Wireless sensor network comprises several small nodes distributed randomly in an area under observation. These nodes sense the assigned physical phenomenon occurring around it and reports back to the base station. WSN is increasingly applied in defense, industrial, health care applications etc. Some applications envisage movement of the nodes. This results in frequent alteration of network topology leading to dynamic network partitioning. Hence, the routing of data from source to sink faces more routing disruptions. Maintaining application specific quality of service (QoS) in terms of Packet delivery ratio in such Mobile wireless sensor network (MWSN) environment is a daunting task. This cannot be achieved by focusing on a solution in only a single layer of sensor network protocol stack. Instead, by incorporating different techniques in each layered OSI protocol stack, a single coherent framework is proposed in this paper. This paper proposes prediction based coding in application layer, opportunistic routing in network layer, Laplacian scaling based on round trip time measurement in session layer. The proposed cross layer solution ensures application specific QoS guaranteed fault tolerance in MWSN.

无线传感器网络包括随机分布在观察区域中的几个小节点。这些节点感知周围发生的指定物理现象，并报告给基站。WSN越来越多地应用于国防，工业，医疗保健等应用。一些应用设想了节点的移动。这导致网络拓扑结构频繁更改，从而导致动态网络分区。因此，从源到接收器的数据路由面临更多的路由中断。在这样的移动无线传感器网络（MWSN）环境中，根据分组传送率来维持应用程序特定的服务质量（QoS）是一项艰巨的任务。仅将重点放在传感器网络协议堆栈的单层解决方案中是无法实现的。代替，通过在每个分层的OSI协议栈中加入不同的技术，本文提出了一个统一的框架。本文提出了基于预测的应用层编码，网络层的机会路由，基于会话层往返时间测量的拉普拉斯定标。所提出的跨层解决方案确保了MWSN中特定于应用程序的QoS保证的容错能力。