Spatial Optimization of Coverage-Type 3-D Heterogeneous Visual Sensor Networks for Urban Security

Visual sensor networks (VSNs) is a key enabling technology for the development of smart cities. Recent years have witnessed great success in the mechanism design, modeling, and optimization techniques for VSNs. In this paper, we revisit the 3-D heterogeneous VSNs for urban security and design a novel spatial optimization framework to optimize the coverage comprehensively. The proposed framework bridges the computational tools with high performance in spatial geography, structured mathematical models, and modern heuristic algorithms, resulting in an emerging multidisciplinary research paradigm. Specifically, a problem-specific improved genetic algorithm combined with local search is embedded as the solver. Computational results through multiple scenarios of various scales show that the proposed algorithm outperforms the traditional solution approaches. Real-world case studies in Beijing, China are also presented and visualized to illustrate further the optimized configurations of VSNs as well as provide some operational insights for decision-makers in smart cities.