Wireless Netw (2011) 17:11731189 DOI 10.

1007/s11276-011-0338-7

On enabling cooperative communication and diversity combination in IEEE 802.15.4 wireless networks using off-the-shelf sensor motes
Muhammad U. Ilyas Moonseong Kim Hayder Radha

Published online: 21 April 2011 Springer Science+Business Media, LLC 2011

Abstract This paper presents the Generalized Poor Mans SIMO System (gPMSS) which combines two approaches, cooperative communication and diversity combination, to reduce packet losses over links in wireless sensor networks. The proposed gPMSS is distinct from previous cooperative communication architectures in wireless sensor networks which rely on a relay channel, and also distinct from implementations in 802.11 networks that require a wired infrastructure or hardware changes for cooperation. gPMSS foregoes the need for any changes to mote hardware and it works within the current IEEE 802.15.4 standard. We describe the gPMSS protocol that governs the cooperation between receivers. Three variants are evaluated including selection diversity, equal gain and maximal ratio combining. First, we demonstrate gPMSS on bit error traces in a fully reproducible manner. This is followed by an implementation of gPMSS in C# on the .NET Micro Framework edition of the recently released Imote2 mote platform. We demonstrate by means of experiments an increase in the packet reception rate from 2230% to 7376%, a relative increase of 150245%. We

also analyzed the power consumed by the transmitter per delivered packet and observe a reduction of up to 68%. We also take into account the retry limit of the IEEE 802.15.4 protocol and demonstrate that gPMSS is able to provide 99% packet delivery at the protocols default retry parameters against 6575% without it. Keywords Wireless sensor networks Receiver cooperation Diversity combining IEEE 802.15.4

1 Introduction Channel fades and interference effects limit the throughput, useful communication range and (in case of battery powered devices) lifetime of nodes. In this chapter we describe the generalized Poor-Mans-SIMO-System (gPMSS), a readily deployable low-cost, low-power, protocol centric approach that enables cooperative communication in IEEE 802.15.4 [1] wireless sensor networks (WSN). We demonstrate that gPMSS reduces the fraction of packets that are received with bit errors or not received at all by an order of magnitude, thus reducing the number of retransmissions. It makes the use of long range links that are unfeasible due to high packet loss and retransmission rates feasible again. We also show that even in instances where gPMSS is not able to correct all errors from a packet it still succeeds in reducing the number of bit errors. At the receiver side gPMSS uses diversity combining methods adapted from their analog domain counterparts of the same name [6] for digital signals. What makes the application of single-input multiple-output (SIMO) diversity combining principles novel from traditional use is that they are applied to the demodulated version of received packets, after Physical layer processing. We demonstrate the efcacy of gPMSS

The preliminary version of this paper titled Reducing Packet Losses in Networks of Commodity IEEE 802.15.4 Sensor Motes Using Cooperative Communication and Diversity Combination was published in the proceedings of the IEEE Conference on Computer Communications (Infocom), Rio de Janeiro, Brazil, Apr. 1925, 2009. M. U. Ilyas (&) M. Kim H. Radha Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI 48824, USA e-mail: ilyasmuh@egr.msu.edu M. Kim e-mail: mkim@egr.msu.edu H. Radha e-mail: radha@egr.msu.edu

123

1174

Wireless Netw (2011) 17:11731189

by applying it to bit error traces collected from IEEE 802.15.4 channels that allow detailed analysis and precise reproduction of results. We also demonstrate gPMSS effectiveness under real-world conditions by implementation on Crossbows Imote2 .NET Micro Framework sensor platform [12]. Enabling the use of long range links (that would otherwise not be used) makes gPMSS a viable protocol due to the benets and utility of such links by several applications in wireless sensor networks. Network lifetime extension Funneling is the effect of network trafc from multiple sources owing to a small number of sink nodes [26]. This trafc surge produces congestion in the region around the sink nodes/base station, forcing nodes near sink nodes to relay more trafc than other nodes and consume power at correspondingly higher rates. Since nodes in WSNs have only limited power resources this means that the sink nodes neighbors will run out of power sooner, leaving the sink node disconnected from the rest of the network. Load balancing techniques like [15] attempt to distribute the burden of relaying trafc to increase the lifetime of sensor networks. Employing gPMSS in such a scenario will grow the set of neighbor nodes of the sink node and allow load balancing among more nodes. Small-world networks Several attempts have been made at building small-world network [27] topologies in wireless networks to simplify resource discovery and reducing average path length to facilitate data dissemination. Proposed architectures required hardware modications such as adding a secondary RF interface [25, 26] or building hybrid networks by augmenting wireless networks with wired shortcuts [9, 22]. Since gPMSS is a protocol centric approach it does not require any hardware modications which adds to its appeal as a low-complexity and low-cost solution. Network connectivity Long range links can be used to add links between two components of a network that are only sparsely connected with one another. gPMSS adopts well-understood diversity combining methods for analog signals and applies them to digital signals (packets). Specically, gPMSS implements selection diversity, equal gain diversity combining and maximal-ratio gain diversity combining. The latter relies on a model of the instantaneous bit error rate (BER) driven by channel state information (CSI) [16], i.e. received signal strength indication (RSSI) and link quality indication (LQI). We provide proof of concept by applying gPMSS to bit-error traces and demonstrate one order of magnitude reduction in packet losses. Applying gPMSS to traces allows more detailed analysis and reproducibility that is not possible in a live setup, i.e. the event when receivers are not able to reconstruct an error-free version of the

transmission. We show that even then we are able to signicantly reduce the average BER of incorrigible packets. Finally, we implement gPMSS on Imote2 sensor motes [12] using C# and demonstrate a clear reduction in packet losses. Experimental results from IEEE 802.15.4 links indicate that using diversity combining raises packet reception rate (PRR) by up to an additional 130% over those in a single receiver. Our contributions are threefold; 1. gPMSS is a protocol centric, cross-layer approach which means it can be used in presently deployed wireless sensor networks by making software changes only. It does not require any modications to hardware but runs on networks of commercial off-the-shelf (COTS) single antenna sensor motes. gPMSS is non-intrusive in the sense that it does not require changes to the pre-existing IEEE 802.15.4 standard. gPMSS is able to reduce power consumed at the transmitter per packet delivered by up to 68%. This represents a signicant increase in the lifetime of sensor node.

2.

3.

Figure 1 illustrates the difference between routes traversed by a packet sent by transmitter T to a distant node R1 when gPMSS is used (dotted arrows represent long range links, solid lines represent links between R1, R2, R3 that form a fully connected graph), and the multi-hop path from node T to R1 when it is not used (solid arrows). The remainder of this chapter is organized as follows: Section 2 reviews some related works. Section 3 describes the three diversity combining techniques for packet recovery. Section 4 describes the gPMSS that enables cooperation between multiple receivers. Section 5

R3 R2 R1 High packet loss communication range

T Low packet loss communication range

Fig. 1 Application of generalized gPMSS in a wireless sensor network with mesh topology. Path from transmitter T to receiver R1 marks the multihop path that would be taken in a network without gPMSS. Dashed line links between T and receivers R1, R2 and R3 denote the longer range but high loss links that are used under Generalized gPMSS

123

Wireless Netw (2011) 17:11731189

1175

describes the trace collection setup and demonstrates a proof of concept of gPMSS in a manner that can be reproduced. Section 6 describes the gPMSS implementation on Imote2 and its results. Section 7 discusses our results in terms of PRR, retransmission attempts and energy consumption per packet. Section 8 concludes this chapter.

recent and most relevant work using cooperative receiver diversity is Bletsas and Lippman [5] and Bletsas et al. [4]. However, this paper offers several improvements over Bletsas et al. approach: Bletsas et al. rely on selection diversity alone, i.e. a transmission can be received successfully only if at least one of the cooperating receivers has an error-free reception. No attempt is made at correcting packets that are received with errors. gPMSS lls this gap by supplementing selection diversity with various diversity combining methods. Bletsas protocol relies on the exchange of IEEE 802.11x like request-to-send/clear-to-send (RTS/CTS) packets prior to the actual data transmission to clear the channel and inhibit interference. Since the gPMSS protocol presented in this paper is based on IEEE 802.15.4, it forgoes use of RTS/CTS packets which reduces power consumption. Bletsas uses a pilot signal transmitted by the sender to select a relay node prior to data transmission based on network conditions. In gPMSS, as long as any one candidate relay node has received a transmitted packet free of errors, the selection of the relay node is performed without any packet transmission overhead, on a packet-by-packet basis. Bletsas et al. used COTS hardware for their cooperating receivers. However, their denition of COTS is very broad in the sense that they use the term to describe custom built mote platforms using COTS components. We use the term COTS in a stricter sense that includes only commercial mote platforms and precludes any specially designed or modied systems, even if built from commercially available components. This paper demonstrates gPMSS on unmodied Crossbow Imote2 [12] platforms [12], a truly COTS platform.

