32-A Wireless MAC Protocol Using Implicit Pipelining

7/27/2019 32-A Wireless MAC Protocol Using Implicit Pipelining

http://slidepdf.com/reader/full/32-a-wireless-mac-protocol-using-implicit-pipelining 1/16

A Wireless MAC ProtocolUsing Implicit Pipelining

Xue Yang, Member , IEEE , and Nitin H. Vaidya, Senior Member , IEEE

Abstract—In distributed multiple access control protocols, two categories of overhead are usually associated with contention

resolution. One is channel idle overhead, where all contending stations are waiting to transmit. Another is collision overhead, which

occurs when multiple contending stations attempt to transmit simultaneously. Either idle overhead or collision overhead being large,

contention resolution algorithm would be inefficient. Prior research work tries to minimize both the idle and the collision overheads

using various methods. In this paper, we propose to apply “pipelining” techniques to the design of multiple access control protocol so

that channel idle overhead could be (partially) hidden and the collision overhead could be reduced. While the concept of pipelined

scheduling can be applied to various MAC protocol designs in general, in this paper, we focus on its application to IEEE 802.11 DCF. In

particular, an implicitly pipelined dual-stage contention resolution MAC protocol (named DSCR) is proposed. With IEEE 802.11, the

efficiency of contention resolution degrades dramatically with the increasing load due to high probability of collision. Using the implicit

pipelining technique, DSCR hides the majority of channel idle time and reduces the collision probability, hence, improves channel

utilization, average access delay, and access energy cost over 802.11 significantly both in wireless LANs and in multihop networks.

The simulation results, as well as some analysis, are presented to demonstrate the effectiveness of DSCR.

Index Terms—Multiple access control (MAC), IEEE 802.11, pipelining, wireless LANs, multihop networks, channel utilization, accessenergy cost, packet access delay.

æ

1 INTRODUCTION

IEEE 802.11 standard [1] defines a distributed coordinationfunction (DCF), which uses a binary exponential backoff

(BEB) algorithm to resolve channel contention. BEB algo-rithm controls the channel contention by adjusting the valueof CW (CW represents contention window and is aparameter of 802.11). When a station successfully transmitsa packet, it resets its CW to a minimum value CW min; when

a collision occurs, the colliding stations exponentiallyincrease their CW by a factor of 2, until CW reaches themaximum value CW max. A station wanting to access thechannel generates a random backoff counter uniformlydistributed over the interval [0, CW]. This backoff countercorresponds to the number of idle slots this station has towait before its transmission attempt.

Clearly, the choice of CW is critical to the performance of

802.11. When there are few contending stations in the

network, a smaller CW will reduce the channel idle time and

enable a better usage of channel bandwidth. When the

number of contending stations increases, a larger CW is

preferred to reduce the collision probability. Some prior

research analyzes the performance of IEEE 802.11 DCF [2],[3], [4], [5], [6], [7], [8], [9], and shows that IEEE 802.11 DCF

operates far from the optimal points of CW, which can be

confirmed from the simulation results shown in Fig. 1 and

other papers as well [2], [5], [10]. The simulation results in

Fig. 1 are obtained using ns-2 network simulator for wirelessLANs, in which all stations can hear each other’s transmis-sion. The payload size is 512 bytes and Constant Bit Ratetraffic is used. We can see that the peak throughput of 802.11 DCF is achieved when there are four contendingstations, and the throughput degrades with fewer or greaternumber of contending stations. With less than four con-

tending stations, the unnecessary channel idle time is theprimary reason for the throughput degradation. On theother hand, when increasing the number of contendingstations from 4 to 256, the aggregate throughput of 802.11 DCF degrades from 91 percent of ideal throughput(defined as the maximum throughput a network can obtainwithout any MAC scheduling idle and collision overhead) to68 percent, which is mainly due to the increased packetcollisions.

Two categories of overhead are associated with conten-tion resolution at MAC layer. One is channel idle overhead,where all stations are waiting to transmit. The other iscollision overhead, which occurs when multiple contending

stations attempt to transmit simultaneously. In 802.11 DCF,stations perform various operations sequentially. Contend-ing stations go through contention resolution procedure todetermine which station has the right of accessing thechannel; then, the winning station transmits its packet. Onlywhen current transmission finishes, a new round of contention resolution begins—for ease of exposition, as-sume a LAN scenario here. In the sequential procedure, both channel idle time and collisions consume the entirechannel bandwidth, resulting in performance tradeoffs. If alarger CW is applied, 802.11 DCF can achieve betterthroughput in heavily contended networks due to thereduced collision probability. However, at the same time,

the channel idle overhead in networks with little contention

258 IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 5, NO. 3, MARCH 2006

. The authors are with the Department of Electrical and ComputerEngineering and the Coordinated Science Laboratory, University of Illinoisat Urbana-Champaign, 1308 West Main Street, Urbana, IL 61801.E-mail: {xueyang, nhv}@uiuc.edu.

Manuscript received 24 Sept. 2003; revised 12 Aug. 2004; accepted 8 Oct.2004; published online 16 Jan. 2006.For information on obtaining reprints of this article, please send e-mail to:

[email protected], and reference IEEECS Log Number TMC-0155-0903.1536-1233/06/$20.00 ß 2006 IEEE Published by the IEEE CS, CASS, ComSoc, IES, & SPS



increases using a larger CW, which leads to the degradedthroughput in such networks.

Traditionally, efforts have been made to dynamically

adjust stations’ channel access behavior based on thenetwork contention status so that both the channel idleand collision overhead can be minimized [10], [11], [12],[13], [14]. However, such algorithms typically requireextensive channel feedback information, which may not be available in wireless networks, to infer the networkcontention status.

We consider an alternative method, which applies“pipelining” techniques to MAC protocol design. Pipelininghas been used successfully in other areas (e.g., computerarchitecture [15]) to improve performance. The key me-chanism used for pipelining is to divide the total task intosubtasks, and to introduce parallelism by allowing different

subtasks of different tasks proceed simultaneously. ForMAC protocols, the total task is to schedule the channelaccess and transmit the packet. By using “pipelining,” thecontention resolution procedure that schedules the channelaccess (partially) overlaps with the packet transmissionduration. That is, when a pair of source and destinationstations are using the channel to exchange packets, theremaining contending stations start the contention resolu-tion procedure in parallel in order to resolve the channelcontention for next packet. Since it is performed in parallelwith packet transmission, the pipelined contention resolu-tion procedure consumes little channel bandwidth, whichcan help to improve performance.

Our previous research has explored explicit pipelining,wherein a control channel is used for explicitly performingpipelined contention resolution [16], [17]. This paperconsiders an implicit pipelining scheme that attempts toretain the benefits of explicit pipelining, but without using aseparate control channel. The performance evaluationresults presented later in this paper indicate that the implicitpipelining scheme is able to achieve good performance.

The rest of the paper is organized as follows: A brief introduction to IEEE 802.11 DCF is presented in Section 2; theproposed scheme bears significantsimilarities to IEEE 802.11.In Section 3, one explicit pipelining scheme is discussed,whichmotivates the proposed “implicit pipelining” mechan-

ism. Section 4 describes the proposed implicit pipelining

MACprotocol, named DSCR, in detail. Section 5 presents thesimulation results of DSCR for wireless LANs. The improvedcontention resolution using DSCR in multihop networks isdiscussed in Section 6. Related work is discussed in Section 7.Conclusions are drawn in Section 8. Finally, theoreticalanalysis regarding the reduced channel contention usingDSCR is given in the Appendix which is provided for free at

http://computer.org/tmc/archives.htm.

2 OVERVIEW OF IEEE 802.11 DCF

We briefly describe the relevant features of IEEE 802.11 DCFhere. For more details, please refer to [1]. IEEE 802.11 DCFdefines two access methods: basic access method and RTS/ CTS access method. The basic access method involves onlyData/ACK exchange, in which data packets are transmittedwhen channel access has been obtained. ACK frames followsuccessful data packet receptions. In the RTS/CTS accessmethod, RTS (Request To Send) and CTS (Clear To Send)frames are exchanged before Data/ACK packets. RTS and

CTS frames contain a duration field that defines the periodof time for which the medium is to be reserved to transmitthe actual Data frame and the returning ACK frame.Stations which overhear RTS/CTS frames defer transmis-sion for this period. This mechanism is referred to as“virtual carrier sensing” and it is implemented using“Network Allocation Vector” (NAV). The duration field isalso available in the MAC header of Data and ACK frames.A station updates the NAV with the duration field specifiedin the overheard frames.

