，那么裤架的深度范围在490- 540mm，宽度不限。 ▌灯具：大吊灯最小高度：2400mm。壁灯高：1500—1800mm。反光灯槽最小直径：等于或大于灯管直径两倍。壁式床头灯高：1200—1400mm。照明开关高：1000mm。 ▌插座、开关：
最完整家装尺寸大全
最完整家装尺寸大全 最完整家装尺寸大全 最完整家装尺寸大全 衣橱：深度：一般60~65;推拉门：70,衣橱门宽度：40~65 推拉门：75~150,高度：190~240 矮柜：深度：35~45,柜门宽度：30-60 电视柜：深度：45-60,高度：60-70 单人床：宽度：90,105,120;长度：200,210 双人床：宽度：135,150,180;长度：200,210 圆床：直径：186,212.5,242.4（常用） 室内门：宽度：80-95,医院120;高度：190,200,210,220,240 厕所、厨房门：宽度：80,90;高度：190,200,210 窗帘盒：高度：12-18;深度：单层布12;双层布16-18（实际尺寸） 沙发：单人式：长度：80-95,深度：85-90;坐垫高：35-42; 背高：70-90 双人式：长度：126-150;深度：80-90 三人式：长度：175-196;深度：80-90 四人式：长度：232-252;深度80-90 茶几：小型，长方形：长度60-75,宽度45-60,高度38-50（38最佳）
中型，长方形：长度120-135;宽度38-50或者60-75 正方形：长度75-90,高度43-50 大型，长方形：长度150-180,宽度60-80,高度33-42（33最佳） 圆形：直径75,90,105,120;高度：33-42 方形：宽度90,105,120,135,150;高度33-42 书桌：固定式：深度45-70（60最佳），高度75 活动式：深度65-80,高度75-78 书桌下缘离地至少58;长度：最少90（150-180最佳） 餐桌：高度75-78（一般），西式高度68-72,一般方桌宽度120,90,75;长方桌宽度80,90,105,120;长度 150,165,180,210,240 圆桌：直径90,120,135,150,180 书架：深度25-40（每一格），长度：60-120;下大上小型下方深度35-45,高度80-90 活动未及顶高柜：深度45,高度180-200 木隔间墙厚：6-10;内角材排距：长度（45-60）*90 1 室内常用尺寸 1、墙面尺寸 （1）踢脚板高；80—200mm. （2）墙裙高：800—1500mm.
家装的标准尺寸大全
在工地 1、标准红砖23*11*6； 标准入户门洞0.9米*2米， 房间门洞0.9米*2米， 厨房门洞0.8米*2米， 卫生间门洞0.7米*2米， 标准水泥50kg/袋。 在厨房 1.吊柜和操作台之间的距离应该是多少？ 60厘米。 从操作台到吊柜的底部，您应该确保这个距离。这样，在您可以方便烹饪的同时，还可以在吊柜里放一些小型家用电器。 2.在厨房两面相对的墙边都摆放各种家具和电器的情况下，中间应该留多大的距离才不会影响在厨房里做家务？ 120厘米。 为了能方便地打开两边家具的柜门，就一定要保证至少留出这样的距离。 150厘米。 这样的距离就可以保证在两边柜门都打开的情况下，中间再站一个人。 3.要想舒服地坐在早餐桌的周围，凳子的合适高度应该是多少？ 80厘米。 对于一张高110厘米的早餐桌来说，这是摆在它周围凳子的理想高度。因为在桌面和凳子之间还需要30厘米的空间来容下双腿。 4.吊柜应该装在多高的地方？ 145至150厘米。 在餐厅 1.一个供六个人使用的餐桌有多大？ 120厘米。