2 Related work The concept of spatial receiver diversity is not new and has been studied extensively in the analog signal domain. Chakraborty et al. proposed the Extended ARQ scheme [7] that recombines spatially diverse versions of a received packet to detect bit errors and an exhaustive search to correct them if their number is less than a threshold value. Extended ARQ has a lot in common with the version of gPMSS that uses equal gain combining and is agnostic of what MAC standard is used, but the results provided in [7] are based on theoretical analysis only. Miu et al. [19] proposed a system that used transmitter diversity to increase packet reception rate in IEEE 802.11 [13] networks with multiple access points (AP) as senders. The scheme roughly corresponds to gPMSS with selection diversity, without diversity combination for error correction. Miu et al. generalized this approach in [21] for applications beyond streaming video. Miu [20] extended the idea further to reduce packet losses on the uplink (mobile device to AP). However, this required modications to IEEE 802.11b AP hardware or deployment of more APs, and uses a dedicated frame combiner connected to all APs through a wired network. It used the equal gain method for detecting bit errors and, like Extended ARQ, relied on an exhaustive search of the correct bit values. Cheng and Valenti [8, 24] extended the idea for improving throughput on uplinks in IEEE 802.11a networks by using maximal ratio combining based on CSI measurements. However, like Mius system it still required a dedicated combiner connected to all APs. Ji et al. [17] proposed an approach for improving the throughput of downlinks by scheduling transmissions to multiple receivers in IEEE 802.11a/b networks based on explicit feedback from receivers while maintaining fairness. Bahl [2] made the case for multi-radio transceivers, but as Fig. 4 in his paper showed, collaboration between network interfaces is possible only when they are all located on the same device. More recently, Woo described SOFT [28] which also exploited receiver diversity for the uplink in IEEE 802.11 networks similar to Mius [20], but with diversity combining being performed using maximal ratio combining. Therefore, it too requires a centralized combiner on the wired network that all APs are connected to. The most

To summarize, the gPMSS system presented is distinct from all these prior works on cooperative communication and diversity combining in wireless networks because it is (1) designed for IEEE 802.15.4 networks, (2) is purely implemented in software and commercial-off-the-shelf motes without modications to mote hardware, (3) is tested on bit error traces collected from real IEEE 802.15.4 channels, (4) as well as actual implementation on motes.

3 SIMO diversity combining techniques The solution that is described in this section is dubbed the Generalized Poor-Mans-SIMO-System because it uses receiver side diversity combining techniques and is built using commercial-off-the-shelf components, without

123

1176

Wireless Netw (2011) 17:11731189

customized or recongured hardware. Receiver diversity improves link quality of wireless channels with high losses. This way we reduce losses and retransmissions and increase throughput and channel utilization. This subsection describes linear diversity combining techniques. All these techniques are derivatives of the techniques by the same names presented by Brennan [6]. Brennan describes scanning diversity, selection diversity, equal gain diversity and maximal-ratio diversity combining. Although the methods described by Brennan were meant for analog signals, we have suitably modied and adapted them for use with demodulated, digital signals. We have included the last three, selection, equal gain and maximal-ratio diversity combining. Readers should know that even when the diversity combining method used is either equal gain or maximal-ratio combining, selection diversity is used whenever at least one receiver possesses an error-free version of a transmissions. Equal gain or maximal ratio combining are only used when none of the gPMSS receivers was able to receive error-free (i.e. the situation described in Fig. 3c). The purposes of diversity combining are twofold. 1. 2. Select an error-free version of a received transmission from among all received versions. If the rst goal is not achievable, obtain another version of the transmission, with fewer errors than any of the individual received versions.

BER Model

BER Model

BER Model

(a)

(b)
BER Model

BER Model

BER Model

(c)
Fig. 2 Illustration of logical functioning of various diversity combining techniques. a Selection diversity. b Equal gain diversity. c Maximal ratio gain diversity

3.1 Selection diversity Selection diversity is the simplest diversity combining technique. Figure 2a is an equivalent system diagram of the selection diversity process. The basic idea in is to select from all received packets the one that is expected to have the fewest errors. This is advantageous when it is used in conjunction with forward error correction (FEC) because fewer bit errors are easier to correct than more bit errors. When all received versions have errors, the best selection diversity can hope to achieve is pick the version with the fewest bit errors. We dene the bit error rate (BER) of the nth packet in a sequence as, BER bn # of error bits in nth recvd pkt : # of bits in nth recvd pkt 1

Thus the underlying random process producing the sequence of BER observations b[n] is called the BER process and is denoted by B. The term BER is not used in its strict traditional sense where it denotes the long term average probability of bit errors, such as in a binary symmetric channel (BSC). Instead the BER is computed over each received packet. Unfortunately, under ordinary circumstances the BER process is not directly observable.

A packets failure to pass the cyclic redundancy check (CRC) test only tells us that the number of bits with errors is non-zero (b [ 0), but it does not give any information about the number of errors. Therefore, we must rely on estimates of the BER. The performance of selection diversity will be determined by the accuracy of the model used to predict the BER of packets that fail the CRC test. We have used Ilyas and Radhas [16] CSI measurementbased model of the BER process on IEEE 802.15.4 links. It models the BER of packets with errors by a random variable with an exponential distribution whose parameters are estimated using maximum likelihood estimation (MLE). Packets are rst classied according to CSI measurements and separate BER distributions are generated for each. Parameters for Ilyas and Radhas CSI measurement-based model is based on an extensive set of bit error traces. For each received packet the model relies on two CSI parameters, i.e. LQI and RSSI. Measurement of both RSSI and LQI is mandated by the IEEE 802.15.4 LR-WPAN standard for every received packet. The RSSI random process is denoted by P, and RSSI measured by a receiver R for the nth packet in a sequence is denoted by qR[n]. We used the MICAz [11] to demonstrate proofof-concept and the Imote2 [12] to demonstrate the

123

Wireless Netw (2011) 17:11731189 Fig. 3 gPMSS protocol operations. a Reception of an error free packet by a gPMSS cluster. b gPMSS message exchanges when parent receiver R1 receives message with errors but child R2 receives error-free. c gPMSS message exchanges for recovery of data when neither parent R1 nor children R2 and R3 receive error-free

1177

D(T) Time ACK

T0

(a)

(b)

(c)

functioning gPMSS protocol implementation, both of which use the Chipcon CC2420 radio transceiver [23]. Note that almost all commercially available wireless sensor platforms with IEEE 802.15.4 RF interfaces currently use the either the Chipcon CC2420 or its newer variant the CC2430 radio transceiver. Therefore, we expect the results and conclusions drawn to hold across a wide range of different mote platforms. Technically, the CC2420 does not measure the LQI directly. Instead, it measures the correlation C between the rst 8 received symbols (of the PHY header) and the corresponding 8 known symbols (Preamble). IEEE 802.15.4 uses 16-ary Offset-Quadrature Phase Shift Keying modulation which encodes 4 bits in one symbol. The rst 8 symbols, 4 bytes, of the PHY header comprise of the Preamble sequence consisting of 32 binary zeros. The LQI is then dened as, LQI C c1 c2 : 2

random process is denoted by K, and LQI measured by a receiver R for the nth packet in a sequence is denoted by kR[n]. Coming back to our description of the CSI-driven BER model of [16], each pair of LQI and RSSI inputs produces a probability density function (PDF) of the BER of packets received with those particular CSI measurements. To be useful in the current context, the output of the CSI-driven BER model has to be mapped to a single value. We use bX% to denote the Xth percentile of the BER process PDF (b50% is Bs mean). The instances of the BER model return BER estimates denoted as b(R1), b(R2) and b(R3). The output selector in Fig. 2a receives as input the estimated BERs b(R1), b(R2) and b(R3). Based on these estimates it selects the receiver with the lowest BER estimate as the least errorprone one and accepts its received copy as the best one and outputs it as D(Sel), i.e. DSel DRr : r arg min bRi :
i

In the Chipcon CC2420 c1 and c2 are functions of the packet error rate (PER) measured over an extended period of time and are determined experimentally. c1 and c2 scale the 7 bit value of the correlation to the range of an 8 bit number. Since equation 2 is merely a shifting and scaling of the measured C we take c1 = 0 and c2 = 1. The LQI

3.2 Equal gain diversity The equal gain diversity combining method described here is depicted by an equivalent system diagram in Fig. 2b.

