摘要
无线传感器网络(wireles ssensor networks,简称WSNs)通常是资源有限的,且具有较高的链路差错率.在这样的网络环境中,高可靠性的传输协议会消耗过多的能量并产生较大的传输延迟.而另一方面,许多部署在WSNs环境中的应用可以容忍一定程度的数据包丢失.因此,在这种特殊的网络背景和应用需求环境中,随机投递传输协议应运而生.这种类型的传输协议在有限的网络资源和传输可靠性之间实现了适当的折衷,符合众多传感器网络应用系统的特殊需要.为了对已有的几种随机投递传输协议进行性能分析与评价,采用带吸收态的有限状态马尔可夫链模型对其进行建模.这种建模方法使得协议性能参数的分析计算过程更加直观、简练、易于理解.分析了不同网络参数条件(如源到目的节点之间的跳距,无线链路位差错率等等)对于协议性能的影响.为了提高随机可靠传输协议的效率,部分协议中引入了逐条应答机制或利用无线链路广播特性.分析结果显示,这些措施对协议性能的影响在不同的网络参数条件下也有所不同.最后,为体现分析模型的作用,基于理论分析结果对部分协议进行了改进,并仿真显示改进的效果.
In general, the Wireless Sensor Networks （WSNs） are resource constrained, and with high Bit Error Rate （BER） links. Highly reliable transport protocol for this kind of network is challenging and costly in terms of energy and delay expenditure. On the other hand, many applications deployed on WSNs can live with some packets losses provided that the loss rate is tolerable. Hence, the stochastic delivery transport protocols emerge as the applications and network constrains require. The stochastic delivery transport protocols carry out a profitable trade-off between the reliability and resource cost, thereby are adopted by many applications in WSNs. To analyze the performance metrics of this kind of protocol in multi-hop WSNs, a Finite State Markov Chain （FSMC）-based model is developed in this paper. By using this model the performance parameters of the protocols can be calculated directly, easily and comprehensibly. The effects of different network parameters （such as number the hops, the bit error rate of the wireless link） on the performance are investigated. To enhance the efficiency of stochastic delivery protocols, hop by hop acknowledgement scheme is introduced in some stochastic reliable transport protocols, and so does the broadcast character of the wireless channel. The analytical results show that the effects of these schemes on performance are diverse with the change of network parameter settings. Finally, the paper presents some advice for improving these protocols based on the analysis. Simulation results also demonstrate the effectiveness of these improvements.
出处
《软件学报》
EI
CSCD
北大核心
2009年第4期942-953,共12页
Journal of Software
基金
国家自然科学基金Nos.60573122
60773138
国家高技术研究发展计划(863)
国家重点基础研究发展计划(973)
新世纪优秀人才支持计划~~
关键词
基础矩阵
带吸收态的有限状态马尔可夫链
建模
性能分析
传输协议
无线传感器网络
fundamental matrix
finite state Markov chains with absorbing states
modeling
performance analysis
transport protocol
wireless sensor networks （WSNs）