The carrier sense mechanism in IEEE 802.11 includesphysical carrier sense and virtual carrier sense. Afterchannel is sensed idle for a DIFS (DCF Interframe Space)

duration, the backoff procedure is invoked by a backloggedstation and its backoff counter is decremented by 1 aftereach idle slot. As we explained in Section 1, 802.11 DCF usesbinary exponential backoff (BEB) algorithm to resolve channelcontention. A shorter interframe space, SIFS, is used toseparate transmissions belonging to a single dialog (e.g.,CTS, Data, and ACK frames in the case of RTS/CTS accessmethod). Fig. 2 illustrates the RTS/CTS access method of IEEE 802.11 DCF.

In terms of channel feedback, a station using 802.11receives feedback from its own packet transmission (successor failure) to adjust its contention window size. In addition,the physical and virtual carrier sensing help to avoid

potential collisions.

3 PIPELINED PACKET SCHEDULING

To perform pipelined packet scheduling, the pipelinedcontention resolution procedure can either be used tocompletely resolve the channel contention or only partiallyresolve the contention [16], [17], [18]. When it is used toresolve the channel contention partially, the purpose of thepipelined contention resolution procedure is to hide a partof channel idle overhead and to reduce collision overhead.In our prior work, we have developed a “partial pipelining”scheme [16], [17], which is briefly discussed below to

motivate the proposed implicit pipelining scheme. The

YANG AND H. VAIDYA: A WIRELESS MAC PROTOCOL USING IMPLICIT PIPELINING 259

Fig. 1. Throughput of 802.11 versus the number of contending stations

(throughput is normalized to the ideal value with perfect scheduling).



partial pipelining scheme uses a narrow-band busy tonechannel for the purpose of pipelining, as explained next.

In partial pipelining, the contention resolution procedureis split into two phases, as shown in Fig. 3. Pipelined stage 1includes only contention resolution phase 1 and is

performed on the busy tone channel. Contention resolutionphase 2 and packet transmissions are performed on the datachannel in pipelined stage 2.

A backoff counter is associated with contention resolu-tion phase 1. When some other station is transmitting on thedata channel, a contending station reduces its phase 1 backoff counter by 1 after each idle slot on the busy tonechannel. Whenever the phase 1 backoff counter reacheszero, the contending station sends out a signal on the busytone channel to claim its winning status and becomes a“pipelined station.” In the rest of this paper, we use theterm “pipelined station” to refer to a station that has woncontention resolution phase 1 and is allowed to entercontention resolution phase 2, while the term “contending

station” is used for any backlogged station that wants toaccess the channel. Only pipelined stations are allowed tocontend for the data channel access when current transmis-sion finishes. Other contending stations freeze themselvesin contention resolution phase 1 upon sensing a busy tonesignal. The data channel contention among pipelinedstations is further resolved in contention resolution phase 2following the procedure similar to 802.11 DCF (but withsmaller contention window sizes). It can be shown that,with a large probability, the number of pipelined stations issmall with perfect busy tone detection. Thus, the collisionprobability among the pipelined stations is small, whichleads to the reduced collision overhead. Also, since a

significant portion of random backoff is performed on the

narrow-band busy tone channel in contention resolutionphase 1 when data channel is busy, the channel idleoverhead associated with the backoff procedure is effec-tively reduced.

Partial pipelining can achieve a significant improvement

over IEEE 802.11 in terms of channel utilization across awide range of network sizes [16], [17]. However, partialpipelining relies on a busy tone channel, which incurs morehardware cost. Additionally, in wireless networks, the busytone signal may not be sensed reliably. In fact, with hiddenterminals, the busy tone transmitted by a station may not besensed by another station at all. With unreliable busy tonedetection, the performance of partial pipelining may beexpected to degrade—to assess the performance impact of imperfect busy tone detection, we evaluated the partialpipelining scheme with perfect busy tone detection (i.e.,100 percent detection probability) and the worst-casescenario of no busy tone detection (0 percent detection

probability). The performance results for wireless LANswith up to 256 contending stations are plotted in Fig. 4. Asexpected, the performance of partial pipelining degradeswith 0 percent detection probability for busy tones.However, surprisingly, the performance remains superiorto IEEE 802.11. A closer look at partial pipelining revealsreasons for this phenomenon. On the one hand, lack of busytone detection implies that there is no way for a pipelinedstation to signal other stations to freeze themselves incontention resolution phase 1; this increases number of contending stations in stage 2, resulting in poorer perfor-mance compared with perfect busy tone detection. On theother hand, contention resolution phase 1 allows only a

subset of stations to enter stage 2 at any given time,


Fig. 2. RTS/CTS access method of IEEE 802.11 DCF.

Fig. 3. Partial pipelining scheme.



resulting in better performance than IEEE 802.11. Theseobservations motivated us to consider development of aMAC protocol that mimics partial pipelining, but withoutusing a separate control channel. Effectively, our goal is toachieve performance of perfect busy tone detection withoutusing a busy tone at all (note that 0 percent detectionprobability is equivalent to not transmitting a busy tone).

Again, consider the partial pipelining scheme with0 percent busy tone detection probability. In this case, whena station countsdownits phase 1 counter to0, and transmits a busy tone, other stations do not sense the busy tone, and cancontinue counting down their backoff counters. Effectively,stations cancount down their phase 1 backoff counters fortheduration of theon-goingdata packettransmission on thedatachannel—if the data packet transmission lasts for L slots,

then stations will be able to count down their counters byL slots. Thus, partial pipelining with 0 percent busy tonedetection can be implemented by simply allowing eachstation to decrement its phase 1 backoff counter by L slotswhenever a data packet transmission is detected as having been completed (instead of decrementing the counter by 1after each slot time). Next, instead of usingL, the data packetduration, one may decrement the backoff counter by anysuitable amount. The implicit pipelining illustrated in Fig. 5is derived from the above observations and incorporatesother adaptations of backoff procedures to improve perfor-mance. The proposed MAC protocol DSCR (pipelined Dual-Stage Contention Resolution) is described in the next section

in detail.

4 PIPELINED DUAL STAGE CONTENTION

RESOLUTION MAC PROTOCOL (DSCR)

DSCR uses backoff mechanism analogous to 802.11 DCF.Each contention resolution phase maintains its owncontention window and backoff counter. As in 802.11 DCF,a station using DSCR receives feedback from its own packet

transmission (success or failure) to adjust its contentionwindow size. Available physical and virtual carrier sense(using NAV mechanism) also help to avoid some potentialcollisions. DSCR differs from 802.11 DCF only in thecontention resolution algorithm, while the other functionsremain the same.

For heavily contended networks, DSCR statisticallycontrols the number of pipelined stations to be relativelysmall so that channel contention in stage 2 can be resolvedefficiently. At the same time, DSCR is carefully designed toavoid unnecessary waste of channel bandwidth for net-works with little contention, which will get clearer in viewof protocol details.

4.1 Contention Resolution Phase 1 (Stage 1)

CW1 is the contention window for contention resolutionphase 1 and it has a minimum valueCW 1min anda maximumvalue CW 1max. The initial value of CW1 is CW 1min.

A backoff counter, named bc1, is associated withcontention resolution phase 1. bc1 is chosen to be uniformlydistributed over the interval [0, CW1]. Whenever a station’sbc1 becomes less than or equal to 0, this station becomes apipelined station and enters stage 2.

Updating of bc1: The backoff counter bc1 is reduced inthe following two ways:

1. At the end of an overheard successful packet

transmission, a contending station reduces its bc1 by a quantity named F . By doing this, the pipelinedcontention resolution phase 1 is implicitly per-formed. With a large value of F , many stationscould have their bc1 0 after reducing it by F ;hence, many stations could enter stage 2 to contendfor the channel. On the other hand, with a smaller F ,there would be fewer stations entering stage 2. InDSCR, we choose F to be a function of tc, where tc

represents the number of successfully transmittedpackets overheard by the contending station eversince the most recent time it enters stage 1. tc will bereset to 1 when a station enters stage 1 for a newly

arriving packet or when returns from stage 2 to


Fig. 4. Performance of partial pipelining with and without busy tone

detection (packet size: 512 bytes).

