Deployment in wireless sensor networks by parallel and cooperative parallel artificial bee colony algorithms

Abstract

Increasing number of cores in a processor chip and decreasing cost of distributed memory based system setup have led to emerge of a new work theme in which the main concern focused on the parallelization of the well-known algorithmic approaches for utilizing the computational power of the current architectures. In this study, the performances of the conventional parallel and cooperative model based parallel Artificial Bee Colony (ABC) algorithms on the deployment problem related to the wireless sensor networks were investigated. The results obtained from the experimental studies showed that parallelized ABC algorithm with the cooperative model is capable of finding similar or better coverage ratios with the increased convergence speeds than its serial counterpart and parallelized implementation in which the emigrant is chosen as the best food source in the current subcolony.

On the performance of quick artificial bee colony algorithm for dynamicdeployment of wireless sensor networks

“…Details of these algorithms can be found in [24,25]. Since the results of the ABC, pABC, and coop-pABC algorithms were taken from [24] and the results of the tlABC and p-tlABC algorithms were taken from [25], similar parameter settings and scenarios were used in this comparison study. The binary detection model is considered with 100 mobile sensors, r d is 7 m, and the area of interest is 10000 m 2 with 100 m length and 100 m width.…”

“…In addition, the other algorithms used in the comparison were applied to this problem by considering the position information of a sensor (both x and y coordinates) as the value in one dimension while the qABC handle the coordinate information in different dimensions. As in [24,25], 20 independent runs were executed, and the means and standard deviations of them are given in Table 9.…”

“…Performance of the qABC was compared with some other optimization methods including the ABC, a parallel ABC (pABC) [24], a cooperative parallel ABC (coop-pABC) [24], a version of the ABC powered by a transition control mechanism (tlABC) [25], and a parallel version of the tlABC (p-tlABC) [25] on the dynamic deployment problem of WSNs. Details of these algorithms can be found in [24,25]. Since the results of the ABC, pABC, and coop-pABC algorithms were taken from [24] and the results of the tlABC and p-tlABC algorithms were taken from [25], similar parameter settings and scenarios were used in this comparison study.…”

“…Yadav et al proposed a modified version of ABC by addressing two drawbacks of the standard ABC algorithm using a crossover operator and a hybrid local search, and they tested its performance on a dynamic deployment problem with mobile sensors and considered a probabilistic detection model [23]. Aslan applied a parallel ABC and a cooperative parallel ABC to this problem by using a binary detection model and a WSN consisting of only mobile sensors [24]. He also proposed a new modified ABC by using a transition control mechanism and used it as a serial version and a parallel version for solving the dynamic deployment problem of WSNs that include mobile sensors with a binary detection model [25].…”