一种密度预测与服务分级的MAC退避算法  

A backoff algorithm of MAC protocol with density prediction and service classification

作　　者：苏海武[1] 程良伦[1] 高锐[1] 苏新凌[1] 孙志敬[1] 

机构地区：[1]广东工业大学自动化学院,广东广州510006

出　　处：《电子技术应用》2013年第10期112-115,共4页Application of Electronic Technique

基　　金：广东省重大科技专项(2009A080207008);广东省科技计划项目(2009B010800010);国家自然科学基金重点项目(U2012A002D01)

摘　　要：为了提高重负载的中高速无线传感器网络性能,深入研究竞争型MAC协议的退避算法,基于PTTL退避算法提出一种密度预测及服务分级的MAC退避算法。该算法对网络邻近节点数目进行加权递推平滑预测,实现竞争窗口自适应节点密度的目的;引入服务分级意识,赋予服务级别高或数据积压或跳数多的节点优先发送权,满足关键数据多跳传输实时性要求。NS2仿真结果表明:本DPSC退避算法在节点高密度与高负载环境下网络性能优于其他三种算法,其平均时延比PTTL算法降低10%,吞吐量提高15%,平均能耗下降5%。In order to improve the performance in high-speed sensor networks which traffic load is high,study backoff algorithm of competitive MAC protocol,and propose a backoff algorithm of MAC protocol with density prediction and service classification which is based on PTTL.The protocol forecasts the number of neighboring nodes by weighted recursive smoothing,to achieve the purpose of the competition window adaptive node density; introducing awareness of the service classification,and giving the right of send priority to the nodes that are the high level of service,or data backlog or multi-hop,to meet the real-time requirements of critical data multi-hop transmission.The NS2 simulation results show that：the network performance of DPSC backoff algorithm in the nodes of high-density and high-load environment is superior to the other three algorithms.The average delay was 10％ lower than the PTTL algorithm,throughput increased by 15％,and the average energy consumption decreased by 5％.

关 键 词：中高速无线传感器网络 退避算法 密度预测 服务分级 优先发送 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]