Node Distribution Optimization in Positioning Sensor Networks through Memetic Algorithms in Urban Scenarios

Abstract

Local Positioning Systems rely on ad-hoc node deployments which fit the environment characteristics in order to reduce the system uncertainties. The obtainment of competitive results through these systems requires the solution of the Node Location Problem. This problem has been assigned as NP-Hard, therefore a heuristic solution is recommended for addressing this complex problem. Genetic Algorithms (GA) have shown an excellent trade-off between diversification and intensification in the literature. However, in Non-Line-of-Sight environments in which there is not continuity in the fitness function evaluation among contiguous solutions, challenges arise for the GA. Consequently, in this paper, we introduce a Memetic Algorithm (MA) with a Local Search strategy for exploring the most different individuals of the population in search of improving the NLP results in urban scenarios for the first time. Results show that the MA proposed outperforms the GA optimization and attains an improvement of 6.51% in accuracy in the scenario proposed.

Memetic Chains for Improving the Local Wireless Sensor Networks Localization in Urban Scenarios

“…This section presents the main differences between the previously proposed methodologies (i.e., GA and MA) and the results obtained for the MA-VND-Chains presented in this paper. These results show that the MA-VND-Chains improves the overall performance of the TDOA architecture and reduces the Root Mean Squared Error (RMSE) of the sensor distribution even more than our previous MA [28]. Table 1 shows the parameters used for the positioning signal link and for the architecture sensors characteristics based on real LPS applications.…”

“…However, the characteristics of the NLP with their inherent discontinuities in the fitness evaluations promotes the adaptation of the MA-Solis Wets-Chains (MA-SW-Chains) algorithm introduced in [52] for the discontinuous optimization of the NLP. We adapt this algorithm in this paper through the consideration of a different LS strategy (i.e., Variable Neighborhood Descent (VND [53]) versus the classical Solis Wets algorithm [54] of the MA-SW-Chains algorithm) and the consideration of the movement of the sensor nodes during the LS process for the definition of an intelligent strategy for analyzing the neighborhood of the NLP potential solutions which also differs from our previous strategies implemented in [28,44]. The introduction of the memory chains must also be adapted in this paper to the particularities of the NLP obtaining finally improved results in the solution presented with regards to the previous methodologies proposed for the NLP.…”

“…The analysis of the changes generated for the MA-VND-Chains algorithm quantifies the improvement introduced with respect to the GA [35] and the MA [28] from our earlier papers. After the LS process, a reevaluation of all the individuals that have been modified through the improvement of the pseudo-function is required.…”

“…Consequently, no a priori suitable time-based architecture can be defined in LPS applications [26] and a deep analysis on the environment characteristics must be performed to objectively determine the most appropriate architecture for a particular scenario. In this paper, we analyze the deployment of the synchronous TDOA architecture in a complex urban scenario for their extended use in the literature [27,28] and due to their trade-off among noise and clock uncertainties in LPS applications.…”

“…Therefore, we first proposed a GA [35] for the NLP in LOS environments and we later improved our algorithm through an intelligent employment of the genetic operators during the diversification and intensification phases of the optimization [49]. The dimensions of the space of solutions of the NLP and the consideration of the NLOS paths in the fitness function generating discontinuities in the quality evaluations suggested the introduction of MA (i.e., hybridizations of GA with Local Search (LS) procedures) in both outdoor [44] and urban environments [28]. This memetic strategy for the NLP proved to be promising for analyzing the space of solutions of the NLP which presents numerous local optima complicating the optimization process.…”