Algorithms for Relaying via Channel Quantization in Finite Rate Feedback Limited Sensor Networks

Abstract

Relaying is often advocated for improving system performance by enhancing spatial diversity in wireless networks. In this paper, we address the issue of energy tradeoff made by relay nodes between transmitting their own data and forwarding other nodes' information in fading channels. First, assuming perfect channel state information (CSI), we propose a power control policy in a two-node relay network under which total energy consumption across both nodes is minimized while meeting both outage probability requirements. However perfect CSI at each node is not possible in general due to bandwidth limitations that prevent the full exchange of precise channel information. Thus in this paper, we develop power control algorithms for relaying under bandwidth constraints, via quantization. Specifically, we develop a quantization protocol that accounts for the asymmetric nature of uplink/downlink communication bandwidths and develop an optimal polynomial time algorithm for channel quantization using this protocol. The quantization algorithm minimizes the sum of expected transmission powers at the source and relay, required for collaborative relaying to satisfy the given outage probability.