Aggregator Modeling with Multiple Active/Sleep Modes for Wireless Sensor Networks

摘要

Data aggregation and active/sleep scheme are wildly used in energy-efficient wireless sensor networks. However, little research has been done on the optimal active/sleep scheme design for an aggregator node. In our previous work, we developed a stochastic model of aggregator node with active/sleep scheme for performance analysis and optimization. That model simulates the direct mode transfer from active to sleep or vice versa. In this paper, we propose a new stochastic analytical model for aggregator with multiple modes in the active period which reflect the wireless sensor's multiple energy saving modes. We derive Complete Aggregation Packet (CAP) rate, Total Aggregation Packet (TAP) rate, the Success Ratio which is defined as CAP rate divided by TAP rate, and Average Packet Delay (APD) to evaluate the aggregator performance. The extensive numerical simulations validate the accuracy of the stochastic analytical model. We would like to optimize the performance measures of APD and TAP rate by Lagrangian Optimization method which has not been finished yet.