123

1178

Wireless Netw (2011) 17:11731189

e e Recall that like D(T), the three received copies D R1 ; D R2 R3 e and D are vectors of binary numbers (representing bits) obtained after demodulation of the received carrier signal. Essentially, equal gain diversity combining uses received e e e data D R1 ; D R2 and D R3 to vote on the value of each output bit. In the example in Fig. 2b performs vector e e e addition of D R1 ; D R2 and D R3 , stores the sums in integers and then adjusts the gain by dividing by the number of receivers N, where N = 3 in this example. The result will be an array of rational numbers in the range [0,1]. These numbers are thresholded such that values less than 0.5 are remapped to binary zeros, and values greater than (or equal to) 0.5 to binary ones. The output of the thresholder is D(EG). If S is a function representing the operation of the binary decision thresholder, then for an N-receiver gPMSS cluster the equal gain diversity combining process can be represented as; ! N 1 X Ri DEG S D : 4 N i1 Equal gain diversity combining has two advantages over the preceding selection diversity combining. 1. 2. It has lower complexity because it does not rely on a BER model. The diversity combining procedure may output a copy of the transmitted packet that has fewer errors or is completely error-free, even when all individual received copies are not.

4 gPMSS protocol This section describes the operation of the gPMSS protocol. Assume a WSN consisting of a large number of singleantenna COTS receivers communicating over multiple hops with the base station collecting data. According to some topology construction algorithm, a node R1 is chosen as an upstream end-point of a link. To use R1 as part of a set of multiple receivers we propose the gPMSS protocol that denes the message exchange between cooperating receiver nodes to handle transmissions that are received with errors or not received at all. The following subsection provides a brief overview of gPMSS protocol message exchanges for four important operations. For illustrative purposes we assume a scenario in which there is a distant transmitter T and a receiver R1 with two neighbor nodes R2 and R3 that are located close enough to communicate with R1 with few losses. 4.1 gPMSS cluster creation The Poor Mans SIMO System (PMSS) described by Ilyas, Kim and Radha [14] differs from gPMSS in the way clusters of receivers are formed. In PMSS, cluster creation is explicit, and involved an exchange of messages between R2 and R1 and also between R3 and R1 after which R2 and R3 would become associated with R1 to act as cooperating receiver. In gPMSS nodes take advantage of CSI of overheard messages. The assistance rendered by neighbors to a node R1 is now ad-hoc. The decision by a neighbor node whether it is in a position to assist R1 is based on historical link conditions between it and R1. Link conditions can be simply assessed by tracking historical packet retransmission rates on a link, or LQI/RSSI measurements. Links exhibiting performance a certain threshold level may be classied as good. Figures 7 and 8 density functions of LQI and RSSI of packets originating from R1, R2 and R3. Nodes in a network with the gPMSS protocol will maintain such histograms for all neighbors from which they overhear trafc. A high mean, median or mode of LQI and RSSI density functions is indicative of a link with high PRR. In this way, once a node determines it enjoys good link conditions with a neighbor it will act as a member of that neighbors cluster of receive nodes. Also, in PMSS cooperating nodes would communicate with R1 in a scheduled, round-robin fashion. In contrast, in gPMSS once it is determined that cooperating receivers need to communicate with R1, transmission times are chosen randomly. For more details about PMSS we refer the reader to [14]. For the following discussion we will assume that this way two nodes R2 and R3 placed close to R1 make the assessment that they enjoy a reliable wireless channel with R1 and volunteer to assist it as cooperating receivers.

3.3 Maximal ratio diversity The maximal ratio diversity combining method described here is depicted by an equivalent system diagram in Fig. 2c. It combines elements from selection and equal gain diversity combining. Maximal ratio combining can be described as equal gain diversity but with weighted addie e e tion. D R1 ; D R2 and D R3 are each multiplied by weights w1, w2 and w3 computed as, wi 1 2bRi 81 i N 5

and added. The sum is then re-normalized by dividing by the number of receivers N (in this case N = 3) and thresholded which returns the output D(MR) of the maximal ratio combining process; ! N  1 X MR Ri D S wi D : 6 N i1 In the following subsection we proceed to describe the gPMSS protocol that enables cooperation between receivers.

123

Wireless Netw (2011) 17:11731189

1179

4.2 Error-free reception by at least one recipient This section describes the exchange of messages under the gPMSS protocol that occurs when at least any one of the receiving nodes receives a transmitted packet without errors. Figure 3a depicts the simplest case. The solid lines represent the transmission and reception of a message between source and destination node. The dotted lines represent communication that occurs implicitly as a result of a receiver operating in promiscuous mode, (deliberately) eavesdropping on messages exchanged between other nodes (marked by solid lines). Here T sends a data message D(T) to R1 at time 0 that is overheard by R2 and R3. R1 will promptly responds to T with an ACK within time T0 from the initial transmission. R2 and R3 overhear the ACK from R1 back to T within time T0 and recognize that the packet was successfully received by R1 and acknowledged, and no further action is required. Figure 3b depicts the case where R1 is not the nal destination. In addition, let us also assume that R1 receives the transmission D(T) with errors (marked by a zigzagged arrow), whereas R2 and R3 receive the same error-free. After the initial timer T0 expires, all receivers that receive D(T) error-free choose a random wait-time t1 from an exponential PDF limited to the range [0,T1]. Let t1(R2) and t1(R3) denote R2 and R3s random wait-times, respectively. Let t1(R2) \ t1(R3), then R2 will transmit ACK back to T before R3. R3 will overhear R2s ACK and cancel transmission of its own ACK. At any time, if an ACK packet is lost and not received by T within time TT of transmitting data packet D(T), Ts fallback behavior will be to retransmit D(T) (although it may already have been received and ACKed). This way the power consumed in nodes forming the gPMSS cluster to relay packets will be more evenly distributed. 4.3 Erroneous reception by all recipients This section describes the exchange of messages under the gPMSS protocol that occurs when all nodes that form a gPMSS cluster receive a transmission with errors. Figure 3c depicts this entire transaction. Here T sends a data message D(T) to R1 that is overheard by R2 and R3. Since all receivers R1, R2 and R3 receive with errors none of them is able to e e respond to T with an ACK within time T1. Let D R1 ; D R2 eR3 denote the different versions of D(T) as they are and D received by R1, R2 and R3, respectively. Thus, there is no error-free copy of the transmitted message at any receiver. Nodes R1, R2 and R3 all wait for one another to respond to T with an ACK. When none of the receivers R1, R2 and R3 overhear an ACK going back to T within T0 ? T1 time of receiving, they infer that none of them received D(T) errorfree. At this point, the lack of an ACK packet from the

