Distributed winner determination protocol for reverse auction-based task allocation in pervasive computing

Reverse Auction is one of the best-known market-based mechanisms for allocation of delay-constrained tasks to distributed sensor nodes in wireless sensor network as a typical application in pervasive computing. However, costly message exchanges with enormous overheads among resource-constrained sensor nodes for decision making are required. In this paper, a novel energy/delay-efficient distributed message exchange protocol for winner determination in reverse auction-based task allocation is proposed. The main objective of this protocol is the reduction in the overhead, delay and energy consumption in message exchanges for determining the winning bidder. Simulation results show that the proposed protocol can achieve superior performance compared with static and online adaptive scheduling schemes with centralized message exchanges.