Fig. 5. Implicitly pipelined packet scheduling.



stage 1, and tc will be increased by 1 after eachsubsequent overheard successful transmission.While there are various choices possible for F , wecurrently define F as F ¼ 2

tc À 1. Thus, the longer astation has stayed in the stage 1, the more aggres-sively it reduces its bc1, hence, the larger probabilityof becoming a pipelined station. Once bc1 0, the

station enters stage 2 to contend for the channel.2. It is possible that, when current transmission of a

station finishes, none of the surrounding contendingstations have their bc1 less than or equal to zero.When such cases occur, no pipelined stationscontend for the channel in stage 2, channel band-width will be unnecessarily wasted. To avoid this, inDSCR, bc1 is also linearly decreased when channel isidle, i.e., bc1 is reduced by 1 after each idle slot.Whenever a station’s bc1 reaches zero, it becomes apipelined station and enters stage 2.

Updating of CW1: Among pipelined stations that

contend for the channel in stage 2, a station that eventually

wins channel access transmits its packet, then resets CW1 toCW 1min, resets tc to 1, regenerates bc1 from the interval [0,

CW1], and returns to stage 1. On the other hand, a pipelined

station that loses channel contention in stage 2 will double

its CW1,1 regenerate bc1, reset tc to 1 and return to stage 1.We will illustrate stage 1 later using some examples.

4.2 Contention Resolution Phase 2

CW2 is contention window for contention resolution phase

2. It has a minimum value CW 2min and a maximum value

CW 2max.Initial value of CW2: The initial value of CW2 is CW 2min

for all stations entering stage 2.

A backoff counter, named bc2, is associated with thecontention resolution phase 2. Whenever bc2 reaches zero, a

transmission is allowed.Initial value of bc2: Depending on how a pipelined

station enters stage 2, there are two different choices for the

initial value of bc2.

1. When a packet transmission finishes, multiplepipelined stations in the neighborhood could occurand enter stage 2 at the same time since they all mayhave bc1 0 after reducing bc1 by F . To furtherresolve the channel contention among these multiplepipelined stations, each of them generates an initialvalue for bc2, which is uniformly distributed over theinterval [0, CW2].

2. As we mentioned in Section 4.1, a contending stationwill also reduce bc1 by 1 after each slot when thechannel is sensed idle. When a station counts downits bc1 to zero in such a way, it will become apipelined station and enter stage 2 with the initialvalue of bc2 being 0. This is due to the followingconsiderations: The objective of contention resolutionphase 2 is to further resolve the channel contentionamong multiple pipelined stations occurred at thesame time. Since the probability that more than

one station count down bc1 to 0 simultaneouslyduring channel idle time is small, a further phase 2 backoff may lead to unnecessary channel waste.

Transmissions and Retransmissions: Despite differentinitial values of bc2 thepipelinedstations in stage 2 may have,they all follow the same rules for transmissions andretransmissions. A pipelined station has to wait for the

channel to be idle for DIFS duration before the backoff procedure of contention resolution phase 2 begins (as in802.11). Then, if a pipelined station’s bc2 reaches zero and thechannel is idle, the station will begin its transmission;otherwise, bc2 will be decreased by 1 after each idle slot.Beforebc2 ofapipelinedstationreacheszero,ifaframesentbysome otherstation is overheard(e.g.,RTS or CTSframes in thecase of the RTS/CTS access method), the former pipelinedstation loses channel contention and returns to stage 1.

When a collision2 happens, the colliding stations willdouble their CW2, and generate a new bc2 value from theinterval [0, CW2]. The colliding stations as well as otherpipelined stations all stay in stage 2 and repeat the above

contention resolution phase 2 until someone eventuallywins the channel.

The pipelined station that wins the channel will transmitthe packet, reset CW1 and CW2 to CW 1min and CW 2min,respectively, reset tc to 1, regenerate bc1 from the interval[0, CW1], and return back to the first stage. A pipelinedstation that loses the channel will double its CW1, resetCW2 to CW 2min, reset tc to 1, regenerate bc1, and return back to the first stage.

4.3 Examples for DSCR

Simulation results show that, for DSCR, CW 1min ¼ 15,CW 2min ¼ 31 are appropriate choices for networks up to

256 contending stations. The choices of CW 1max andCW 2max have no major impact on the performance of

DSCR, provided that they are large enough to accommodatethe maximum network size. We set both CW 1max andCW 2max to 1,023.

Fig. 6 illustrates the dual stage contention resolution of DSCR. For ease of exposition, in the example, we pretendthat all five stations observe the end of a packet transmis-sion at time t0, and they all contend for the channel witheach other. Notice it is possible that, in multihop networks,different stations observe different packet transmissionsdepending on their locations. We also pretend that allstations have tc ¼ 4 (i.e., four successfully transmittedpackets have been overheard by the contending stationsince the most recent time it enters stage 1) at time t0, hence,each station reduces bc1 by 2

tc À 1 ¼ 15. Since stations 3 and5 now have bc1 0, they enter stage 2 (becoming pipelinedstations), generate bc2 from the interval [0, CW2] (the initialvalue of CW2 is CW 2min ¼ 31), and begin to count downbc2.3 Stations 1, 2, and 4 stay in stage 1, reducing their bc1 by1 after each idle slot. After seven slots, at time t2, station 5counts down bc2 to zero and wins the channel access.Station 3 doubles its CW1 on losing stage 2, resets tc to 1and returns back to stage 1. It also generates a new bc1 value


1. When we say “double” CW1, we mean that CW1 = 2*CW1 + 1. For

instance, if CW1 is originally 7, on doubling, it becomes 15.

2. When an RTS is not followed by a CTS, or Data is not followed by anACK, a collision is assumed to have occurred.

3. To simplify the explanation, we do not mention DIFS in this example.



which is uniformly distributed over the interval [0, CW1].

At time t3 when station 5 finishes its transmission, station 5

resets CW1 to CW 1minðCW 1min ¼ 15Þ, resets tc to 1, returns

back to stage 1, and generates a new value for bc1 from the

interval [0, CW1]. Upon station 5 finishing its transmission

at time t3, all stations again reduce their bc1 by 2tcÀ1. Notice

this time, stations 1, 2, and 4 have tc ¼ 5, hence, bc1 of these

three stations is reduced by 31. On the other hand, stations 3

and 5 have tc ¼ 1 and their bc1 are reduced by 1. Now that

all stations have their bc1 larger than 0, none of them enterstage 2 at time t3. However, bc1 of each station is also

decreased by 1 after each idle slot, at time t4, bc1 of station 2

is counted down to zero after 5 idle slots. Station 2 then

becomes a pipelined station, enters stage 2, sets the initial

value of bc2 to 0 and transmits at time t4.A special case happens when there is only a single flow in

the network. In the example of Fig. 7, at time t0, station 0 is

the only backlogged source station in the network and its bc1is 8. Assuming tc ¼ 1 at time t0, station 0 then has bc1 ¼ 7

after reducing bc1 by 2tc À 1 ¼ 1, and it continues to stay in

stage 1 because bc1 > 0. During channel idle time, bc1 is

decreased by 1 after each slot. After seven idle slots, at

time t1, bc1 is counted down to zero. The initial value of bc2 is

set to 0 and it begins to transmit its packet.At time t2 when station 0 finishes its transmission, it

resets CW1 to CW 1minðCW 1min ¼ 15Þ, resets tc to 1, and

generates a new value, say 9, for bc1. Again, at time t2,

station 0 reduces bc1 by 2tc À 1 ¼ 1. Then, after eight idle

slots, the bc1 of station 0 is counted down to zero at time t3,

and station 0 transmits again. As we can see, when there is

only a single flow in the network, DSCR performs very

similar to IEEE 802.11 DCF if CW 1

min of DSCR has the samevalue as CW min of 802.11.

4.4 Summary of DSCR

DSCR uses two stages to resolve the channel contention,

which is not a new idea. For instance, European Tele-communications Standards Institute (ETSI) High Perfor-

mance European Radio LAN (HIPERLAN/1) MAC

protocol [19] uses two stages, namely, the elimination

stage and the yield stage, to resolve channel contention. In

the elimination stage of HIPERLAN/1, a contending

station transmits elimination burst for a random durationand then listens to the channel in the elimination survival

verification interval. A contending station survives the


Fig. 6. An example for the dual stage contention resolution of DSCR (time axis is not drawn to scale).

