Information Fusion 30 (2016) 36–51 Contents lists available at ScienceDirect Information Fusion journal homepage: www.elsevier.com/locate/inffus A centralized immune-Voronoi deployment algorithm for coverage maximization and energy conservation in mobile wireless sensor networks Mohammed Abo-Zahhad a , Nabil Sabor a,b,∗ , Shigenobu Sasaki b , Sabah M. Ahmed a a Electrical and Electronics Engineering Department, Faculty of Engineering, Assiut University, Assiut 71516, Egypt b Department of Electrical and Electronic Engineering, Niigata University, Niigata 950-2181, Japan article info Article history: Received 29 April 2015 Revised 3 September 2015 Accepted 15 November 2015 Available online 22 November 2015 Keywords: Mobile wireless sensor networks Immune algorithm Voronoi Diagram Coverage area Node deployment abstract Saving energy is a most important challenge in Mobile Wireless Sensor Networks (MWSNs) to extend the lifetime, and optimal coverage is the key to it. Therefore, this paper proposes a Centralized Immune-Voronoi deployment Algorithm (CIVA) to maximize the coverage based on both binary and probabilistic models. CIVA utilizes the multi-objective immune algorithm that uses the Voronoi diagram properties to provide a better trade-off between the coverage and the energy consumption. The CIVA algorithm consists from two phases to improve the lifetime and the coverage of MWSN. In the first phase, CIVA controls the positions and the sensing ranges of Mobile Sensor Nodes (MSNs) based on maximizing the coverage and minimizing the dis- sipated energy in mobility and sensing. While the second phase of CIVA adjusts the radio (sleep/active) of MSNs to minimize the number of active sensors based on minimizing the consumption energy in sensing and redundant coverage and preserving the coverage at high level. The performance of the CIVA is compared with the previous algorithms using Matlab simulation for different network configurations with and without obstacles. Simulation results show that the CIVA algorithm outperforms the previous algorithms in terms of the coverage and the dissipated energy for different networks configurations. © 2015 Elsevier B.V. All rights reserved. 1. Introduction In recent years, Wireless Sensor Networks (WSNs) have found to monitor the environment, track targets on a battlefield, measure traffic on roads, monitor patients or track the location of personnel inside buildings [1]. One of the key points in the design of WSNs that is related to the sensing attribute is the coverage of the sensor field. Coverage has a direct effect on the network performance, thus it considered as the measure of quality of service in WSNs. The deployment strategy of sensor nodes in the field is the most critical factor related to the coverage and the connectivity. Sensors can be deployed either randomly or deterministically. Random deployment is usually preferred in large scale WSNs not only because it is easy and less expensive but also because it might be the only choice in remote and hostile environments. However, random deployment is not efficient approach and can cause coverage holes in the field. On the other hand, deterministic deployment is very complex in large and harsh environments and costs time [2]. ∗ Corresponding author at: Department of Electrical and Electronic Engineering, Niigata University, Niigata 950-2181, Japan. Tel.: +8108092746344. E-mail addresses: zahhad@yahoo.com (M. Abo-Zahhad), nabil_sabor@aun.edu.eg (N. Sabor), shinsasaki@ieee.org (S. Sasaki), sabahma@yahoo.com (S.M. Ahmed). Rearrangement the sensor nodes after initial deployment by at- tached them with vehicles or mobile robots can improve the net- work coverage and eliminate the coverage holes. However, moving the sensor nodes introduces new challenges in saving the consump- tion energy because the mobility systems of nodes consume more en- ergy [3]. Thus, many deployment algorithms based on a binary sens- ing model [1,2,4–9] and a probabilistic sensing model [10–12] have been developed recently to improve the network coverage. Some of these algorithms took in their consideration the mobility cost of all nodes besides improving the coverage, while other algorithms con- sidered the sensing range adjustment to save the dissipated energy in sensing and improve the coverage. However, these is no deployment algorithm considers the dissipated energy in the mobility, the sens- ing and the redundant coverage at the same time besides improving the network coverage. In order to provide a better trade-off between the coverage and the energy consumption, one of the recently evo- lutionary algorithms which called the Multi-Objective Immune Algo- rithm (MOIA) is considered here. The main features of MOIA algo- rithm compared to other algorithms are [13,14]: (1) It is the global search performance; (2) It produces the solution sets that are highly competitive in terms of convergence, diversity and distribution; (3) It has elitism which inherently embedded in the selection mechanism to preserve good solutions and not lose them during generations; http://dx.doi.org/10.1016/j.inffus.2015.11.005 1566-2535/© 2015 Elsevier B.V. All rights reserved.