Exploiting Interference for Efficient Distributed Computation in Cluster-based Wireless Sensor Networks

TL;DR

This paper introduces a novel clustering-based consensus algorithm for wireless sensor networks that leverages interference and broadcast properties to improve convergence speed and reduce energy consumption.

Contribution

It presents a new interference-exploiting consensus algorithm optimized via semidefinite programming for efficient distributed computation.

Findings

Enhanced convergence speeds in sensor networks

Reduced energy consumption through interference exploitation

Feasible offline optimization via semidefinite programming

Abstract

This invited paper presents some novel ideas on how to enhance the performance of consensus algorithms in distributed wireless sensor networks, when communication costs are considered. Of particular interest are consensus algorithms that exploit the broadcast property of the wireless channel to boost the performance in terms of convergence speeds. To this end, we propose a novel clustering based consensus algorithm that exploits interference for computation, while reducing the energy consumption in the network. The resulting optimization problem is a semidefinite program, which can be solved offline prior to system startup.