这是对圆形餐桌的直径要求。 140*70厘米。 这是对长方形和椭圆形捉制的尺寸要求。 2.餐桌离墙应该有多远？ 80厘米。 这个距离是包括把椅子拉出来，以及能使就餐的人方便活动的最小距离。 3.一张以对角线对墙的正方形桌子所占的面积要有多大？ 180*180平方厘米 这是一张边长90厘米，桌角离墙面最近距离为40厘米的正方形桌子所占的最小面积。 4.桌子的标准高度应是多少？ 72厘米。 这是桌子的中等高度，而椅子是通常高度为45厘米。 5.一张供六个人使用的桌子摆起居室里要占多少面积？ 300*300厘米。 需要为直径120厘米的桌子留出空地，同时还要为在桌子四周就餐的人留出活动空间。这个方案适合于那种大客厅，面积至少达到600*350厘米。 6.吊灯和桌面之间最合适的距离应该是多少？ 70厘米。 这是能使桌面得到完整的、均匀照射的理想距离。 在卫生间 1．卫生间里的用具要占多大地方？ 马桶所占的一般面积：37厘米×60厘米 悬挂式或圆柱式盥洗池可能占用的面积：70厘米×60厘米 正方形淋浴间的面积：80厘米×80厘米 浴缸的标准面积：160厘米×70厘米 2．浴缸与对面的墙之间的距离要有多远？ 100厘米。想要在周围活动的话这是个合理的距离。即使浴室很窄，也要在安装浴缸时留出走动的空间。总之浴缸和其他墙面或物品之间至少要有60厘米的距离。 3．安装一个盥洗池，并能方便地使用，需要的空间是多大？ 90厘米×105厘米。这个尺寸适用于中等大小的盥洗池，并能容下另一个人在旁边洗漱。4．两个洗手洁具之间应该预留多少距离？ 20厘米。这个距离包括马桶和盥洗池之间，或者洁具和墙壁之间的距离。 5．相对摆放的澡盆和马桶之间应该保持多远距离？ 60厘米。这是能从中间通过的最小距离，所以一个能相向摆放的澡盆和马桶的洗手间应该至少有180厘米宽。 6．要想在里侧墙边安装下一个浴缸的话，洗手间至少应该有多宽？ 180厘米。这个距离对于传统浴缸来说是非常合适的。如果浴室比较窄的话，就要考虑安装小型的带座位的浴缸了。
家装用尺寸一览表
家装用尺寸一览表 ■标准入户门洞0.9m*2m, ■房间门洞0.9m*2m, ■厨房门洞0.8m*2m ■卫生间门洞0.7m*2m ■客厅：长沙发：240*90*75cm?长方形茶几：130*70*45cm 电视柜：200*50*180cm?? 电视离沙发：3m?电视高度与电视柜高差：40到120cm徒道宽度：100至120cm ■厨房：橱柜操作台：台面高80cm左右？面积90*46 (最小20最大60) cm?吊柜： 离台面60cm左右？?高度在145cm到150cm餐桌：餐桌高：750—790mm餐椅高；450 —500mm圆桌直径：二人500mm二人800mm四人900mm五人1100mm六人1100-1250mm 八人1300mm十人1500mm 十二人1800mm方餐桌尺寸：二人700 x 850(mm),四人1350x 850(mm),八人2250X 850(mm) ■卫生间：浴缸长度：一般有三种1220、1520、1680mm宽：720mm高：450mm 坐便：750x 350(mm> 冲洗器：690x 350(mm> 盟洗盆：550x 410(mm)o 淋浴器高： 2100mm 化妆台：长：1350mm 宽450mm ? ■卧室：标准双人床尺寸：150*190、150*200厘米，被套的尺寸应配180*215和 200*230之间的。加大双人床尺寸：180*200厘米，被套一般为200*230或220*240。 床头柜宽：400毫米-600毫米，深：350毫米-450毫米高：500毫米-700毫米。衣柜：柜门尺寸，单扇一门宽度不超过1200mm高度不超过2400mm挂衣区尺寸，上衣区高度在100cm-120cm不低于90cm宽度在40cm长衣区是140cm-150cm指间，
家装各种最佳尺寸标准大全
家装各种最佳尺寸标准大全
