Clustered IEEE 802.15.4 sensor networks with data aggregation: energy consumption and probability of error

In this paper, we analyze an IEEE 802.15.4-compliant wireless sensor network, organized in clusters with fusion centers (FCs) acting as cluster heads. Sensors observe the same common binary phenomenon, decide on its status, and send their decisions to the FCs, which perform a majority fusion. FCs' decisions are then sent to an access point, where a final estimate is obtained. Data aggregation (DA) over consecutive observations can be also performed at sensors, with the aim of reducing the number of sensors simultaneously competing for the channel, while increasing the packet size. A novel mathematical framework for deriving the energy consumption in IEEE 802.15.4 networks is also proposed. The performance is evaluated in terms of energy consumption and probability of decision error, jointly accounting for decentralized detection, medium access, and DA.