Wireless Sensor Network deployment using stochastic optimization techniques - a comparative study

For over a decade, the Minimum Cost Coverage Deployment Problem (MCCDP) has been an active research topic in the field of Wireless Sensor Networks (WSNs). The MCCDP is modeled as a combinatorial constrained optimization problem which was proven to be NP-Complete. Consequently, many of the existing deployment algorithms designed to solve the MCCDP are based on stochastic optimization techniques. These techniques can be heuristic such as Greedy Heuristics (GHs) or metaheuristic such as Genetic Algorithms (GAs) and Ant Colony Optimization (ACO). However, the performance of these algorithms has not been evaluated thoroughly, especially for solving MCCDPs of large scales, i.e. for deploying large-scale WSNs. Moreover, quantitative comparisons of the different algorithms that belong to these techniques are yet to be conducted, to the best of our knowledge. These comparisons are quite important in providing some insights into the suitability of a technique for a specific WSN deployment problem at hand. In this paper, we present a comparative performance evaluation of four of the existing algorithms designed to solve the MCCDP which are based on the aforementioned stochastic optimization techniques. Performance is evaluated in terms of quality of obtained results, computational cost, convergence speed and deployment scalability. Based on the statistical analysis of the experimental results, a comparison is conducted to highlight the comparative strengths and shortcomings of each algorithm.