Fig. 7. A single flow.



elimination stage if and only if the channel is sensed idle inits elimination survival verification interval; otherwise, thisstation is eliminated and withdraws from current channelcompetition. The yield stage then further resolves conten-tion and reduces the number of stations allowed totransmit to 1.

The fundamental difference between DSCR and conven-

tional two-stage contention resolution algorithms is that, inDSCR, stage 1 proceeds in parallel with stage 2 usingpipelining techniques, as illustrated in Fig. 5. Whenobserving the channel activities, only stage 2, whichincludes the contention resolution phase 2 and packettransmissions, actually consumes the channel bandwidth.Stage 1 reduces both the channel idle and collision overheadwithout introducing any additional cost.

On the other hand, in conventional two-stage contentionresolution algorithms, both stages consume channel band-width. For example, in HIPERLAN/1, the elimination stagecannot effectively reduce the channel contention if it is tooshort. At the same time, since the elimination stage also

consumes channel time, the longer the elimination stage,the more wasted channel bandwidth in elimination stage.As a result, the length of elimination stage of HIPERLAN/1has to be carefully chosen to trade-off the correspondingchannel wastage with the effect of reducing channelcontention, which is very similar to the tradeoff of choosingCW value faced by 802.11 DCF, as mentioned at the beginning of this paper (Section 1).

Furthermore, without requiring any additional signalingmechanism (e.g., burst signals or busy tones), DSCRexploits the dynamic feedback between two contentionresolution stages to effectively control the number of pipelined stations in stage 2. As Fig. 8 illustrates, with atotal of N contending stations at the input, stage 1 generatesM pipelined stations. The value of M largely depends onthe distribution of CW1 among all N contending stations. If all contending stations have large values for CW1, then bc1among them tends to be widely distributed and only a smallfraction of them will have their bc1 reaching zero at anypoint of time. Thus, only a small number of stations will become pipelined stations. More stations may becomepipelined stations if all contending stations have smallervalues for CW1. At the same time, stage 2 selects 1 winnerfrom M pipelined stations to access the channel, and theremaining M À 1 pipelined stations double their CW1. Theoutput of stage 2 thus affects the distribution of CW1among N contending stations (i.e., input of stage 1), which,

in turn, affects the number of pipelined stations that may

occur next time. Thus, despite the large variation of the totalnumber of contending stations, the number of pipelinedstations entering stage 2 remains within a small range.

The interaction between the distribution of CW1 and thenumber of pipelined stations in stage 2 is captured by theanalysis presented in the Appendix, which can be found onthe Computer Society Digital Library at http://computer.

org/tmc/archives.htm. Both the analysis and our simula-tion results show that DSCR successfully controls thechannel contention in stage 2. With up to 256 contendingstations, the average number of pipelined stations contend-ing for the channel in stage 2 is less than 28.

5 PERFORMANCE EVALUATION OF DSCR IN

WIRELESS LANS

As wireless LAN environment gives us a simple context toreveal some key properties of DSCR, in this section, we usesome simulation results from wireless LANs to continueour discussions on DSCR protocol. Extensive simulationresults for multihop wireless networks will be presented in

Section 6 to show the performance improvement of DSCRover 802.11 in multihop networks.

All the simulation results in this paper are based on amodified version of ns-2 network simulator [20]. Theeffective transmission range is limited to 250 meters andcarrier sense range (the range in which carrier can besensed) is limited to 550 meters based on the settings of Lucent WaveLAN DSSS radio interface. Two-ray groundradio propagation model is assumed. The channel bit rate isset to 11 Mbps and RTS/CTS access method is used in thesimulations. Physical layer preamble and header aretransmitted at 1 Mbps, thus, have a total length of 192saccording to IEEE 802.11 standard (with Direct SequenceSpread Spectrum) [1].4 Packet payload size is 512 bytes. Weuse Constant Bit Rate traffic and traffic rate is aggressiveenough to keep a contending station always backlogged.The total number of contending stations (N) is increasedfrom 1 to 256. Each simulation lasts for 30 seconds and thepresented results are averaged over 20 runs.

We measure the results of DSCR and 802.11 in terms of aggregate throughput, average access delay and accessenergy cost. In the simulated scenario, taking into accountthe overhead introduced by data packet header (48 bytes),RTS (20 bytes), CTS (14 bytes), ACK (14 bytes), DIFS (50s),SIFS (10s), physical layer preamble and header (192s),respectively, for each of RTS, CTS, DATA and ACK, thetotal transmission time is 1; 290:18s for each payload

packet (512 bytes). Therefore, without any cost of channelcontention resolution, we can expect the ideal throughput to be 3,100.3 Kbps. The aggregate throughput of DSCR and802.11 shown below is normalized to the ideal throughput.

The average access delay is measured as the delayexperienced by a packet from the time it arrives at MAClayer to the time the source station receives acknowl-edgment from the destination. To measure the energyconsumption, we assume the power consumption model of 2.4 GHz DSSS Lucent IEEE 802.11 WaveLAN PC Cardoperating in ad-hoc mode with channel bit rate of 11 Mbps


Fig. 8. Dynamic feedback of DSCR.

4. In this paper, we do not consider the case that different portions of apacket may have different transmission ranges due to their different

transmission rates.



[21], as shown in Table 1. As both successful andunsuccessful transmission attempts consume energy, aMAC protocol with more retransmissions will be lessenergy-efficient. We define the “access energy cost” as thetotal consumed energy by all stations divided by theaggregate throughput (unit: Joule/Kbps).

5.1 Throughput, Access Delay, andAccess Energy Cost

The aggregate throughput of DSCR and IEEE 802.11 is norm-alizedwith respect to theidealthroughput (3,100.3 Kbps) andis presented in Fig. 9a. The standard deviation of thenormalized throughput for DSCR and 802.11 is less than

0.0013 and 0.0015, respectively. As we can see, 802.11 has theclosest performance to DSCR when there arefour contendingstations, where 802.11 obtains 91 percent of ideal throughputand DSCR obtains 92 percent. With fewer or greater numberof contending stations, the performance gap between DSCRand 802.11 becomes larger. Particularly, with 32 contendingstations, DSCR achieves 93 percent of ideal throughputwhile802.11 gets 84 percent. When N is 256, DSCR still retains88 percent of the ideal throughput. Meanwhile, 802.11 hasdropped its normalized throughput to 68 percent.5

DSCR not only improves the channel utilization over802.11 DCF, but as shown in Figs. 9b and 9c, the averageaccess delay and access energy cost of DSCR is also better

than 802.11 DCF.For a distributed contention resolution algorithm, the

access delay of a packet consists of channel waiting time before its transmission attempts and the time spent inpacket retransmissions. In DSCR, many retransmissions areavoided due to reduced collision probability. On the otherhand, the channel waiting time does not increase due to thepipelining procedure of DSCR. These two features togetherexplain why DSCR has shorter average access delay whileimproving the channel utilization for large networks. WhenN is 256, the average access delay of 802.11 is 0.466 s, whileDSCR has average access delay of 0.376 seconds. Thestandard deviation of the measured access delay for DSCRand 802.11 is 0.001 and 0.019 (seconds), respectively.

The reduced collision probability of DSCR also con-tributes to its reduced access energy cost. As manyretransmissions are avoided, each successfully deliveredpacket costs fewer transmission attempts, thus consumesless energy. When N is 256, the access energy cost of 802.11 DCF is 3.25 Joule/Kbps while that of DSCR is2.48 Joule/Kbps. The standard deviation of the measuredaccess energy cost for DSCR and 802.11 is 0.003 and 0.077(Joule/Kbps), respectively.

5.2 Key Properties of DSCR

To explain why DSCR works better, we take a closer look atthe behavior of DSCR with regard to the contention degreein stage 2 and the number of collisions experienced.

The total number of contending stations is increasedfrom 1 to 256, and we plot the average number of pipelinedstations in stage 2, as shown in Fig. 10a. Given the range of

N from 1 to 256, the number of pipelined stations iscontrolled to be within the range from 1 to 28, whichconfirms that DSCR controls the channel contention instage 2 effectively. In the Appendix, which can be found onthe Computer Society Digital Library at http://computer.org/tmc/archives.htm, we give theoretical analysis todeduce the average number of pipelined stations in stage 2.

