Research On Mobility Schemes Of Sink In Wireless Sensor Networks

Wireless Sensor Network (WSN) has become one of the hottest research problems ofmany countries. Since many sensor nodes usually are deployed over where people can notreach, the sensor nodes can not be changed fluently. Energy is the most important resource inthe network. Nodes located near sink deliver packets more fluently that those which arefarther from sink. In order to solve the problem advanced above, idea of sink mobility cameout. If sink changes its location after it harvests data packets for a while, then set of nodes thatnear sink will be changed as well. How the mobile sinks move is an important factor thatinfluences how long the lifetime of wireless sensor networks is last. Currently, there existssome autonomous motion like random mobility, fixed traffic mobility and somenon-autonomous motion, such as moving towards max left energy area. Non-autonomousmotion is not sensitive about real time network information. It is kind of mobility scheme,that is not depended on real time information. On the contrary, autonomous motion is partly orabsolutely depends on the whole states of sensor nodes. It always needs to collect enoughinformation to decide which location is the next destination for sink. And this situation needsmore time and energy of sensor nodes, leading to network delay and cost. At the same time,the motion of sink brings new challenge, especially how to maintain the breakage of data linkbetween source nodes and sink because of sink mobility. The main work of this article islisted below:1. Work out a new routing maintain algorithm to support sink mobility, it is a simplealgorithm related to left energy of sensor nodes. We name it ERSMA(Energy Related SimpleMaintain Algorithm). In order to remain the link between source and sink, it uses old set ofneighbor nodes of sink and the new one to work out the intersection, then remains part of theold path and fixes the break part.2. ERSMA is used to support random motion and mobile scheme towards the area ofmost left energy. Then we compare network metrics under different network parameters.Network parameters include the communication radius of sensor nodes, the density of sensornodes deployed in the scene, the speed of sink and the query interval of sink. The networkmetrics include average left energy of sensor nodes, the variance of left energy, lifetime of network and packet lost radio.3. To avoid the blindness of non-autonomous motion and the vast overload caused bycollecting extra information to support autonomous motion, a pattern of moving towardssource area is proposed in this paper which is called lightweight autonomous motionaccording to the fact that events are concentrated in some region.4. A protocol called FTR (Focus Tracking Routing) is designed to support this mobilityscheme. A route forwarding method that only depends on local Euclidean distancecomputation is provided, aiming to avoid collecting global information and reduce thecomputational complexity. We analyze the effects of the width of transmission area on thenetwork lifetime mathematically. Simulation results show that the proposed sink mobilityscheme supported by FTR can prolong network lifetime, reduce average routing hop countand packet delivery ratio and balance energy consumption.