receiver informs T that the receivers are about to collectively attempt to recover the packet by means of diversity combining. That process will involve the exchange o multiple packets between R1, R2 and R3 which can be overheard by T and will, if the RF transceiver is left active, result in consumption of signicant amounts of energy. Therefore, T will disable its RF transceiver for the time period T2 during which receivers attempt diversity combining. At the receiver side, instead of requesting a retransmission from T, R1 collects the error-prone versions of D from cooperating receivers, acknowledging each one as it receives them. R2 e will transmit D R2 ; kR2 ; qR2 , which denotes the concateR2 e nation of D , the LQI k(R2) and RSSI q(R2) with which it was received from T, to R1 in time interval [T0 ? T1, e T0 ? T1 ? T2] after it received D R2 . Similarly, R3 will eR3 ; kR3 ; qR3 between [T0 ? T1, T0 ? T1 ? transmit D e T2] after it received D R3 . Once R1 has received e eR2 ; kR2 ; qR2 g and f D R3 ; kR3 ; qR3 g it executes one fD of the diversity combining algorithms described in the preceding section in an attempt to recover D(T). If the CRC computed from the recovered packet matches the appended CRC the attempt is successful. On the receiver side T waits for an ACK, any ACK from any of the receivers R1, R2 or R3, for a timeout period of TT until it attempts retransmission of D(T). Note that TT [ T0 ? T1 ? T2. It should also be noted that the reduction in retransmissions by T is achieved at the expense of an increase in the time between when a data packet is transmitted and a matching ACK is received. However, it should be noted that IEEE 802.15.4 protocol explicitly forgoes the use of an IEEE 802.11-like exchange of request-to-send(RTS)/clearto-send(CTS) packets. Therefore, any delays experienced by the transmitting node T in receiving an ACK do not unduly hold up the communication of other nodes not participating in the above described exchanges. However, other nodes are affected by the transmissions between receiver nodes that happens when diversity combining is attempted. Most current environmental monitoring, infrastructure monitoring, surveillance and other systems enabled by WSN try to keep packet transmission rates low to maximize the lifetime of power constrained sensors. Therefore, the reduction in capacity that results from diversity combining is assumed to be of little consequence for most applications. When selection diversity is used to avoid a retransmission by T capacity does not decrease.

5 Trace based proof of concept In this section we provide proof of concept of gPMSS by testing its performance on bit error traces. We collected several different sets of bit error traces totaling a few

123

1180

Wireless Netw (2011) 17:11731189

Octets: 1 Len

million packets in a way that provides, to the authors best knowledge, the BER a packet is subjected to and the LQI and RSSI with which it is received. The results shown in this section are generated from one of those traces. 5.1 Experimental setup

2 Frame Control

1 Sq No

2 Dest PAN ID

2 Dest Addr

29 Data / Payload

2 FCS

Typ Grp

4 SeqNo (2)

4 SeqNo (3)

4 SeqNo (4)

4 SeqNo (5)

4 SeqNo (6)

1 0x 00

The trace-collection setup is depicted in Fig. 4 and consists of a Crossbow MPR2400 MICAz mote [11] transmitter and another three MICAz motes mounted on Crossbow MIB600 Ethernet gateways [10] as receivers. The three receivers R1, R2 and R3 are connected to a host PC running three instances of Xlisten (a data logging application), one for each receiver. The link between transmitter and receiver was non-line-of-sight, with a wall, a door and several furniture items in the direct line between them. The receivers were separated by a distance of 0.25m. The transmitter was congured to transmit at 0 dBm. This way a data collection session produces three traces. All traces were collected while operating in channel 26 in the 2.480 GHz band. The reason for choosing channel 26 was that it is least prone to interference from any 802.11b/g frequency channels. Our own experience shows that selecting channel 26 does not completely eliminate interference from co-located 802.11b/g WLANs, but reduces it signicantly. 5.1.1 Packet payload TinyOS [18] is one of the most widely used open source operating system in WSN devices. TinyOS v1.1 allows various packet formats to be transmitted. We suitably modied code to enable the standard 802.15.4 frame format which TinyOS v1.1 labels CC2420 Frame Format (after the Chipcon CC2420 chipset [23] used in MICAz devices). Strictly speaking, the term packet refers to the Protocol Data Unit (PDU) exchanged between network layers of the transmitter and receiver while the term frame is used for PDUs exchanged between MAC layers. However, since our analysis is restricted to the MAC layer

0x8 Src Dst SeqNo 401 Adr Adr (1)

Fig. 5 CC2420 MAC frame format used for experiments

there is little cause for confusion and we use these terms interchangeably to refer to MAC layer PDUs. The exact MAC frame format used is shown in Fig. 5. The size of the frame is 41 bytes and comprises of a 1 byte Length Field, 2 byte Frame Control Field (FCF), 1 byte Sequence Number, 2 byte Destination PAN ID, 2 byte Destination Address, 1 byte Type eld, 1 byte Group eld, 29 bytes of data followed by a 2 byte Frame Check Sequence (FCS) containing a CRC. The contents of the payload eld are of our own choosing and consist of 3 unused bytes, the Source Address, the Destination Address and 6 copies of a 32 bit sequence number. The sequence number in the payload is used to keep track of lost packets. If the sequence number between two consecutively received packets skips one or more numbers that is indicative of a packet loss. The sequence number eld alone proves too small for this task in the face of long fades. Note that transmitted packets differ only in the 1 byte sequence number in the header and the six 32 bit sequence numbers in the payload, and the CRC. For a particular trace all remaining bits remain unchanged. However, since the wireless channel will introduce bit errors the copies of the sequence number used to track packet losses in the received packet may differ. For this purpose we use a majority vote of the received sequence numbers to reconstruct the transmitted sequence number and from it the entire packet. 5.1.2 Trace generation Bit-level error traces can be generated by comparing a transmitted packet with its received version. A simple bitwise XOR operation of the transmitted and received packets yields a bit pattern in which a zero (0) signies a bit that is received without error while a one (1) represents an inverted bit. We observe that in some cases the length of the received packet is shorter than the transmitted packets. This constitutes a partial loss and we use the term partially lost packets to refer to such packets. Partially erased packets are logged when bits in the MAC headers Length Field are inverted and the receiver stops listening to the wireless channel prematurely. It has also been observed that if bits in

Ch

MICAz Mote Ethernet Gateway Receiver 1 MICAz Mote Ethernet Gateway Receiver 2 Host PC MICAz Mote Ethernet Gateway Receiver 3

an

ne

l1

Channel 2
n an Ch el 3

MICAz Mote Transmitter

IEEE 802.15.4 Channel 26 (2.480 GHz)

Fig. 4 Equipment setup for trace collection

123

Wireless Netw (2011) 17:11731189

1181

pB()

the Length Field are inverted in such a way that the length of the incoming packet appears longer than actual the length of the logged packet still equals that of the transmission. Although the Length Field in the received packet may falsely indicate a longer packet, the absence of a carrier signal allows the receiver to detect the end of transmission. 5.2 Channel state information Each received packets logged entry is accompanied with three pieces of packet level CSI parameters. The rst is the FCS status of the packet modeled by random variable U with the nth packets FCS status is represented by /[n]. Ordinarily receivers only distinguish between two states, i.e. FCS Pass (denoted / = 0) if the CRC value in the FCS eld matches the CRC of the received packet, and FCS Fail (denoted / = 0) if it does not. Since we have knowledge of packet erasures and size of transmitted packets we extend the denition of FCS status to accommodate the reason for failure. We restrict the denition of FCS Fail BE (denoted / = 1) to mean that the size of a received packet matches the size of the transmitted packet and the CRC failure is due to Bit Errors (BE). Furthermore we classify a packet as being FCS Fail PL (denoted / = 2) and FCS Fail CL (denoted / = 3), where PL and CL are abbreviations for Partial Loss and Complete Loss respectively. Packets that are partially lost cannot pass the CRC test and are marked FCS Fail PL. Packets that are not received at all, i.e. when the decoded Sequence Number at receiver skips, are marked FCS Fail CL. Among other CSI there are RSSI and LQI which we described in earlier sections. Completely lost packets, with / = 3, are assigned q = -128, k = 0, and b = 1. Thus each received packet is characterized by its FCS Status, LQI, RSSI and BER processes. 5.3 Implementation results Using the above detailed setup we collected . The particular trace used to demonstrate proof of concept of gPMSS consists of 891,070 data packets collected from 7:12:42 p.m. on November 21, 2007 to 7:23:02 p.m. on November 22, 2007 in the Engineering Building at Michigan State University. This particular data set was collected in an ofce environment. The gPMSS cluster consisted of three receivers, also Crossbow MICAz motes mounted on MIB600 Ethernet gateways. Figure 6 is a cropped portion of the PDF of BERs observed in packets at gPMSS receivers R1, R2 and R3 that excludes b = 0 for enhanced visibility. Figure 7 depicts the PDF of the LQI of all received packets at R1, R2 and R3. Figure 8 depicts the PDF of their RSSI. These three gures clearly show that all three receivers experience different channel conditions.

