QS-Sift:一种基于服务质量和空间相关性的无线传感器网络MAC协议

在大规模、稠密无线传感器网络中传输的数据大都具有高度的空间相关性．近年来，国内外学者提出通过在网络中筛选出一部分节点来发送数据以降低能量开销．本文在网络节点的筛选策略方面开展研究，基于Vuran等人提出的理论模型比较分析了三种节点筛选策略．数值实验结果表明传输低质量的监测数据意义不大，因为它们几乎不能降低甚至反而会增加在数据汇集端的重建失真（QoS）．基于此结论，本文提出了一种新型的分布式、能量高效MAC协议-QS-Sift，该协议赋予监测数据质量高的节点在接入信道时的高优先权，并抑制低质量监测数据的传输．以此进一步降低需要发送监测数据的节点数量，从而降低网赂能耗．模拟实验结果表明QS-Sift能够以较低的网络能耗、传输时延提供更好的QoS．

QS-Sift: QoS and Spatial Correlation-based Medium Access Control in Wireless Sensor Networks

Data transmitted in large scale, densely deployed wireless sensor networks (WSNs) is often highly spatial correlated. Recently it has been proposed to select a subset of sensor nodes to transmit their data so as to reduce energy consumption. In this paper, we investigate the ways of selecting sensor nodes in depth. Using the theoretical framework proposed by Vuran, we compare the performance of three sifting methods. We show that the transmission of poor sensed data (low SNR due to the far distance to the event) is unnecessary because the contribution of poor data to the given maximum distortion (QoS) is very small if not negative. Based on this observation, a new fully distributed, energy efficient MAC protocol is designed to assure QoS requirement and exploit spatial correlation, referred as QS-Sift. QS-Sift ensure that nodes with good sensed data (high SNR being close to the event) have higher priority than nodes with poor data when contending for the shared channel, and besides, poor data has fewer chance to be delivered. All these processing are performed by individual nodes without any centralized control. Extensive simulation has been done to verify the effectiveness of the propose schema in terms of energy consumption, transmission delay while ensuring a given QoS requirement.