Ajit Warrier

Z-MAC: a hybrid MAC for wireless sensor networks

This paper presents the design, implementation and performance evaluation of a hybrid MAC protocol, called Z-MAC, for wireless sensor networks that combines the strengths of TDMA and CSMA while offsetting their weaknesses. Like CSMA, Z-MAC achieves high channel utilization and low latency under low contention and like TDMA, achieves high channel utilization under high contention and reduces collision among two-hop neighbors at a low cost. A distinctive feature of Z-MAC is that its performance is robust to synchronization errors, slot assignment failures, and time-varying channel conditions; in the worst case, its performance always falls back to that of CSMA. Z-MAC is implemented in TinyOS.

DRAND: distributed randomized TDMA scheduling for wireless ad-hoc networks

This paper presents a distributed implementation of RAND, a randomized time slot scheduling algorithm, called DRAND. DRAND runs in O(delta) time and message complexity where delta is the maximum size of a two-hop neighborhood in a wireless network while message complexity remains O(delta), assuming that message delays can be bounded by an unknown constant. DRAND is the first fully distributed version of RAND. The algorithm is suitable for a wireless network where most nodes do not move, such as wireless mesh networks and wireless sensor networks. We implement the algorithm in TinyOS and demonstrate its performance in a real testbed of Mica2 nodes. The algorithm does not require any time synchronization and is shown to be effective in adapting to local topology changes without incurring global overhead in the scheduling. Because of these features, it can also be used even for other scheduling problems such as frequency or code scheduling (for FDMA or CDMA) or local identifier assignment for wireless networks where time synchronization is not enforced. We further evaluate the effect of the time-varying nature of wireless links on the conflict-free property of DRAND-assigned time slots. This experiment is conducted on a 55-node testbed consisting of the more recent MicaZ sensor nodes.

DRAND: Distributed Randomized TDMA Scheduling for Wireless Ad Hoc Networks

This paper presents a distributed implementation of RAND, a randomized time slot scheduling algorithm, called DRAND. DRAND runs in O(delta) time and message complexity where delta is the maximum size of a two-hop neighborhood in a wireless network while message complexity remains O(delta), assuming that message delays can be bounded by an unknown constant. DRAND is the first fully distributed version of RAND. The algorithm is suitable for a wireless network where most nodes do not move, such as wireless mesh networks and wireless sensor networks. We implement the algorithm in TinyOS and demonstrate its performance in a real testbed of Mica2 nodes. The algorithm does not require any time synchronization and is shown to be effective in adapting to local topology changes without incurring global overhead in the scheduling. Because of these features, it can also be used even for other scheduling problems such as frequency or code scheduling (for FDMA or CDMA) or local identifier assignment for wireless networks where time synchronization is not enforced. We further evaluate the effect of the time-varying nature of wireless links on the conflict-free property of DRAND-assigned time slots. This experiment is conducted on a 55-node testbed consisting of the more recent MicaZ sensor nodes.

How much energy saving does topology control offer for wireless sensor networks? - A practical study

Topology control is an important feature for energy saving, and many topology control protocols have been proposed. Yet, little work has been done on quantitatively measuring practical performance gains that topology control achieves in a real sensor network. This is because many existing protocols either are too complex or make too impractical assumptions for a practical implementation and analysis. A rule of thumb or a practical upper bound on the energy saving gains achievable by topology control would assist engineers in estimating the overall energy budget of a real sensor system. This paper proposes a new topology control protocol simple enough to permit a straightforward stochastic analysis and also a real implementation in Mica2. This protocol is currently deployed in our testbed network of 42 Mica2 nodes. Our contribution is not on the novelty of this protocol but on a practical performance bound we can study using this protocol. The stochastic analysis reveals that topology control can achieve a power gain proportional to network density divided by a factor of eight to ten. Our experiment result from the real testbed tests confirms this finding. We also find a tradeoff in terms of throughput loss due to reduced density by topology control which amounts to about 50% throughput loss. These performance figures represent rough rules of thumb on energy efficiency achievable even by a very simple, unoptimized protocol.

Cross-layer optimization made practical

Limited resources and time-varying nature of wireless ad hoc networks demand optimized use of resources across layers. Cross-layer optimization (CLO) for wireless networks, an approach that coordinates protocol behaviors at different layers with a goal to maximize a utility function, has received considerable attention lately. However, most existing work remains as theory and no practical CLO based on utility optimization exists today. The main difficulties in implementing theoretical CLO designs arise often from impractical assumptions about the characteristics of the wireless medium and also from computational and communication overhead of proposed solutions to achieve or approximate the optimality. In contrast, many existing practical approaches for CLO (not necessarily utility optimization) are rather ad hoc in nature and developed mostly based on intuitions. Thus, a clear gap between theory and practice in CLO exists. This paper addresses this dichotomy to close the gap by taking an optimal solution from utility-based CLO and applying practical approximation to enable a practical implementation in a wireless mesh network where nodes are statically positioned in an ad hoc fashion. We focus on the utility of maximizing throughput. We identify the impractical or computationally-intensive components of a theoretically-derived optimal throughput-maximizing solution and then propose, in most cases, practical approximation with O(1) complexity for MAC, scheduling, routing and congestion control. The result is a practical CLO solution that approximates the theoretically-derived optimal solution, but achieves much improved performance over existing practical CLO implementations.