0.08
R1

0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.02 0.04 0.06 0.08 0.1 0.12

R2 R3

0.14

BER
Fig. 6 PDF of BER experienced by receivers R1, R2 and R3 (pB (b = 0) is cropped out for better view of non-zero range

R1

0.08

R2 R3

0.06

p ()

0.04 0.02 0

40

60

80

100

LQI
Fig. 7 PDF of LQI experienced by receivers R1, R2 and R3

5.3.1 PER and PLR analysis We dene two quantities based on the FCS status, the packet error rate (PER) and the packet loss rate (PLR); PER PLR # of rcvd packets with / 1; 2 ; # of transmitted packets # of rcvd packets with / 3 ; # of transmitted packets 7 8

The packet reception rate (PRR) as PRR = 1 - (PER ? PLR). In Fig. 9 the rst three entries on the horizontal axis plot the PER, PLR and the sum of the two, PER?PLR, for R1, R2 and R3. For individual receivers PER?PLR happens to be approximately 7, 17 and 12%. These gures are followed by plots of these same quantities for the three diversity combining techniques. The simplest technique, selection diversity, appears to track the PER?PLR of the best performing receiver, in this case R1. Equal gain and maximal ratio diversity combining both perform better than

123

1182
60,000

Wireless Netw (2011) 17:11731189

0.6 0.5 0.4

R1 R2 R3

50,000

# of Packets

40,000 30,000 20,000 10,000 0

R1 R2 R3 SIMOSelection Div. SIMOEqual Gain Div. SIMOMaximal Ratio Div.

pP()

0.3 0.2 0.1 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

BER
95 90 85 80

RSSI (dBm)
Fig. 8 PDF of RSSI experienced by receivers R1, R2 and R3

Fig. 10 Histogram of BERs observed by receivers R1, R2 and R3 without gPMSS diversity combining and with selection, equal gain, and maximal ratio diversity combining

0.2
PER PLR PER + PLR

0.15

0.1

0.05

0 R1

R2

R3

Select

Equ Gain

Maxratio

Receiver
Fig. 9 PER, PLR and PER?PLR experienced by receivers R1, R2 and R3 without gPMSS diversity combining and with selection, equal gain, and maximal ratio diversity combining

exhibited by diversity combining methods are the same across all traces. Figure 10 shows that the histogram of the selection diversity combining closely matches that of the best receiving individual receiver, i.e. R1. The close match of the histogram of selection diversity with that of R1 shows it manages to bring a gPMSS BER performance up to that of the best receiving node. Thus, the BER model that is at the heart of this diversity combining technique delivers good performance. The result of equal gain and maximal gain diversity combining are even better. For every BER bin in the histogram, both equal gain and maximal-ratio combining are able to reduce the number of corrupt packets. Both are very close in their performance, but equal gain is consistently beating maximal-ratio combining across all BER bins in Fig. 10, and is also able to maintain this performance across different trace sets.

6 gPMSS protocol implementation This section describes our implementation of the gPMSS protocol for motes and analyzes its performance. For the mote platform, we selected the Crossbows Imote2 with the pre-installed .NET Micro Framework edition [12]. Using this edition of the Imote2 enabled us to implement gPMSS in the C# programming language which simplied and accelerated development. At this point we would like to clarify that although the Imote2 used for the actual implementation in this section is different from the MICAz we used for trace collection in Sect. 5, both use the same Chipcon CC2420 radio transceiver [23] which makes them equivalent for the purpose at hand. As the description of the gPMSS protocol above showed, in a situation when a transmission is received correctly by at least one recipient, gPMSS implements selection diversity described in Sect. 3.1. But when a transmission is received with errors

any individual receiver and selection diversity. This was to be expected. Recall that selection diversity merely tries to pick out the least corrupted version among a set, whereas equal gain and maximal ratio actually attempt to correct errors in received messages by un-weighted and weighted voting, respectively. This is adequately reected in the plot of PER, PLR and PER?PLRs. Both are able to reduce the PER. 5.3.2 BER analysis In Fig. 10 we plot the histogram (not PDF) of packets with non-zero BER as experienced by individual receivers R1, R2 and R3 without any diversity combining, as well as with different diversity combining methods. Again, the trends

123

Wireless Netw (2011) 17:11731189

1183

by all receivers, gPMSS either implements the functionality of an equal gain diversity or maximal ratio diversity combiner. We have implemented both in C# for Imote2. Using the Imote2, we conducted three experiments to collect performance data that would enable us to evaluate various gPMSS variants (selection diversity, selection diversity with diversity combining). The three experiments were conducted on August 16, 17 and 18, 2008 in a residential environment on the campus of Michigan State University. Receivers were arranged in a linear array with inter-receiver separation of 0.25m. Each experiment spanned a period of approximately 7 hours. The transmitter was placed at a distance of 8 m outside of the line-of-sight of the receivers. The maximal ratio diversity combiner depends on the CSI-driven BER model by Ilyas and Radha [16]. Since the BER model takes as input an LQI, RSSI pair k, q we still need to map it to a probability value. In the rst instance we nd the 90th percentile value of the BERs predicted PDF, i.e. the BER for which the value of the cumulative distribution function (CDF) is 0.9. In the second instance we map PDFs of the BER to their corresponding 50th percentile. We analyze the performance of the gPMSS protocol in a setting with one transmitter and N = 3 receivers. The receivers run a complete implementation of the gPMSS protocol described in Sect. 4. For the experiment the timeout constants were set to T0 = 2 sec, T1 = 10 sec, T2 = 12 sec and TT = 30 sec. We deliberately chose large values for T1, T2 and TT to avoid synchronization issues and justify them by the low-rate nature of target applications for IEEE 802.15.4. For the time being we have not attempted to optimize them to maximize throughput while still avoiding synchronization problems. The experiment was conducted at a residence with moderate Wi-Fi network interference.

experiment using a variant of gPMSS specied in the rst column. The results presented here are for three variants, (a) Maximal-ratio combining using b90% for the BER point estimate, (b) Maximal-ratio combining using the b50% for the BER point estimate, and (c) equal gain combining. To make sense of the packet counts in Table 1 and quantitatively assess the benets of using only selection diversity, and using selection diversity in conjunction with maximalratio/ equal gain combining we look at PRRs, denoted by h. Columns (1), (2) and (3) in Table 2 contain the PRRs of the baseline conguration in which receivers R1, R2 and R3 do not cooperate. Column (4) contains the PRR when gPMSS is used with the diversity combination method in column (0). Some of the packets received using gPMSS will have been received as a result of selection diversity, while others will have been recovered as a result of diversity combining. The following columns separate the gain in PRR over that in the baseline conguration by providing the additive increase in PRRs of individual receivers. Columns (5), (6) and (7) are additive contributions of selection diversity in hgPMSS to the PRRs of individual receivers. Thus, DhSD;R1 ; DhSD;R2 and DhSD;R3 are the increments in the PRR with respect to their respective baseline performances hR1, hR2 and hR3 in non-cooperating mode. Finally, column (8) is the additive contribution of diversity combining DhDC to the PRR hgPMSS of the system with gPMSS. Thus, since the PRR gains in columns (5), (6), (7) and (8) are all additive the relationship between the terms in Table 2 is, hgPMSS hR1 DhSD;R1 DhDC hR2 DhSD;R2 DhDC hR3 DhSD;R3 DhDC : 7.2 Channel capacity In this section we compute the channel capacity of using a single hop to R1, R2 or R3, use selection diversity as well as the full implementation of gPMSS. Capacity is denoted by K and computed as, Capacity K # of information bytes transferred : # of bytes transmitted by all nodes 10 9

7 Results and analysis This section analyzes and compares PRR, energy per packet and effect of retransmission limits on packet delivery rate with and without gPMSS. 7.1 Packet reception rate We denote the total number of transmissions made from transmitter T by CT, and the number of retransmissions among them by CR. Similarly, the number of transmitted packets that are received at R1, R2 and R3 without errors are denoted by C1, C2 and C3, respectively. Finally, CS denotes the number of packets for which diversity combining was attempted and succeeded, and CF the number of packets for which it failed. All these values are tabulated in Table 1. Each row in the table corresponds to a trial

