Fault-Tolerant Relay Deployment Based on Length-Constrained Connectivity and Rerouting Centrality in Wireless Sensor Networks

Author(s):  
Lanny Sitanayah ◽  
Kenneth N. Brown ◽  
Cormac J. Sreenan
2018 ◽  
Vol 14 (2) ◽  
pp. 131-137
Author(s):  
Lanny Sitanayah

Wireless Sensor Networks (WSNs) are subject to failures. Even though reliable routing protocols for WSNs exist and are well-understood, the physical network topology must ensure that alternate routes with an acceptable length to the sinksare in fact available when failures occur. This requires a sensor network deployment to be planned with an objective of ensuring some measure of robustness in the topology, so that when failures do occur the protocols can continue to offer reliable delivery. To ensure that sensor nodes have sufficient paths, it may be necessary to add a number of additional relay nodes, which do not sense, but only forward data from other nodes. In this paper, we review a range of existing algorithms to deploy relay nodes for fault tolerance. We classify the state-of-the-art relay placement algorithms based on routing structures, connectivity requirements, deployment locations, and fault-tolerant requirements.


Fault Tolerant Reliable Protocol (FTRP) is proposed as a novel routing protocol designed for Wireless Sensor Networks (WSNs). FTRP offers fault tolerance reliability for packet exchange and support for dynamic network changes. The key concept used is the use of node logical clustering. The protocol delegates the routing ownership to the cluster heads where fault tolerance functionality is implemented. FTRP utilizes cluster head nodes along with cluster head groups to store packets in transient. In addition, FTRP utilizes broadcast, which reduces the message overhead as compared to classical flooding mechanisms. FTRP manipulates Time to Live values for the various routing messages to control message broadcast. FTRP utilizes jitter in messages transmission to reduce the effect of synchronized node states, which in turn reduces collisions. FTRP performance has been extensively through simulations against Ad-hoc On-demand Distance Vector (AODV) and Optimized Link State (OLSR) routing protocols. Packet Delivery Ratio (PDR), Aggregate Throughput and End-to-End delay (E-2-E) had been used as performance metrics. In terms of PDR and aggregate throughput, it is found that FTRP is an excellent performer in all mobility scenarios whether the network is sparse or dense. In stationary scenarios, FTRP performed well in sparse network; however, in dense network FTRP’s performance had degraded yet in an acceptable range. This degradation is attributed to synchronized nodes states. Reliably delivering a message comes to a cost, as in terms of E-2-E. results show that FTRP is considered a good performer in all mobility scenarios where the network is sparse. In sparse stationary scenario, FTRP is considered good performer, however in dense stationary scenarios FTRP’s E-2-E is not acceptable. There are times when receiving a network message is more important than other costs such as energy or delay. That makes FTRP suitable for wide range of WSNs applications, such as military applications by monitoring soldiers’ biological data and supplies while in battlefield and battle damage assessment. FTRP can also be used in health applications in addition to wide range of geo-fencing, environmental monitoring, resource monitoring, production lines monitoring, agriculture and animals tracking. FTRP should be avoided in dense stationary deployments such as, but not limited to, scenarios where high application response is critical and life endangering such as biohazards detection or within intensive care units.


Sign in / Sign up

Export Citation Format

Share Document