Wireless network performance for residential demand-side participation

Abstract

In this paper, we investigate the performance of wireless technology for monitoring and controlling the electrical load of a residential neighborhood. An event-based power metering scheme is assumed where transmissions are sent only when a change occurs in household consumption. Analysis of high-resolution, empirical data suggests that consumption transitions can be modeled as a Poisson process in the time-domain. A peer-to-peer 802.11 wireless network was simulated to determine latency between houses and a local sub-station. Probability distributions for monitoring and control traffic show that packet latency is exponential distributed and neighborhoods larger than 400 homes may not be able to respond within a fraction of the 60 Hz cycle.