- مبلغ: ۸۶,۰۰۰ تومان
- مبلغ: ۹۱,۰۰۰ تومان
Recent technological advances and an increasing thrust toward automation have resulted in a rapid adoption of Wireless Sensor Networks (WSNs) as the de facto choice in monitoring and surveillance applications. Their low cost, versatility and ability to operate around the clock in inhospitable terrains without endangering human life make WSNs ideal for applications like space exploration, environmental monitoring and combat zone surveillance. In these applications WSNs are to operate autonomously for prolonged durations; thus self-healing from failures becomes a requirement to ensure robustness through sustained network connectivity. The paucity of resources makes node repositioning the method of choice to recover from failures that partition the network into numerous disjoint segments. In this paper we present a Geometric Skeleton based Reconnection approach (GSR) that exploits the shape of the deployment area in order to restore connectivity to a partitioned WSN in a distributed manner. GSR decomposes the deployment area into its corresponding two dimensional skeleton outline, along which mobile relays are populated by the surviving disjoint segments to reestablish connectivity. The performance of GSR is validated through mathematical analysis and simulation.
In this paper we have presented GSR, a distributed algorithm that 781 enables a WSN to restore connectivity after the failure of multiple col- located nodes that partitions the network into disjoint segments. The main idea is to exploit the pre-failure network topology to determine the skeleton of the WSN which can be utilized as a template along which mobile nodes can be deployed in case of failure in order to form a connected inter-segment topology. Unlike previously published ap- proaches GSR forms a resource-efficient inter-segment topology and does not require further optimization once initial connectivity has 789 been established amongst the segments. The simulation results have demonstrated that GSR scales well and outperforms competing ap- proaches in terms of the number of required mobile nodes and the distance they need to travel. In the future, we plan to tackle the dis- tributed connectivity restoration problem under resource and secu- rity constraints that may make certain locations unsafe for MN de- ployment and avoid making the topology structure predicable to an adversary.