DiffQ: Differential Backlog Congestion Control for Wireless Multi-hop Networks

In this demo, we showcase DiffQ - a congestion control protocol inspired by theoretical cross-layer optimization approaches. DiffQ can support congestion control for network flows that use either single-path or opportunistic multi-path routing. Our demo will focus on the performance in single-path routing environments, where contemporary end-point congestion control algorithms like TCP face severe unfairness or even starvation. This is primarily due to the interaction of such protocols with MAC layer unfairness. We demonstrate micro (5 flows) as well as macro-evaluations (60 flows) of such cases. Our demo is conducted on WiSeNet - a 70-node wireless mesh test-bed hosted in the computer science building at NCSU. Distributed over a 100,000 sq ft building, this is one of the largest test-bed installations both in terms of number of nodes and coverage area, hence an ideal testing ground for such scenarios. Experimental results like throughput, MAC-layer statistics, delay, routing path flaps and network buffer overflows are recorded and displayed in real-time and enable a birds eye-view of the entire network status and allow us to point out various phenomenon as they happen.

Performance of a Concurrent Link SDMA MAC Under Practical PHY Operating Conditions

Space division multiple access (SDMA)-based medium access control (MAC) protocols have been proposed to enable concurrent communications and improve link throughput in multi-input-multi-output (MIMO) ad hoc networks. For the most part, the works appearing in the literature make idealized and simplifying assumptions about the underlying physical layer and some aspects of the link adaptation protocol. The result is that the performance predicted by such works may not necessarily be a good predictor of the actual performance in a fully deployed system. In this paper, we look to introduce elements into the SDMA-MAC concept that would allow us to better predict their performance under realistic operating conditions. Using a generic SDMA MAC, we look at how the network sum throughput changes with the introduction of the following: (1) use of the more practical MMSE algorithm, instead of the zero-forcing or singular-value-decomposition-based nulling algorithms used for receive beamnulling; (2) impact of channel estimation errors; (3) introduction of link adaptation mechanism specifically designed for concurrent SDMA MACs; and (4) incorporation of TX beamforming along with RX beamnulling. Following on the transmission window during which concurrent transmissions are allowed by the MAC, we qualify the impact of each of these four elements in isolation. At the conclusion, the performance of a system that incorporates elements 1-4 is presented and compared against the baseline system, showing an improvement of up to five times in the overall network sum throughput.

Mitigating Starvation in Wireless Sensor Networks

Wireless sensor nodes deployed in hostile network conditions are required to report events such as enemy movements quickly and reliably. Unfortunately, current CSMA-based MAC protocols for wireless sensor networks suffer from varying degrees of starvation. Such starvation could lead to significant portions of the sensed area being invisible to the sink. It has been shown by prior research that such starvation is caused due to spatial bias and CSMA-based channel access. In this experimental work, we study how the use of B-MAC, the default CSMA-based MAC protocol can be the cause of starvation in a 35 node multi-hop wireless sensor testbed. We then show that the use of Z-MAC, a hybrid MAC, can alleviate such starvation, while at the same time giving high channel utilization

Adaptively switching between CSMA and SD-CSMA in multi-antenna based Ad Hoc Networks

Carrier Sense Multiple Access (CSMA) is the conventional medium access method used in wireless Ad Hoc networks. It can be enhanced as a Space-Division based CSMA (namely, SD-CSMA) by using multi-antenna techniques, which allows concurrent link communications in the network. In this paper, we present a MAC design that can adaptively switch between CSMA and SD-CSMA in multi-antenna based Ad Hoc Networks. We first investigate the CSMA mode and the SD-CSMA mode by considering several practical constraints from MIMO physical layer, which include imperfect channel estimation, modulation and coding scheme, MMSE detection, power constraint and packet length. We then propose a selection criterion that adaptively switches between CSMA and SD-CSMA based on current wireless channels among nodes. Simulation results verify that the proposed design can outperform both the CSMA mode and the SD-CSMA mode in terms of aggregate throughput

Experimental evaluation of MAC protocols for fairness and QoS support in wireless networks

Existing wireless MACs are known to have fairness and QoS problems 1. In response, the wireless community have come up with numerous ldquopoint solutionsrdquo, with one-to-one comparison of their solution with existing MACs for the specific fairness model they target. However, with some tradeoff of flexibility and efficiency, many of these MACs can be made relatively free of a a specific fairness model. This decoupling of the MAC from the fairness model is desirable because it allows objective testing of the MAC under different fairness models and also allows independent evolution of fairness policies and research into medium access issues. However, there is no cross-model evaluation of such MACs especially in the context of real multihop wireless testbeds. In this paper we address this issue by testing three MACs - 802.11e, EY-NPMA and DWOP, (which represent different ways of enabling fairness in the MAC) across a number of fairness and QoS models - temporal/rate and proportional fairness, and EDF and static priority scheduling. In the process we also extend EY-NPMA significantly (which we call Siren) for multihop wireless environments. We perform these comparisons on a 30 node multihop wireless sensor testbed comprising of MicaZ devices equipped with the CC2420 radio.