多包接收无线传感器网络的跨层优化策略研究

Maximizing Throughput for Wireless Sensor Network With Multi-Packet Reception

多包接收能有效提高无线传感器网络的网络吞吐量,但由于该思想在物理层实现的复杂性,决定了在上层难以建立充分发挥多包接收性能的最优传输方案。本文根据基于信噪比的物理层数据发送模型,通过研究可同时传输数据的节点集合,建立了链路层的分时传输模型。为了使该模型可在多项式时间内求解,提出了最大可行集合的概念,减少了求解的时间和空间复杂度,最终提出基于多包接收的无线传感器网络跨层优化策略。仿真显示,采用该策略后数据传输速率比传统方案提高了1倍。

Multi-packet reception（MPR） technique is a promising approach to increase throughput for wireless networks.Due to its complex physical layer behavior,optimal solutions are yet to be determined.Based on the physical interference model,we analyze all constraints for concurrent transmissions under MPR and thus identify all feasible sets of concurrent transmissions.By scheduling these feasible sets into different time slots,the throughput maximization problem can be formulated as a linear program.However,the problem size is exponential due to the large number of feasible sets.We propose a concept of largest feasible set to decrease the problem size,which enables us to obtain a cross-layer optimal solution in polynomial-time.Comparing with optimal solutions without MPR,we show that throughput can be increased about 100% by using MPR in simulation.