Mobile Wireless Sensor Networks

output_kcirjh

Mobile Wireless Sensor Networks (Mobile WSNs) are widely used for various civilian and military applications, and thus have attracted significant interest in recent year. Our work investigates the important problem of optimal deployment of Mobile WSNs in terms of coverage and energy consumption. Both centralized and distributed algorithms are developed for sensor relocations after randomly deployment.

A.  Genetic Algorithm based Centralized optimizations

Genetic Algorithms are a type of evolutionary, optimization algorithm famous for global optimal search. We apply Multi-objective Genetic Algorithms to optimize the coverage and minimize the travel of sensors in order to save the energy and prolong the sensor network.

Figure 1 Simulation for Multi-objective Genetic Algorithms

(a) Initial Deployment                       (b) Case A                                (c) Case B
Figure 1 Simulation for Multi-objective Genetic Algorithms

Figure 1 shows the simulation results in coverage before and after our Multi-objective Genetic Algorithms. 10 sensors with fix range of 25 meters are randomly deployed in 100 meters by 100 meters sensing field. The initial coverage is 83%. Case A has an average travelled distance of 7.9 m and the coverage is about 97.6%, versus case B that has an average travelled distance of 16 m and 100% coverage.

B.  Average relative position based Distributed relocation algorithm

In this distributed algorithm, the relative positions are used in each sensor to calculate its optimal position so that a better coverage can be achieved for the entire sensor network. Figure 2 shows an example while two obstacles existed in a sensing field. The coverage in this case increased from around 30% to more than 90%.

C.  Optimization without Localization System Figure 2 Example of sensor relocation

C. Optimization without Localization System
Figure 2 Example of sensor relocation

This is also a distributed relocation algorithm. It does not require location or relative location information. Sensors use received signal strength to obtain the relative distance between neighbor sensors.

Three different initial deployments are simulated: In the first scenario, all sensors are deployed in the central 50 by 50 meters area; the second scenario has all the sensors randomly deployed in the entire area; and in the third scenario, sensors are split into two groups and deployed near the left and right boundary. The initial coverage of the three scenarios are 50%, 68% and 51%, respectively with a sensing radius of 15.9 meters. Multiple simulations are done in order to find out the performance of the algorithm. The cumulated density function of the coverage after 20 rounds of algorithm is shown in figure 3.

 

Figure 3 Coverage after 20 rounds in statistic probability

Figure 3 Coverage after 20 rounds in statistic probability

About Kun Bao