Multihop wireless mesh networks can provide Internet access over a wide area with minimal infrastructure expenditure. In this work, we present a measurement driven deployment strategy and a data-driven model to study the impact of design and topology decisions on network-wide performance and cost. We perform extensive measurements in a twotier urban scenario to characterize the propagation environment and correlate received signal strength with application layer throughput. We find that well-known estimates for pathloss produce either heavily overprovisioned networks resulting in an order of magnitude increase in cost for high pathloss estimates or completely disconnected networks for low pathloss estimates. Modeling throughput with wireless interface manufacturer specifications similarly results in severely underprovisioned networks. Further, we measure competing, multihop flow traffic matrices to empirically define achievable throughputs of fully backlogged, rate limited, and web-emulated traffic. We find that while fully backlogged flows produce starving nodes, rate-controlling flows to a fixed value yields fairness and high aggregate throughput. Likewise, transmission gaps occurring in statistically multiplexed web traffic, even under high offered load, remove starvation and yield high performance. In comparison, we find that well-known noncompeting flow models for mesh networks over-estimate network-wide throughput by a factor of 2. Finally, our placement study shows that a regular grid topology achieves up to 50 percent greater throughput than random node placement.

In this work, we propose and analyze a two-way training system that exploits the fact that most wireless nodes are capable of both transmitting and receiving signals. That is, the underlying channel is by design a two-way channel, even if only one of the nodes has data to communicate. For half-duplex nodes with one antenna at the source node and M antennas at the destination node, we show that a novel training system can double the maximum diversity order of the system. The key departure from existing work is that (a) all resources used to obtain channel information at the source and destination are accounted for, and (b) channel estimates at the source and destination are mismatched and noisy. We further derive the full diversity-multiplexing tradeoff which demonstrates a diversity order gain at almost all multiplexing gains due to information at the source, even when that information is based on noisy estimates and subjected to full resource accounting.

No abstract

Abslrcf-The performance of 80211b wireless LANs is well understood for indoor environments. However, their behavior In outdoor and mobile environments has remained largely unexplored. We have performed experiments to bridge this knowledge gap by empirically measuring the performance of an 802.11h system in a broad spectrum of emulated environments. The goal of our work Is to contribute to the evolution of currently existing wireless standards by measuring the effects of different channel phenomena on 802.11h. Our results show that current Implementations of the 80Z.llb standard may not he well suited to use in a mobile environment, but that they collld conceivably be m&ed to have better performance in such situations.

Coherent signaling over a fading channel requires the receiver to have knowledge of the current fading state (CSIR). In order to proactively prevent outages, the transmitter requires knowledge of the fading state (CSIT) to perform power control. Typically, the channel state is estimated by using a training signal, and the channel state information contains some error. In this work, we examine the performance of a symmetric, single-input, multipleoutput (SIMO) channel in which both the transmitter and the receiver have independent, imperfect estimates of the channel. We give expressions for mutual information and outage probability, and we use those expressions to evaluate the diversity order of the system for a constant target rate. We show that diversity order is dominated by the transmitter error and that the diversity order is equal to that of a system with perfect CSIR. I. INTRODUCTION Knowledge of the channel state at the transmitter (CSIT) can be used to improve the performance of wireless systems operating over fading channels [1], [2], [3], [4], [5], [6], [7], [8].In particular, it is well known that CSIT can be used to avoid outages through power control. However, relatively little is known about the performance of power control in the presence of channel estimation error [8], [9]. While existing works have examined different aspects of channel estimation error at either the transmitter or the receiver, we address the relationship between errors at the transmitter and the receiver and compare their impacts on outage performance.In this work, we account for errors in channel state information at both the transmitter and the receiver as well as the resources used for training. Our results show that, though improving CSIR provides some gains, it is the quality of the CSIT that largely determines the diversity order of the system. Therefore, we conclude that securing adequate CSIT is far more important than obtaining CSIR and worthy of greater resource allocation.The remainder of the work is structured as follows. Section II defines our channel model and basic assumptions. Section III contains our principal contributions. We numerically evaluate the outage performance of our system in Section III-E. Section IV contains a summary of our results and ideas for future extensions.