This way the channel capacities KR1, KR2 and KR3 observed when R1, R2 and R3 receive can be computed from the packet counts in Table 1 as,

Table 1 Packet counts Div comb CT CR 855 879 C1 937 819 C2 893 844 C3 CS CF

Exp1: max-ratio b90% 3,170 Exp3: equal gain 3,683

957 597 0 825 497 0

Exp2: max-ratio b50% 4,167 1,039 1,254 1,322 1,198 739 0

123

1184

Wireless Netw (2011) 17:11731189

Table 2 PRR of individual nodes without gPMSS, PRR with gPMSS protocol, PRR gain for individual receivers R1, R2 and R3 due to selection diversity, and the PRR gain due to diversity combining (0) Exp1: max-ratio b90% Exp2: max-ratio b50% Exp3: equal gain (1) hR1 0.29 0.30 0.22 (2) hR2 0.28 0.31 0.23 (3) hR3 0.30 0.29 0.22 (4) hgPMSS 0.73 0.75 0.76 (5) DhSD;R1 0.25 0.27 0.40 (6) DhSD;R2 0.26 0.26 0.39 (7) DhSD;R3 0.24 0.28 0.40 (8) DhDC 0.19 0.18 0.13

Table 3 Channel capacity Div comb Exp1: max-ratio b90% Exp2: max-ratio b50% Exp3: equal gain KR1 0.285 0.290 0.217 KR2 0.272 0.305 0.223 KR3 0.291 0.278 0.218 KSD 0.508 0.536 0.582 KgPMSS 0.483 0.504 0.546

7.3 Energy per packet In this section we compute separately the energy expended by the transmitter T as well as the receiver cluster R1, R2 and R3 per error free packet communicated to any one receiver. We begin by computing the power consumed in transmitting and receiving data packets (DAT) and acknowledgement packets (ACK). Most of the additional power consumption during transmission/reception operations in an Imote2 occurs in the TI Chipcon CC2420 RF transceiver. The bulk of the remaining power consumption occurs in the Intel PXA271 XScale processor. According to measurements performed by Barton-Sweeney at Yale Universitys ENALAB [3], the power management IC (PMIC) on the Imote2 operates at approximately 90% efciency, supplying on-board devices (XScale processor, CC2420 RF transceiver) approximately 4.0 V. When the processor operates at 104 MHz, the total current drawn by the Imote2 is reported to be 68.70 mA when the radio is active, and 48.10 mA when it is idle. The difference of 20.60 mA is the current drawn by the CC2420 RF transceiver when it is transmitting/receiving. The Imote2 datasheet reports that the current drawn by the Imote2 with processor running at 104 MHz and active radio to be 66 mA, which is in close agreement with the measured value [12]. Furthermore, the CC2420 RF transceivers datasheet states that current drawn during transmission is 17.40 mA and during reception is 19.70 mA [23]. These two values are very close to each other and are also in close agreement with the measured value of 20.60 mA. Since the Intel XScale processor is not put into any low-power mode at any time its power consumption remains constant. The variations in power consumption due to gPMSS are due to variations in power consumption by the CC2420 RF transceiver produced by transmit/receive operations. Although Barton-Sweeneys measurements do not distinguish between transmit and receive operations of the RF transceiver, they are made in the conguration it is used by the Imote2, whereas the numbers provided in the CC2420 datasheet are for a wide range of supply voltages. For this reason, after verifying Barton-Sweeneys reported measurements with [12] and [23], we rely on them for the remainder of the paper. Thus, the Imote2 consumes 274.80

C1 LDAT ; CTLDAT C1LACK C2 LDAT 11 KR2 ; CTLDAT C2LACK C3 LDAT KR3 : CTLDAT C3LACK Here, LDAT and LACK denote the lengths (in bytes) of data (51 bytes) and acknowledgement packets (5 bytes). When selection diversity is used the channel capacity KSD is computed as, KR1 KSD CT CR CS LDAT : CTLDAT CT CR CS LACK 12

When selection diversity is used in conjunction with diversity combining the channel capacity KgPMSS is computed as, KgPMSS CT CR LDAT : CT CR LACK CS2LDAT 2LACK 13 Table 3 displays the channel capacities for 1-hop, selection diversity and gPMSS. Clearly, channel capacity is signicantly higher than the 1-hop communication congurations for both selection diversity and gPMSS. The channel capacities can be better evaluated by plotting each against the corresponding PRR in Table 2. This is shown in Fig. 13. This gure shows the tradeoff that comes with using selection diversity and gPMSS. The cluster of data points produced by selection diversity increases both channel capacity as well as PRR. Adding further complexity and using gPMSS increases PRR further, but at the cost of a slight drop in channel capacity. Data points for the same mechanisms (SISO, SD, gPMSS) are clustered together, demonstrating consistency across experiments.

123

Wireless Netw (2011) 17:11731189

1185
Columns (1a) and (ab) correspond to the baseline scenario using only retransmissions. Columns (2a) and (2b) are for the scenario where the single hop link from T to R1 is replaced by a 2 hop link, i.e. from T to T0 to R1. Columns (3a) and (3b) correspond to the case when only selection diversity is used by receivers. Columns (4a) and (4b) corresponds to the case where a full implementation of gPMSS is used that employs diversity combining (equal gain or maximal-ratio) in addition with selection diversity

1.369 EDAT ? EACK = 537.55 lJ

EDAT

274:8 103 W 418 bits 250 103 bits=s 360:54 lJ

(4a) PT

14

1.332 EDAT ? EACK = 524.21 lJ 5.232 EDAT ? 3 EACK = 2018.26 lJ 1.744 EDAT ? EACK = 672.75 lJ 6 EDAT ? 6 EACK = 2427.07 lJ 9.969 EDAT ? 3 4 EDAT ? 4 EACK = 3726.13 lJ EACK = 1618.04 lJ 3.323 EDAT ? EACK = 1242.04 lJ Exp2: maxratio b50%

Similarly, the energy expended in transmitting or receiving a 5 byte ACK packet is, 15
(3b) PR

EACK

Table 4 Energy consumed by transmissions at transmitter and receiver side per error-free received packet

PT PR

ET : # of packets recvd wo errors ER : # of packets recvd wo errors

16 17

Thus, PT and PR are energy consumption rates of transmitter and receivers obtained by normalizing by number of successfully delivered packets. The number of successfully delivered packets is R1 for 1 and 2 hop SISO, CT - CR - CS for selection diversity, and CT - CR for diversity combining. Table 4 lists PT, the per decodable packet energy at the transmitter, and PR, the per decodable packet energy at all receivers (R1, R2 and R3) combined for all three experiments (listed in column (0)) . In normal operating mode, RF transceivers receive all packets transmitted by nodes within communication and interference range. Motes inspect the MAC address in received packet headers to match its own. If it is determined that it is the intended recipient the packet is passed

(1b) PR

10.149 EDAT ? 3 4 EDAT ? 4 EACK = 3791.03 lJ EACK = 1618.04 lJ

(1a) PT

3.383 EDAT ? EACK = 1263.68 lJ

(0) Div comb

Exp1: maxratio b90%

Exp3: equal gain

4.497 EDAT ? EACK = 1665.32 lJ

13.491 EDAT ? 3 4 EDAT ? 4 EACK = 4995.96 lJ EACK = 1618.04 lJ

More generally, the per bit energy consumed by the RF transceiver is 1.099 lJ/b. Then the energy ET spent by the transmitter T to transmit CT data packets during the course of an experiment is CT 9 EDAT, and the energy expended to acknowledge C1 acknowledgements from R1 is C1 9 EACK. The energy spent by receivers R1, R2 and R3 in acknowledging these are ER1 = C1 9 EACK, ER2 = C2 9 EACK and ER3 = C3 9 EACK. Note that although energy is consumed by motes in tasks other than radio transmissions, the power consumed by computations is orders of magnitude less. Since the gPMSS protocol has computational complexity of O(N). We compute the energy per packet consumed at the transmitter PT and the sum of energy consumed by all receivers together PR as,

