Efficient matrix multiplication on wireless sensor networks

Abstract

Wireless Sensor Networks (WSN for short) have been widely used for data collecting and processing in various applications. This paper studies the problem of how to efficiently realize matrix multiplication on a WSN. Based on a widely accepted model of WSNs, we show the lower bound on the number of parallel steps needed by any algorithm for nxn matrix multiplication on a WSN of n2 sensors with transmission radius r. We then present two parallel algorithms for matrix multiplication on a WSN which have the same time complexity as the lower bound. For both algorithms we analyze their main performance parameters including time cost, memory cost and energy-efficiency in detail. © 2008 IEEE.