A Framework for Edge Computing on Smart Meters

Abstract

Smart meters are connected through slow wireless or wired networks, e.g., multi-hop wireless networks. To reduce the number of hops in data transmission from multiple smart meters, we present a framework for enabling edge computing to be available in smart meters. The framework is designed for supporting MapReduce processing to calculate and analyze data, the amount of power usage, in smart meters by using mobile code and mobile agent technologies. It can dynamically deploy programs for processing data at smart meters as a map step and executes the programs with local data. Finally, it aggregates the results of the programs to certain smart meters as a reduce step.