(3a) PT

1.845 EDAT ? EACK = 709.17 lJ

(2a) PT

(2b) PR

6 EDAT ? 6 EACK = 2427.06 lJ

6 EDAT ? 6 EACK = 2427.04 lJ

1.596 EDAT ? EACK = 619.39 lJ

274:8 103 W 58 bits 250 103 bits=s 43:97 lJ

5.535 EDAT ? 3 EACK = 2127.50 lJ

4.788 EDAT ? 3 EACK = 1858.18 lJ

1.314 EDAT ? EACK = 517.72 lJ

mW power when the RF transceiver is active (and 192.40 mW when it is inactive). We denote the energy consumed by the RF transceiver in transmitting/receiving a single DAT packet by EDAT. Similarly, the energy consumed in transmitting/receiving a single ACK packet is denoted by EACK. Although IEEE 802.15.4 supports multiple data rates, in the Imote2 it is xed at the maximum 250 kb/s. That means, to transmit or receive a data packet that is 41 bytes long, the RF transceiver expends approximately,

3.774 EDAT ? 4.548 EACK = 1560.65 lJ

(4b) PR

3.708 EDAT ? 4.416 EACK = 1531.05 lJ

3.531 EDAT ? 4.062 EACK = 1451.67 lJ

123

1186
20 Exp1: w/o gPMSS
Exp1: MaxRatio
90%

Wireless Netw (2011) 17:11731189

15

Exp2: w/o gPMSS Exp2: MaxRatio

50%

10

Exp3: w/o gPMSS Exp3: Equal Gain

20

40

60

80

100

g (%)
Fig. 11 Maximum number of transmission attempts m versus delivery guarantee g(%)

on to higher layers. Otherwise it is discarded. Therefore, unless otherwise noted, RF transceivers receive all transmissions within communication range. Thus, they expend energy to receive a packet, even when they are not the intended recipient. 1-hop SISO Columns (1a) and (1b) in the table correspond to the baseline case when gPMSS is not used and packets received by R1 are retransmitted. Lossless 2-hop SISO Columns (2a) and (2b) assume a 2 hop link from T to R1 with an intermediate node acting as a relay. This scenario is an alternative basis for comparison of gPMSS. It is assumed that both links (from T to relay, and from relay to R1) are perfect, i.e. no retransmissions are needed. Obviously, as the number of hops on the multihop path used to replace a gPMSS link increases so does the consumed energy. Energy consumed by the T and the relay node are lumped together into PT.

Selection diversity Columns (3a) and (3b) correspond to the case when only selection diversity is used by cooperating receivers. Diversity combining ? selection diversity Columns (4a) and (4b) corresponds to the case where a full implementation of gPMSS is used that employs diversity combination (equal gain or maximal-ratio) in addition to selection diversity. To keep the relationship general the tabulated values are in terms of EACK and EDAT. Figure 12 plots PT and PR (in Joules) expended in experiments 1, 2 and 3 when using maximal-ratio combining with b90%, maximal-ratio combining with b50% and equal gain combining, respectively. As in Table 4 we also evaluate energy for the cases when 1-hop SISO, 2-hop SISO and only selection diversity were used. The ordering of transmitter power consumption rate PT and receiver power consumption rates PR remains of schemes remains mostly the same across experiments across experiments and gPMSS variants. However, there is signicant variation in PT and PR when gPMSS is not used versus selection diversity versus gPMSS. For all three experiments PT is highest when gPMSS is not used while the corresponding receiver power consumption rate PR is lowest. Opting to use selection diversity alone signicantly reduces PT for maximal ratio gain variants (Exp 1 and 2) by about 42% and about 64% for equal gain variant (Exp 3). PR remains unchanged. Note from the previous section that this is accompanied by a 25% (for Exp 1 and 2) and 40% (for Exp 3) increase in PRR. Thus selection diversity is able to provide signicant power savings while increasing PRR at the same time. When gPMSS is employed PT is reduced by about 58% (for Exp 1 and 2) and 68% (for Exp 3) over the baseline conguration not using gPMSS. However, this is accompanied by an increase of approximately the same amount of energy on the receiver side. Thus, it appears that

Fig. 12 The energy in lJ consumed by transmitter and receivers per successfully delivered packet

123

Wireless Netw (2011) 17:11731189

1187

Fig. 13 Plot of channel capacity K vs. PRR h for all three experiments when R1, R2 and R3 are receivers at the end of a single hop link, when selection diversity is used, and when gPMSS is used

gPMSS shifts some of the power consumption from the transmitter side to the receiver side. 7.4 Packet transmission attempts The number of times the IEEE 802.15.4 MAC will retry transmitting a packet is controlled by the maxMaxFrameRetries attribute whose default value is set to 3 but can be varied from 0 to 7 (refer to IEEE 80215.4 standard [1]). This limit on the number of transmission attempts m for a packet limits the maximum PRR that can be guaranteed to g. Conversely, we may ask what is maximum number of transmission attempts m that the MAC must be allowed in order to ensure that at least g% of packets are received without errors? Figure 11 plots m against g for all three experiments. Clearly, to achieve any delivery guarantee g%, fewer transmissions are required with gPMSS, regardless of whether maximal-ratio or equal gain diversity combination is used, compared to the case where gPMSS is not enabled. For example, Fig. 11 shows that to achieve a 95% delivery guarantee we have to allow 9, 9, 13 transmission attempts for the channel conditions observed in experiments 1, 2 and 3. Using gPMSS, however, the maximum number of transmission attempts required to achieve the same delivery guarantee g = 95% are 3, 3, and 3, respectively. Clearly, the values of m required to achieve g = 95% without gPMSS exceeds IEEE 802.15.4s capabilities. From the plot in Fig. 11 we see that at IEEE 802.15.4s default value of m = 4 the maximum achievable delivery guarantee for the three experiments lies in the range 6575%.

without requiring any changes to mote hardware or the IEEE 802.15.4 LR-WPAN standard. We described three principal mechanisms enabled by gPMSS, namely selection diversity, equal gain and maximal-ratio gain diversity combination. We provide proof-of-concept and demonstrate gPMSS efcacy by applying these diversity combining techniques on bit error traces collected from a network of IEEE 802.15.4 motes. We demonstrate gPMSS by implementing it on the Intel Imote2 sensor mote running the .NET Micro framework. We analyze the performance of gPMSS in terms of PRR, retransmission attempts and power consumption per delivered packet. We saw that gPMSS raises the PRR from 2230% to 7376%, a relative increase of 150245%. Since gPMSS is a protocol-based solution it implies a messaging overhead. We observe that power consumption by the transmitter per correctly delivered packet is reduced up to 68%. We evaluated the effect of retry limit imposed by the IEEE 802.15.4 standard of the on the packet delivery rate that can be achieved. At the default retry limit of 3 (m = 4), gPMSS can achieve delivery rates of greater than 99%, against only 6575% when gPMSS is not used. Thus we demonstrate that gPMSS is capable of raising PRR, making use of highly lossy links feasible, thus reducing the number of required retransmission attempts and reducing the energy consumption rate of the transmitter per packet delivered. gPMSS has direct application in the design of small-world topologies in wireless networks to reduce the characteristic path length and diameter of networks which facilitates service discovery and the routing of high priority data in a network. This has the advantage of not needing any additional hardware [25, 26], or adding wired connections [9, 22]. The extension of the effective communication range also has applications in extending the lifetime of nodes surrounding the base station in wireless sensor networks subject to the funneling effect. The larger communication range allows more nodes to communicate with the base station directly and reduces the trafc load from nodes positioned closer to the base station. More generally, gPMSS can be used to connect weakly connected components of a network by adding more links between nodes farther apart.
Acknowledgments This work was supported in part by NSF Award CNS-0721550, NSF Award CCF-0728996, NSF Award CCF0515253, and an unrestricted gift from Microsoft Research.