Since only pipelined stations in stage 2 contend for thechannel at any given time, we expect the collisionprobability using DSCR can be reduced compared with802.11. Using RTS/CTS access method with ideal channelconditions, RTS retransmissions are the direct results of collisions. We count the average number of RTS retransmis-

sions per second for DSCR and 802.11, and plot the resultsin Fig. 10b. As we can see, the number of RTS retransmis-sions of DSCR is much less than 802.11. Particularly, with256 contending stations, there is an average of 1461 RTSretransmissions per second using 802.11 and only 414 usingDSCR. The results reveal that through the implicitlypipelined contention resolution phase 1, DSCR reducescollision overhead dramatically.

6 PERFORMANCE EVALUATION OF DSCR IN

MULTIHOP WIRELESS NETWORKS

In wireless LANs, all stations can hear each other’s transmis-

sion. On the other hand, in multihop networks, two hiddenstations that cannotreceive from each other maystill competefor the channel. Lack of correct channel feedback due to“hidden terminal” problem causes difficulties in efficientcontentionresolution in multihop networks.Some priorwork[22], [23], [24] discuss the problems of 802.11 in multihopnetworks, mainly dealing with the interactions amongphysical layer, MAC layer, routing layer and transport layer.The proposed DSCR is not intended to address all theseproblems existing in multihop networks. Instead, the effortsof DSCR are focused on how to improve the channelutilization via more efficient contention resolution, andvarious multihop scenarios are evaluated in this section.

Recall that the effective transmission range in thesimulations is limited to 250 meters and carrier sense rangeis limited to 550 meters.

6.1 Saturated Multihop Random Networks

In this section, 30 different topologies are generated byplacing 80 stations randomly in a 1; 000mÂ 1; 000m area. Inthe simulations, for each generated topology, each stationpicks one of its one hop neighbors (if there is any) to sendpackets to. The total number of flows varies from 70 to 75depending on the corresponding topology.6 Fig. 13 shows


5. In order to provide meaningful measurements for access delay andaccess energy cost, in all simulations of this paper, there is no retransmis-sion limit for each packet. On the other hand, the results in [16], which showa larger improvement of DSCR over 802.11 in terms of aggregate

throughput, are obtained with short retransmission limit of 7 (as in 802.11).

6. In Section 5, we see that the DSCR achieves greater performanceimprovement in larger networks. To avoid exaggerating the performance

gain of DSCR, we choose a moderate network size here.

TABLE 1Power Consumption of Lucent IEEE 802.11 WaveLAN PC Card



one example of the generated flow patterns. Each sourcestation is always backlogged.

Defining “throughput ratio” as the aggregate throughputof DSCR, divided by the aggregate throughput of 802.11, weshow the results of “throughput ratio” for the 30 randomtopologies in Fig. 11a using payload packet size of 512 bytesand the RTS/CTS access method. DSCR achieves 10 percentto 49 percent more throughput compared with 802.11 in

these multihop networks (i.e., throughput ratio is in therange from 1.10 to 1.49). The standard deviation of “throughput ratio” is less than 0.047. The average numberof RTS retransmissions (per second) for DSCR and 802.11are plotted in Fig. 11b. We can see that the number of retransmissions experienced by DSCR is much smaller than802.11, which confirms that DSCR gains more throughput by effectively reducing the collision overhead.

Average access delay and energy cost are reported inFigs. 12a and 12b, respectively. DSCR achieves 11 percent to69 percent less average access delay for each deliveredpacket, and spends 9 percent to 46 percent less energy perKbps of obtained throughput. The standard deviation of

access delay for DSCR and 802.11 is less than 0.0014 and

0.0008 seconds, respectively. The standard deviation of access energy cost for DSCR and 802.11 is less than 0.0051and 0.0035 Joule/Kbps, respectively.

Comparing with the results for wireless LANs in Fig. 9,we notice that a larger performance improvement over802.11 can be expected from DSCR in multihop networks.The reason is that, in multihop networks, two sourcestations that can interfere with each other may not be able to

sense each other due to their distance or communicationobstacles. Hence, the time period during which a transmis-sion is vulnerable to the possible collisions lasts not only thepropagation delay, but also the packet transmission dura-tion (e.g., RTS transmission duration in the case of RTS/CTSaccess method). As a result, compared with 802.11, thereduced channel contention using DSCR has larger impacton reducing the collision probability, which leads to moreperformance improvement.

6.2 Multihop Networks with Various Load

Having shown the performance improvement of DSCR over802.11 for saturated networks, we further identify its

performance in networks that have less traffic load. The


Fig. 9. Throughput, access delay, and access energy cost. (a) Normalized throughput. (b) Average access delay. (c) Access energy cost.



purpose of this set of simulations focuses on comparing theperformance of DSCR with 802.11 when varying the trafficload from low to high. Poisson traffic with various packetarrival rate is used, and the traffic is generated at theapplication layer. The length of interface queue between thelink layer and the MAC layer is set to one to mitigate theimpact of packet buffering at the interface queue. Note thatthe arrival traffic at the MAC layer is not required to followany specific distribution since neither DSCR nor 802.11exploits any particular traffic property to aid the contentionresolution.

We perform simulations for one of the randomly

generated topologies, as shown in Fig. 13. There are a totalof 80 stations and 74 one hop flows. Each source stationgenerates its packets independently and the packet arrivalrate at each station is (unit: packets per second).

With ¼ 1 (i.e., one packet arrives per second for eachflow) and 512 byte packet size, the traffic demand is far below the network capacity. When gradually varying

from 1 to 20,000 packets/second, offered load is increasedfrom small to very large. The corresponding “aggregatethroughput,” “average access delay,” and “access energycost” are presented in Fig. 14. When the network load isvery low, channel contention is not a major concern for theperformance, DSCR behaves very similar to 802.11 DCF.

Their performance starts to diverge when the network is

loaded more heavily. In this particular example, the

aggregate throughput grows slowly beyond ¼ 100 pack-

ets/second using 802.11 while DSCR continues to deliver

more traffic until ¼ 1; 000 packets/second. The saturation

throughput of DSCR is 1.41 times of 802.11.It is interesting to observe that, when ¼ 100 packets/

second, both 802.11 DCF and DSCR have the largest access

delay. When offered load increases further, the access delay

drops. The main reason is because of the location dependent

contention in multihop networks. When ¼ 100, the

contending channels are not saturated, but the collision

probability among competing flows is high because of hidden terminals (Fig. 13 shows that many hidden terminals

exist). Many packets are delivered after multiple retransmis-

sions and, thus, the average access delay is large. When

traffic arrival rate increases further, it is likely that some of

the flows with less neighborhood contention will grab more

channel resource, while the other flows with more severe

contention will get fewer packets delivered (i.e., location

dependent contention of multihop networks). Since access

delay only accounts for the delay of successfully delivered

packets, the overall average access delay is decreased. At

saturation state, the average access delay of DSCR is

0.025 seconds, while 802.11 DCF has delay of 0.040 seconds.


Fig. 10. Properties of DSCR. (a) Average number of pipelined stations in

stage 2. (b) Average number of RTS retransmissions (per second).Fig. 11. Throughput and retransmissions in random topologies.

(a) Throughput ratio. (b) Number of retransmissions.



When network load is low, most of the energy isconsumed at idle state, which leads to the measured highenergy cost in terms of Joule/Kbps for both DSCR and802.11. At saturation state, the access energy cost of DSCR is0.23 Joule/Kbps, while that of 802.11 is 0.32 Joule/Kbps.

For above simulation results with ¼ 20; 000 packets/second, the standard deviation of “aggregate throughput”for DSCR and 802.11 is 80 and 35 Kbps, respectively; thestandard deviation of “average access delay” for DSCR and802.11 is 0.0008 and 0.0004 seconds, respectively; thestandard deviation of “access energy cost” for DSCR and802.11 is 0.0016 and 0.0014 Joule/Kbps, respectively.

