The Research On Exploiting Physical Sensing Information For Bypassing Routing Holes

Wireless sensor network(WSN)consists of large number of low-power sensor nodes and is usually deployed in harsh environments.The quality of WSN communication links can be easily affected by changes in environmental characteristics.Mobile obstacles in the network can block the signal propagation path can cause data transmission failures,which forms routing holes.Existing routing protocols generally avoid failed paths by monitoring historical data transmissions.Due to low data transmission rate in WSN systems,path switching is usually quite time-consuming,and thus inevitably incurs significant data transmission delay and even failures.To address these issues,this thesis proposes to incorporate physical measurements obtained using sensors to bypass routing holes caused by obstacles.The major contributions of this thesis lie in three areas.(1)We verified the correlations between obstacles,physical measurements,and communication failures via empirical study.(2)We propose an algorithm which leverages physical measurements to learn the relative positions between nodes.The algorithm constructs an optimal obstacle movement trajectory inside the sensor grid,to allow the nearby sensor nodes to learn relative positions between each other directly or indirectly.(3)We present a physical-assisted routing algorithm.When an obstacle is detected within the monitored area,the algorithm uses physical measurements to determine the relative orientation of the obstacle with respect to nodes,and to quickly establish a hole bypassing path along the obstacle's boundary.We have verified the effectiveness of the proposed algorithms via extensive experiment on simulation platform and real testbed.Compared to existing collection tree protocol(CTP)with different link estimators(4Bit,PACE),the proposed routing protocol can significantly improve data transmission performance in terms of delivery ratio,hops,data transmission overhead and control overhead.