在客厅 1.长沙发与摆在它面前的茶几之间的正确距离是多少？30厘米 2.一个能摆放电视机的大型组合柜的最小尺寸应该是多少？200*50*180高厘米 3.如果摆放可容纳三、四个人的沙发，那么应该选择多大的茶几来搭配呢？140*70*45高厘米 4.在扶手沙发和电视机之间应该预留多大的距离？3米 5.摆在沙发边上茶几的理想尺寸是多少？方形：70*70*60高厘米。椭圆形：70*60高厘米。 6.两个面对面放着的沙发和摆放在中间的茶几一共需要占据多大的空间？ 两个双人沙发（规格 160*90*80高厘米）和茶几（规格100*60*45高厘米）之间应相距30厘米。 7.长沙发或是扶手沙发的的靠背应该有多高？85至90厘米。 8.如果客厅位于房间的中央，后面想要留出一个走道空间，这个走道应该有多宽？100至120厘米。 9.两个对角摆放的长沙发，它们之间的最小距离应该是多少？10厘米。 在餐厅 1.一个供六个人使用的餐桌有多大？ 120厘米。这是对圆形餐桌的直径要求。 140*70厘米。这是对长方形和椭圆形捉制的尺寸要求。 2.餐桌离墙应该有多远？80厘米。 3.一张以对角线对墙的正方形桌子所占的面积要有多大？180*180平方厘米 4.桌子的标准高度应是多少？72厘米。 5.一张供六个人使用的桌子摆起居室里要占多少面积？300*300厘米。 6.吊灯和桌面之间最合适的距离应该是多少？70厘米。 在卧室 1、双人主卧室的最标准面积是多少？12平方米 2、如果把床斜放在角落里，要留出多大空间？360*360厘米 3、两张并排摆放的床之间的距离应该有多远？90厘米 4、如果衣柜被放在了与床相对的墙边，那么两件家具这间的距离应该是多少？90厘米 5、衣柜应该有多高？240厘米 6、要想容的下双人床、两个床头柜外加衣柜的侧面的话，一面墙应该有多大？420*420厘米 在厨房 1.吊柜和操作台之间的距离应该是多少？60厘米。 2.在厨房两面相对的墙边都摆放各种家具和电器的情况下，中间应该留多大的距离才不会影响在厨房里做家务？120厘米。 3.要想舒服地坐在早餐桌的周围，凳子的合适高度应该是多少？80厘米。 4.吊柜应该装在多高的地方？145至150厘米。
家装用尺寸一览表
家装用尺寸一览表 ▌标准入户门洞0.9m*2m， ▌房间门洞0.9m*2m， ▌厨房门洞0.8m*2m， ▌卫生间门洞0.7m*2m ▌客厅：长沙发：240*90*75cm 长方形茶几：130*70*45cm电视柜：200*50*180cm 电视离沙发：3m 电视高度与电视柜高差：40到120cm 走道宽度：100至120cm ▌厨房：橱柜操作台：台面高80cm左右面积90*46（最小20最大60）cm 吊柜：离台面60cm左右高度在145cm到150cm 餐桌：餐桌高：750—790mm。餐椅高；450—500mm。圆桌直径：二人500mm．二人800mm，四人900mm，五人1100mm，六人1100-1250mm，八人1300mm，十人l500mm，十二人1800mm。方餐桌尺寸：二人700×850(mm)，四人1350×850(mm)，八人2250×850(mm) ▌卫生间：浴缸长度：一般有三种1220、1520、1680mm；宽：720mm，高：450mm。坐便：750×350(mm)。冲洗器：690×350(mm)。盟洗盆：550×410(mm)。淋浴器高：2100mm。化妆台：长：1350mm；宽450 mm。
▌卧室：标准双人床尺寸：150*190、150*200厘米，被套的尺寸应配180*215和200*230之间的。加大双人床尺寸：180*200厘米，被套一般为200*230或220*240。床头柜宽：400毫米-600毫米，深：350毫米-450毫米高：500毫米-700毫米。衣柜：柜门尺寸，单扇一门宽度不超过1200mm，高度不超过2400mm。挂衣区尺寸，上衣区高度在100cm-120cm，不低于90cm，宽度在40cm；长衣区是 140cm-150cm指间，不低于130cm，宽度在40cm。裤架尺寸。柜子的深度一般在600-650mm之间，那么裤架的深度范围在490-540mm，宽度不限。 ▌灯具：大吊灯最小高度：2400mm。壁灯高：1500—1800mm。反光灯槽最小直径：等于或大于灯管直径两倍。壁式床头灯高：1200—1400mm。照明开关高：1000mm。 ▌插座、开关： 客厅： 1）除特殊要求以外一般低插300mm、增加插座要与原插座持平。总电箱1850mm 2）背景墙插座。在电视柜下面的200一250mm.在电视柜上面的450一500mm.在挂电视中的1100mm. 卧室