6.3 Two Simple Scenarios

As seen from the results above, DSCR works well inmultihop networks despite the presence of hidden term-inals. From the description of DSCR, it should be clear thatDSCR benefits when a station can hear the successfultransmissions of other nearby stations. However, withhidden terminals, such detection may not always befeasible. We now consider two simple scenarios, shown inFig. 15, to explore the impact of hidden terminals further.

The location of each station is shown in the form of (x, y)coordinates.

The first scenario in Fig. 15a has six constantly back-logged source stations 1, 2, 3, 4, 5, and 7, where station 7 isout of the transmission range of stations 1-5, but within thetransmission range of station 6. Via CTS or ACK sent bystation 6, station 7 can learn of the transmissions fromstations 1-5. Using DSCR, when in stage 1, station 7 canreduce its bc1 whenever a transmission from stations 1-5finishes. Once becoming a pipelined station and contendingthe channel in stage 2, station 7 can tell whether it has lostthe channel or not via the overheard CTS from station 6. Asa result, from all contending stations 1, 2, 3, 4, 5, and 7,

DSCR only select a subset of them at any point of time tocontend for the channel in stage 2. The throughput obtained by each individual flow, the aggregate throughput and thenumber of RTS retransmissions occurred per second arepresented in Fig. 15a for both DSCR and 802.11 DCF. Noticethat, when station 7 collides with one of the stations from 1to 5, the packet reeception at station 6 will be interfered, butthe packet reception at station 8 can proceed successfully,which explains why flow 6 has much larger throughputthan other flows using 802.11 DCF. On the other hand,when DSCR is used, station 7 does not always get chancesto contend for the channel access in stage 2, which mitigatesits channel access advantage over others to some extent.

The second scenario in Fig. 15b is almost identical to thefirst scenario except that station 7 is further moved out of the transmission range of station 6. Now, station 7 cannotoverhear any successful transmissions from stations 1-6, butit is still within their carrier sense range. In DSCR, a stationhas two ways of reducing bc1. One way is throughoverheard successful transmissions, the other is to reducebc1 by 1 after each channel idle slot. Therefore, even thoughstation 7 cannot overhear any successful transmissions, itcan still enter stage 2 and contend for the channel when itsbc1 reaches zero via the second way. At the same time,DSCR only selects a subset from stations 1-5 to becomepipelined stations and enter stage 2 at any point of time.

When contending for the channel in stage 2 with other


Fig. 12. Access delay and access energy cost in random topologies. (a) Average access delay. (b) Access energy cost.

Fig. 13. One random multihop network.



pipelined stations, station 7 does not know whether it has

lost the channel access in stage 2 or not without overhearing

any successful transmissions. Consequently, station 7 will

not return to stage 1 until it has won the channel access and

transmitted a packet, which explains why flow 6 has larger

throughput than others in Fig. 15b using DSCR. On the

other hand, when using 802.11, station 7 contend for the

channel with all the stations 1-5. Whenever the transmis-sions from stations 1-5 are sensed, station 7 has to defer its

transmissions, which lead to a relatively small throughput

of flow 6 compared with other competing flows.The key observation from both examples is that, despite

the presence of hidden terminals in multihop networks,

DSCR is still able to select only a subset of contending

stations to become pipelined stations and contend for the

channel access in stage 2. Due to the reduced channel

contention, DSCR experiences less number of RTS retrans-

missions and has better throughput than 802.11, even

though the difference is not significant due to the limited

number of flows in the chosen scenarios. When the number

of contending flows is increased, the throughput benefit of DSCR over 802.11 becomes more significant.

7 RELATED WORK AND SOME COMPARISONS

WITH DSCR

Contention resolution for multiple access control has been

extensively studied for wired networks and infrastructure based wireless networks (i.e., networks in which mobilestations communicate via the base station). One category of the proposed algorithms (e.g., [11], [12]) makes use of ternary channel feedback (i.e., channel idle (0), collision (e),and successful transmission (1)) to estimate the networkload and stabilize the contention resolution algorithm. Someother tree splitting protocols (e.g., [13], [25], [26]) try toresolve collisions fast and improve the throughput. In treesplitting protocols, when a collision happens, the collidingstations are split into two subsets based on a certain rules,whereas the first subset transmits in the next slot. Thestations in the second subset must wait until all the stations

in the first subset have succeeded. If the collision is not


Fig. 14. Poisson traffic in random topologies (horizontal axis uses log-scale). (a) Aggregate throughput. (b) Average access delay. (c) Access energy

cost (Y axis uses log-scale).



resolved, then a further splitting into subsets takes place.Unfortunately, in tree splitting algorithms, a station alsoneeds ternary channel feedback to get hold of the conten-tion resolution progress.

In wireless networks, especially multihop networks,channel feedback is not reliable in terms of reflecting actualmedium statusdue to thewell-known “hidden terminal”and

“exposed terminal” problems. Consequently, the existingtreesplitting protocols and stabilization techniques cannot beused [14], and researchers continue to search for efficientcontention resolution algorithms for wireless networks.

Among them, some prior research work, e.g., [27],proposes to use one common subchannel to schedulepackets for multiple data subchannels. Compared with thepipelined packet scheduling schemes discussed in thispaper, the fundamental difference lies in that, in the priorschemes, the exchange of control messages to decide whichstation will transmit on an available data channel occurswhen at least one of the channels is perceived as idle.Contrary to this, the contention resolution of the pipelined

schemes proceeds for packets to be transmitted in the futurewhen the channel is currently busy.

Some reservation schemes are also proposed (e.g., [28]).To use a reservation scheme, a station has to know beforehand when the next packet will arrive, which maynot be feasible in many applications. Additionally, con-tending stations within a station’s two-hop neighborhoodneed to be aware of the channel reservation made by thisstation, which involves significant overhead when thenetwork topology changes often due to mobility.

Yuang et al. [29] proposes a hexanary-feedback conten-tion access scheme for infrastructure based wireless net-works. In their proposed scheme, mobile stations are

allowed to have full knowledge of the exact number of

users involved in a collision with the support of hardware.A two-phase contention resolution process is then used to

efficiently resolve channel contention. Tay et al. [30] noticethat some wireless networks have the bursty event-basedtraffic patterns and propose a nonpersistent carrier sensemultiple access protocol that attempts to minimize colli-

sions between contending stations for such networks.

Cali et al. [9], [10] propose a stabilized “Dynamic 802.11”algorithm specifically for a wireless LAN, which is a specialtype of wireless networks in which all stations can heareach other’s transmission and contend for a common

channel. “Dynamic 802.11” estimates the channel idle timeand collision time from the observed channel activity, thus,minimizes the contention resolution overhead. However,due to the location-dependent contention and hidden

terminal problems in multihop networks, local channelactivities observed do not reflect the actual status of thecontended channel. Additionally, “Dynamic 802.11” onlydeals with the situations where all contending stationsalways have packets ready for transmissions. [31] proposesan enhanced p-persistent IEEE 802.11 protocol to achieve

both the channel access efficiency and energy saving. Othercontention resolution optimization schemes for 802.11 DCFinclude [32], [33].

The advantages of DSCR when compared with existingwork lie in that, by applying pipelining techniques, DSCR isable to adapt to the limited available channel feedback inwireless networks and improves the contention resolutionefficiency over 802.11 DCF in both wireless LANs andmultihop networks.

In the rest of this section, we discuss “Dynamic 802.11”[9], [10] and a fast collision resolution algorithm (FCR) [33]

in more details.


Fig. 15. Example scenarios for multihop networks. (a) Scenario 1. (b) Scenario 2.



7.1 Discussion of “Dynamic 802.11”

Some prior research work (e.g., [5], [7], [9], [10]) proposes to

tune the contention window of 802.11 to improve itsperformance in wireless LANs. Particularly, the “Dynamic802.11” proposed in [9] dynamically adjusts each station’stransmission probability based on estimated channel idletime and collision time. In [9], it is argued that the networkcan operate at a point close to the optimal if the channeltime wasted on idle periods is equal to the channel timespent on collisions. Based on this argument, [9] proposes ascheme in which each station continuously observes thechannel activities to estimate the average idle/collisionperiod (“Dynamic 802.11” is proposed for basic accessmethod of 802.11 and a channel busy period is assumed to be a collision if an ACK does not immediately follow). Then

at the end of every transmission attempt, each stationcomputes its current estimation for the number of contend-ing stations and the optimal value of transmission prob-ability using the derived analytical results in [9]. Theoptimal transmission probability can be mapped to theaverage contention window value for each station, and thechannel utilization of “Dynamic 802.11” can be optimized.Intuitively, in “Dynamic 802.11,” a station will increase itstransmission probability when observed idle period islonger than collision period and decrease its transmissionprobability when the idle period is shorter than the collisionperiod.