References
1. ANSI/IEEE. (2006). Ansi/ieee std 802, part 15.4: Low rate wireless personal area networks. Technical report, ANSI/IEEE. 2. Bahl, P., Adya, A., Padhye, J., & Walman, A. (2004). Reconsidering wireless systems with multiple radios. ACM SIGCOMM Computer Communication Review, 34(5), 3946. 3. Barton-Sweeney, A. (2006). Power modes and energy consumption for the imote2 sensor node. Technical Report SOS

8 Conclusions We presented the gPMSS, a protocol-centric approach to enable receiver cooperation and diversity combining

123

1188 iMote2 Port, Embedded Networks and Applications Lab (ENALAB), Yale University, September. Bletsas, A., Khisti, A., & Win, M. Z. (2008). Opportunistic cooperative diversity with feedback and cheap radios. IEEE Transactions on Wireless Communications, 7(5), 1823. Bletsas, A., & Lippman, A. (2006). Implementing cooperative diversity antenna arrays with commodity hardware. IEEE Communications Magazine, 44(12), 33. Brennan, D. G. (2003). Linear diversity combining techniques. Proceedings of the IEEE, 91(2), 331356. Chakraborty, S. S., Yli-Juuti, E., & Liinaharja, M. (1998). An arq scheme with packet combining. IEEE Communications Letters, 2(7), 200202. Cheng, S., & Valenti, M. C. (2005). Macrodiversity packet combining for the ieee 802.11 a uplink. In Proceedings of the IEEE wireless communications and networking conference (WCNC05), Vol. 1. Chitradurga, R., & Helmy, A. (2004). Analysis of wired short cuts in wireless sensor networks. In IEEE/ACS international conference on pervasive services (ICPS04) (pp. 167177), July. Crossbow. (2006). MIB600CA, Ethernet Interface Board Datasheet. Crossbow Technology Inc., http://www.xbow.com/ Products/productdetails.aspx?sid=179, 6020-0055-04 rev a edition, Oct. Crossbow. (2006). MPR2400, ZigBee, 2.4 GHz, MICAz Mote Datasheet. Crossbow Technology Inc., http://www.xbow.com/ Products/productdetails.aspx?sid=164, 6020-0060-04 rev a edition, Oct. Crossbow Technology. (2007). Crossbow Imote2 Datasheet, 6020-0117-02 rev a edition, April. IEEE. (2003). Ieee 802.11b-1999, part 11: Wireless LAN: Higher speed physical layer (phy) extension in the 2.4 ghz band. Technical report, IEEE. Ilyas, M. U., Kim, M., & Radha, H. (2009). Reducing packet losses in networks of commodity ieee 802.15.4 sensor motes using cooperative communication and diversity combination. In Proceedings of the 28th IEEE international conference on computer communications (INFOCOM09), Rio de Janeiro, Brazil, April. Ilyas, M. U., & Radha, H. (2006). Increasing network lifetime of an ieee 802.15. 4 wireless sensor network by energy efcient routing. In Proceedings of IEEE international conference on communications. Ilyas, M. U., & Radha, H. (2008). Measurement based analysis and modeling of the error process in ieee 802.15.4 lr-wpans. In Proceedings of the 27th IEEE conference on computer communications (INFOCOM08). IEEE, IEEE, April. Ji, Z., Yang, Y., Zhou, J., Takai, M., & Bagrodia, R. (2004). Exploiting medium access diversity in rate adaptive wireless lans. In International conference on mobile computing and networking (pp. 345359). ACM, New York, NY, USA. Levis, P., Madden, S., Polastre, J., Szewczyk, R., Whitehouse, K., Woo, A., Gay, D., Hill, J., Welsh, M., & Brewer, E. (2005). TinyOS: An operating system for sensor networks. Springer Verlag, Berlin. Miu, A., Apostolopoulos, J. G., Tan, W., & Trott, M. (2003). Low-latency wireless video over 802.11 networks using path diversity. In Proceedings of the IEEE international conference on multimedia & Expo (Vol. 3, pp. 441444) July. Miu, A., Balakrishnan, H., & Koksal, C. E. (2005). Improving loss resilience with multi-radio diversity in wireless networks. In Proceedings of the the 11th annual international conference on mobile computing and networking (MobiCom) (pp. 1630). ACM, AugSep.

Wireless Netw (2011) 17:11731189 21. Miu, A., Tan, G., Balakrishnan, H., & Apostolopoulos, J. (2004). Divert: ne-grained path selection for wireless lans. In Proceedings of the 2nd international conference on mobile systems, applications, and services (pp. 203216). ACM, New York, NY, USA. 22. Sharma, G., & Mazumdar, R. (2005). Hybrid sensor networks: a small world. In ACM international symposium on mobile Ad Hoc networking and computing (pp. 366377). ACM, New York, NY, USA. 23. TI and Chipcon (2004). Chipcon AS SmartRF CC2420 Preliminary Datasheet (rev 1.2), 2004-06-09. TI Chipcon, 1.2 edition, June. 24. Valenti, M. C. (2003). Improving uplink performance by macrodiversity combining packets from adjacent access points. In Proceedings of the IEEE wireless communications and networking conference (WCNC03) (Vol. 1). 25. Wan, C. Y., Campbell, A. T., & Crowcroft, J. (2004). A case for all-wireless, dual-radio virtual sinks. In International conference on embedded networked sensor systems (pp. 267268). ACM Press, New York, NY, USA, July. 26. Wan, C. Y., Eisenman, S. B., Campbell, A. T., & Crowcroft, J. (2005). Siphon: overload trafc management using multi-radio virtual sinks in sensor networks. In International conference on embedded networked sensor systems (pp. 116129). ACM Press, New York, NY, USA. 27. Watts, D. J. (1999). Small worlds: The dynamics of networks between order and randomness. Princeton: Princeton University Press. 28. Woo, G. R., Kheradpour, P., Shen, D., & Katabi, D. (2007). Beyond the bits: Cooperative packet recovery using physical layer information. In Proceedings of the 13th annual ACM international conference on mobile computing and networking (MobiCom) (pp. 147158). ACM Press, New York, NY, USA, September.

4.

5.

6. 7.

8.

9.

10.

11.

12. 13.

14.

Author Biographies
Muhammad U. Ilyas received his Ph.D. and MS degrees in Electrical Engineering from Michigan State University (MSU), East Lansing, MI in 2009 and 2007, the MS degree in Computer Engineering from the Lahore University of Management Sciences (LUMS), Lahore, Pakistan in 2004, and the BE (Honors) degree in Electrical Engineering from the National University of Sciences & Technology (NUST), Rawalpindi, Pakistan in 1999. He is currently Associate Professor in the Department of Electrical Engineering of the School of Electrical Engineering & Computer Science (SEECS) at the National University of Sciences & Technology (NUST), Islamabad, Pakistan. Prior to that he was a Post-doctoral Research Associate appointed jointly by the Department of Electrical & Computer Engineering (ECE) and the Department of Computer Science & Engineering (CSE) at MSU where he worked under the joint supervision of Prof. Hayder Radha (IEEE Fellow) and Prof. Alex X. Liu.

15.

16.

17.

18.

19.

20.

123

Wireless Netw (2011) 17:11731189 Moonseong Kim has received the M.S. degree in Mathematics from Sungkyunkwan University, Korea in August 2002, and the Ph.D. degree in Electrical and Computer Engineering from Sungkyunkwan University, Korea in February 2007. He was a research professor at Sungkyunkwan University in 2007 and a visiting research associate in Department of Computer Science and Engineering, Michigan State University, USA in 20082009. Currently, he is a patent examiner, Information and Communications Examination Bureau, Korean Intellectual Property Ofce, Korea. His research interests include wired/wireless networking, sensor networking, mobile computing, and simulations/numerical analysis. Hayder Radha received the B.S. degree (with honors) from Michigan State University (MSU), East Lansing, in 1984, the M.S. degree from Purdue University, West Lafayette, IN, in 1986, and the Ph.M. and Ph.D. degrees from Columbia University, New York, in 1991 and 1993, respectively (all in electrical engineering). He joined MSU in 2000 as Associate Professor in the Department of Electrical and Computer Engineering. From 1986 to 1996, he was with Bell Laboratories. From 1996 to 2000, he worked at Philips Research USA and became a Philips Research Fellow in

1189 2000. His research interests include wireless and multimedia communications and networking, stochastic modeling, and image and video coding and compression. He has more than 25 patents in these areas. He served as co-chair and editor of the ATM and LAN Video Coding Experts Group of the ITU-T in 19941996. Dr. Radha is a member of the IEEE Signal Processing Multimedia Technical Committee. He is a recipient of the Bell Labs Distinguished Member of Technical Staff Award (1993), the Withrow Distinguished Scholar Award (2003), and the Microsoft Research Content and Curriculum Award (2004).

123