Maximizing Precision for Energy-Efficient Data Aggregation in Wireless Sensor Networks with Lossy Links

Two main factors that impact the performance of data aggregation in wireless sensor networks (WSNs) are data quality and energy efficiency. This paper exploits the tradeoff between data quality and energy consumption to maximize the data aggregation precision under heterogeneous per-node energy constraints. Unlike previous work, we explicitly account for link loss in the optimization framework. To tackle link unreliability, we need to appropriately allocate the limited energy across the incoming and outgoing links of each individual node. We present a centralized algorithm based on the Immune-Genetic heuristic to find near-optimal energy allocation strategy such that the precision of the aggregated data received by the sink is maximized. The algorithmic complexity and implementation issues are also discussed. Furthermore, we develop a localized alternative algorithm based on the Gibbs sampler, which is more scalable and can adapt to large-scale distributed WSNs. Finally, we conduct numerical simulations to demonstrate the convergence as well as the data aggregation precision performance of the proposed algorithms.