However, the optimality of “Dynamic 802.11” relies on

stations’ capability of obtaining channel status correctly.“Missed ACK,” “Carrier Sensing Fault,” and “Not-detectedTransmissions” will cause idle periods/collision periods to be overestimated or underestimated, as noted in [10] aswell. Especially in multihop networks, incorrect observationof channel status will fundamentally mislead the algorithmand result in poor performance of “Dynamic 802.11.” Thedifference between “Dynamic 802.11,” DSCR, and 802.11 isfurther illustrated in Fig. 16 using simulation results forthree simple scenarios.

We use the basic access method (i.e., without the RTS/CTS handshake) because currently “Dynamic 802.11” isdesigned only for the basic access method [9]. Recall that

transmission range is assumed to be 250 meters, and carrier

sensing range is 550 meters for all four stations in thesimulations.

The first one is a typical wireless LAN scenario, in whichall four stations can hear each other’s transmissionscorrectly. In this case, “Dynamic 802.11” reaches its optimalpoint and achieves the best channel utilization, followed byDSCR and then 802.11.

In the second scenario, four stations are still within eachother’s carrier sensing range, but station 1 cannot correctlyreceive the packets transmitted by stations 3 and 4.Similarly, station 4 cannot correctly receive the packetstransmitted by stations 1 and 2. As shown in Fig. 16, theresult for “Dynamic 802.11” is that one of the two sourcestations (station 4 in this particular simulation run) willgrab the channel completely and the other one will be

starved. The reason is explained as follows: During asuccessful packet transmission from station 4, station 1 cansense the DATA transmission from station 4 and ACKtransmission from station 3, but can correctly receiveneither of them. Therefore, station 1 considers the transmis-sion from station 4 as collision, and its transmissionprobability will be decreased accordingly. On the otherhand, station 4 can receive ACK from station 3 correctly andthen increases its transmission probability. Consequently,station 4 has more chances than station 1 to transmit nextpacket. Next time when station 4 transmits, station 4 willagain increase its transmission probability and station 1 willagain decrease its transmission probability. In this way,

station 4 eventually grabs the channel and station 1 isstarved due to over-estimation of collision time. DSCR andIEEE 802.11 do not have such a problem, and DSCRachieves more throughput than 802.11, as the simulationresults show.

In the third scenario in Fig. 16, stations 1 and 4 aremoved further away and they cannot sense each other’stransmissions anymore. When station 4 is transmitting,station 1 will sense the channel as idle. Since station 1overestimates the channel idle period, it tends to transmitmore aggressively than it should. A similar case happens tostation 4 as well. When stations 1 and 4’s transmissionsoverlap in time, enough interference is caused at stations 2

and 3 and collisions occur. The collision probability is high


Fig. 16. Comparison of dynamic 802.11 with DSCR and IEEE 802.11 (without RTS/CTS, payload size: 512 bytes).



since both stations 1 and 4 have inappropriately hightransmission probability on average. The performance of “Dynamic 802.11” degrades dramatically compared with itsperformance in scenario 1.

The primary problem of “Dynamic 802.11” is that itentirely relieson theestimated channel idletimeand collisiontimeto calculate the transmission probability. If complete and

accurate channel feedback is available, it can achieve theoptimal performance. Otherwise, its performance can bequite poor being misled by incorrect channel feedback.

On the other hand, using DSCR or IEEE 802.11, stations 1and 4 mainly exploit channel feedback from their owntransmissions (success or failure) to adjust the contentionwindow. When collisions occur, stations 1 and 4 will doubletheir contention window, thus reduce the collision prob-ability for next transmission attempt. The probabilisticprocedure can resolve the contentions between stations 1and 4 most of the time, even though the collision probabilityis higher than scenarios 1 and 2 due to the lack of ability forstations 1 and 4 to sense each other’s data transmissions. As

the simulation results show, the performance degradationof DSCR and IEEE 802.11 is much less than “Dynamic802.11” in this scenario. In this case, DSCR achieves morethroughput than both 802.11 and “Dynamic 802.11.” Insummary, “Dynamic 802.11” does not perform well inmultihop networks, compared with the proposed implicitlypipelined DSCR protocol.

7.2 Discussion of Fast Collision Resolution (FCR)

Kwon et al. [33] have proposed a Fast Collision Resolution(FCR) algorithm for wireless LANs. With FCR, whenever anew busy period is detected (could be either a collision or apacket transmission), all deferring stations will exponentiallyincrease their own contention window and generate a new

backoff counter. In this way, all deferring stations havelarger contention window compared with the winningstation; thus, the winning station has higher probability towin channel access again. As FCR reduces the collisionprobability by letting one of the contending stations occupythe channel for a certain period of time, unfairness issuearises. Comparing with 802.11, fairness among contendingstations could get worse and worse as the number of contending stations increases. To overcome this issue, FCRdefines a maximum transmission limit and uses distributedSCFQ algorithm to dynamically adjust this maximumtransmission limit. If a station consecutively succeeds inchannel access over this transmission limit, it will change its

contention window to a maximum value in order to giveother stations opportunities to transmit.

The fundamental difference between FCR and DSCR isthat, FCR reduces collision probability by letting one stationoccupy the channel for a relatively long time, while DSCRreduces collision probability by statistically selecting asubset of stations to contend for the channel at any giventime. As the group of pipelined stations staying in stage 2 of DSCR generally change after every successful transmission,DSCR is able to maintain comparable fairness to 802.11.

Currently, FCR is proposed only for wireless LANs [33].As FCR uses SCFQ fair scheduling algorithm to resolve theunfairness issue, it is not clear how to extend FCR further to

multihop networks. On the other hand, DSCR can work

well both in wireless LANs and multihop networks, as we

showed earlier using simulation results.

8 CONCLUSION

Pipelined packet scheduling is discussed in this paper. In

particular, a MAC protocol—DSCR (pipelined Dual-Stage

Contention Resolution)—that uses an implicitly pipelinedcontention resolution algorithm is proposed. By applying

implicit pipelining in DSCR, most of the channel idle

overhead is hidden and the channel contention is reduced

without introducing any additional cost. Extensive simula-

tion results show that DSCR significantly improves the

performance of 802.11 DCF in both wireless LANs and

multihop networks in terms of channel utilization, average

access delay and access energy cost.Overall, without dependence on extensive channel feed-

back information (e.g., as in “Dynamic 802.11” [9]), without

relying on elimination burst or other signaling mechanisms

(e.g., as in HIPERLAN/I [19]), DSCR provides an improvedperformance for IEEE 802.11 DCF and, hence, demonstrates

that pipelining techniques can be useful in improving the

performance of multiple access control protocols.

ACKNOWLEDGMENTS

The authors would like to thank the referees for their

helpful comments. Preliminary versions of this paper

appeared at the ACM International Symposium on Mobile

Ad Hoc Networking and Computing (Mobihoc ’03) poster

session and IEEE Semiannual Vehicular Technology Con-

ference Fall 2003 (invited paper). This research is supported

in part by US National Science Foundation grant 01-96410.

REFERENCES

[1] “Wireless LAN Medium Access Control (MAC) and PhysicalLayer (PHY) Specifications,” IEEE Standard 802.11, June 1999.

[2] H.S. Chhaya and S. Gupta, “Performance Modeling of Asynchro-nous Data Transfer Methods of IEEE 802.11 MAC Protocol,”Wireless Networks, vol. 3, pp. 217-234, 1997.

[3] B.P. Crow, I. Widjaja, J.G. Kim, and P.T. Sakai, “IEEE 802.11Wireless Local Area Networks,” IEEE Comm. Magazine, vol. 2,pp. 116-126, Sept. 1997.

[4] Y.C. Tay and K.C. Chuan, “A Capacity Analysis for the IEEE802.11 MAC Protocol,” ACM/Baltzer Wireless Networks, vol. 7, no. 2,pp. 159-171, Mar. 2001.

