A Hybrid Memetic Framework for Coverage Optimization in Wireless Sensor Networks

Abstract

One of the critical concerns in wireless sensor networks (WSNs) is the continuous maintenance of sensing coverage. Many particular applications, such as battlefield intrusion detection and object tracking, require a full-coverage at any time, which is typically resolved by adding redundant sensor nodes. With abundant energy, previous studies suggested that the network lifetime can be maximized while maintaining full coverage through organizing sensor nodes into a maximum number of disjoint sets and alternately turning them on. Since the power of sensor nodes is unevenly consumed over time, and early failure of sensor nodes leads to coverage loss, WSNs require dynamic coverage maintenance. Thus, the task of permanently sustaining full coverage is particularly formulated as a hybrid of disjoint set covers and dynamic-coverage-maintenance problems, and both have been proven to be nondeterministic polynomial-complete. In this paper, a hybrid memetic framework for coverage optimization (Hy-MFCO) is presented to cope with the hybrid problem using two major components: 1) a memetic algorithm (MA)-based scheduling strategy and 2) a heuristic recursive algorithm (HRA). First, the MA-based scheduling strategy adopts a dynamic chromosome structure to create disjoint sets, and then the HRA is utilized to compensate the loss of coverage by awaking some of the hibernated nodes in local regions when a disjoint set fails to maintain full coverage. The results obtained from real-world experiments using a WSN test-bed and computer simulations indicate that the proposed Hy-MFCO is able to maximize sensing coverage while achieving energy efficiency at the same time. Moreover, the results also show that the Hy-MFCO significantly outperforms the existing methods with respect to coverage preservation and energy efficiency.

An efficient genetic algorithm for large-scale transmit power control of dense and robust wireless networks in harsh industrial environments

“…Furthermore, compared to most coverage-related optimization problems that only rely on a 2D environment [1,14,15,17,20,21], an obstacle is modeled as a 3D geometrical model in the decision making of line-ofsight propagation between a GP-AP pair (i.e., the logical signal blockage variable k ij  ). Both the j-th AP and the Rx placed on the i-th GP have their own heights.…”

“…If the required coverage rate is not yet achieved (line 2), this random TPC solution will be corrected (lines [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. The idea of repair is to iteratively cover the uncovered GPs using the potential APs (whose current transmit power is below the maximal level) while maintaining the increase in transmit power as slight as possible (to comply with interference minimization).…”

“…At the first step of correction (lines 3-9), uncovered GPs are all found and their GP-AP links are all established. The second step of correction (lines [10][11][12][13][14][15][16][17][18][19][20][21][22] is an iteration. In each iteration, a GP is randomly selected from uncovered GPs and its nearest potential AP is found (line 11).…”

“…While large-scale optimization is increasingly desired [18,19], most research on coverage optimization problems neglects the scalability of an optimization algorithm [14,[20][21][22][23]. Large-scale problems are characterized in at least one of the following dimensions [24].…”

Problem Specific MOEA/D for Barrier Coverage with Wireless Sensors

“…Traditional algorithms might not be applicable and only a few techniques address these objectives simultaneously. Multiobjective evolutionary algorithm (MOEA), which is a population based algorithm that naturally leads to Pareto optimal solutions, has been successfully applied to deal with MOPs in WSNs [10]- [12]. In recent years, several MOEAs have been proposed to give approximations on the Pareto optimal solutions (Pos).…”