Effect of data aggregation on the delay of demand-response communications within a Wi-Fi home area network with and without other traffic

Abstract

Home area networks play an important role in the demand-response function of smart grids. It is responsible for responding to requests received from the grid by controlling the devices in the home in a predefined manner. Communication within a Home Area Network should be efficient in terms of both delay and energy. Delay matters since the devices need to respond to the request within the stipulated delay. Energy matters since thousands of Home Area Networks are likely to create a significant energy footprint on the global level. In order to reduce energy consumption, the number of communications needs to be reduced and data aggregation can achieve this goal. However, data aggregation introduces a prolonged delay and may thus render the system unfit for its purpose. Therefore, it is required to determine the variation of delay when data aggregation is performed at different levels. This paper presents algorithms for data aggregation and device clustering optimization. Finally, the delay distribution was studied in a simulation environment with one level of data aggregation. The results show that an existing Wi-Fi network can be used for Smart Grid communications with in-network data aggregation provided that there is a spare (unused) bandwidth of 3 Mbit/s in the network.