[5] J. Weinmiller, H. Woesner, J.P. Ebert, and A. Wolisz, “Analyzing

and Tuning the Distributed Coordination Function in the IEEE802.11 DFWMAC Draft Standard,” Proc. Int’l Workshop Modelling MASCOT, Feb. 1996.

[6] J. Weinmiller, M. Schlager, A. Festag, and A. Wolisz, “Perfor-mance Study of Access Control in Wireless LANs IEEE 802.11DFWMAC and ETSI RES 10 HIPERLAN,” Mobile Networks and

Applications, vol. 2, pp. 55-67, 1997.[7] G. Bianchi, “Performance Analysis of the IEEE 802.11 Distributed

Coordinated Function,” IEEE J. Selected Areas in Comm., vol. 18,no. 3, pp. 535-547, Mar. 2000.

[8] G. Bianchi, L. Fratta, and M. Oliveri, “Performance Evaluation andEnhancement of the CSMA/CA MAC Protocol for 802.11 WirelessLANs,” Proc. Ann. IEEE Int’l Symp. Personal Indoor and Mobile RadioComm. (PIMRC), vol. 2, pp. 392-396, Oct. 1996.

[9] F. Cali, M. Conti, and E. Gregori, “IEEE 802.11 Protocol: Designand Performance Evaluation of an Adaptive Backoff Mechanism,”IEEE J. Selected Areas in Comm., vol. 18, no. 9, pp. 1774-1786, Sept.

2000.




[10] F. Cali, M. Conti, and E. Gregori, “Dynamic Tunning of the IEEE802.11 Protocol to Achieve a Theoretical Throughput Limit,” IEEE/

ACM Trans. Networking, vol. 8, no. 6, pp. 785-799, Dec. 2000.[11] B. Hajek and T. VanLoon, “Decentralized Dynamic Control of a

Multiaccess Broadcast Channel,” IEEE Trans. Automatic Control,1982.

[12] R.L. Rivest, “Network Control by Bayessian Broadcast,” TechnicalReport MIT/LCS/TM-285, Laboratory for Computer Science,Massachusetts Inst. of Technology, Cambridge, Mass., 1985.

[13] J.I. Capetanakis, “Tree Algorithms for Packet Broadcast Chan-nels,” IEEE Trans. Information Theory, vol. 25, no. 5, pp. 505-515,Sept. 1979.

[14] D. Bertsekas and R. Gallager, Data Networks, second ed. PrenticeHall, 1992.

[15] J. Hennessy and D. Patterson, Computer Architecture: A Quantitative Approach, third ed. Morgan Kaufmann, 2002.

[16] X. Yang and N.H. Vaidya, “Explicit and Implicit Pipelining forWireless Medium Access Control,” Proc IEEE Semiann. VehicularTechnology Conf., Fall 2003.

[17] X. Yang and N.H. Vaidya, “Pipelined Packet Scheduling inWireless LANs,” technical report, Coordinated Science Labora-tory, Univ. of Illinois at Urbana-Champaign, Aug. 2002.

[18] J. Deng, Y.S. Han, and Z.J. Haas, “Analyzing Split ChannelMedium Access Control Schemes with ALOHA Reservation,”Proc. Second Int’l Conf. AD-HOC Networks and Wireless (ADHOC-NOW ’03), Oct. 2003.

[19] ETSI TC-RES, “Radio Equipment and Systems (RES); HIghPErformance Radio Local Area Network (HIPERLAN) Type 1;Functional Specification,” Oct. 1996, European Telecomm. Stan-dard ETS 300 652.

[20] VINT Group, “UCB/LBNL/VINT Network Simulator NS (Ver-sion 2),” http://www.isi.edu/nsnam/ns, 1996.

[21] L.M. Feeney and M. Nilsson, “Investigating the Energy Con-sumption of a Wireless Network Interface in an Ad HocEnvironment,” Proc. IEEE INFOCOM Conf., 2001.

[22] Z. Fu, P. Zerfos, H. Luo, S. Lu, L. Zhang, and M. Gerla, “TheImpact of Multihop Wireless Channel on TCP Throughput andLoss,” Proc. IEEE INFOCOM Conf., 2003.

[23] S. Xu and T. Saadawi, “Does the IEEE 802.11 Mac Protocol WorkWell in Multihop Wireless Ad Hoc Networks,” IEEE Comm.

Magazine, vol. 39, no. 6, pp. 130-137, June 2001.[24] S. Xu and T. Saadawi, “Revealing the Problems with 802.11

Medium Access Control Protocol in Multihop Wireless Ad HocNetworks,” Computer Networks, vol. 38, no. 4, pp. 531-548, Mar.2002.

[25] R.G. Gallager, “Conflict Resolution in Random Access BroadcastNetworks,” Proc. AFOSR Workshop Comm. Theory and Applications,Sept. 1978.

[26] B.W. Wah and X. Su, “An Efficient Multiaccess Protocol forWireless Networks,” Proc. Int’l Symp. Internet Technology, pp. 173-178, Apr. 1998.

[27] A. Tzamaloukas and J.J. Garcia-Luna-Aceves, “A Receiver-Initiated Collision-Avoidance Protocol for MultiChannel Net-works,” Proc. IEEE Infocom Conf., 2001.

[28] G.R. Hiertz, J. Habetha, P. May, E. Weiss, R. Bagul, and S.Mangold, “A Decentralized Reservation Scheme for IEEE 802.11Ad Hoc Networks,” Proc. IEEE Personal Indoor Mobile Radio Conf.(PIMRC), Sept. 2003.

[29] M.C. Yuang, B.C. Lo, and J.-Y. Chen, “Hexanary-Feedback

Contention Access with PDF-Based Multiuser Estimation forWireless Access Networks,” IEEE Trans. Wireless Comm., vol. 3,no. 1, pp. 278-289, Jan. 2004.

[30] Y.C. Tay, K. Jamieson, and H. Balakrishnan, “Collision-Minimiz-ing CSMA and Its Applications to Wireless Sensor Networks,”IEEE J. Selected Areas in Comm., vol. 22, no. 6, pp. 1048-1057, Aug.2004.

[31] R. Bruno, M. Conti, and E. Gregori, “Optimization of Efficiencyand Energy Consumption in p-Persistent CSMA-Based WirelessLANs,” IEEE Trans. Mobile Computing, vol. 1, no. 1, pp. 10-31, Jan.2002.

[32] H. Kim and J.C. Hou, “Improving Protocol Capacity with Model-Based Frame Scheduling in IEEE 802.11-Operated Wlans,” Proc.

ACM MobiCom, Sept. 2003.[33] Y. Kwon, Y. Fang, and H. Latchman, “A Novel Medium Access

Control Protocol with Fast Collision Resolution for Wireless

Xue Yang received the BE and MS degreesfrom University of Electronic Science andTechnology of China and the PhD degree fromthe University of Illinois at Urbana-Champaign(UIUC). She is awarded Vodafone-U.S. Founda-tion Graduate Fellowship from 2003 to 2005. Hercurrent research is in the areas of wirelessnetworking and mobile computing, with the focuson medium access control, quality of service,

and topology control. For more information,please visit http://www.crhc.uiuc.edu/~xueyang/. She is a member ofthe IEEE.

Nitin H. Vaidya received the PhD degree fromthe University of Massachusetts at Amherst. Heis presently an associate professor of electricaland computer engineering at the University ofIllinois at Urbana-Champaign (UIUC). He hasheld visiting positions at Microsoft Research,Sun Microsystems, and the Indian Institute ofTechnology-Bombay. His current research is inthe areas of wireless networking and mobilecomputing. His research has been funded by

various agencies, including the National Science Foundation, DARPA,BBN Technologies, Microsoft Research, and Sun Microsystems. He is arecipient of a CAREER award from the US National Science Foundation.

He has served on the program committees of several conferences andworkshops, and served as program cochair for the 2003 ACM MobiCom.He has served as editor for several journals, and presently serves aseditor-in-chief of the IEEE Transactions on Mobile Computing . For moreinformation, please visit http://www.crhc.uiuc.edu/~nhv/. He is a seniormember of the IEEE and the IEEE Computer Society.

. For more information on this or any other computing topic,please visit our Digital Library at www.computer.org/publications/dlib.


32-A Wireless MAC Protocol Using Implicit Pipelining

Documents

Transcript of 32-A Wireless MAC Protocol Using Implicit Pipelining