3D space detection and coverage of wireless sensor network based on spatial correlation

With an increasing number of the WSN applications in 3D space such as in outer space, atmosphere or underwater, the 3D space signal detection and coverage problems become more and more important. However, the 2D assumption, such as the binary detection model, is still used in the research referred to the 3D circumstance to determine whether the event to be detected occurs or not, leading to major inaccuracies. This assumption maybe useful, but in most circumstances, where sensor nodes are distributed in 3D space, this is not the case because the signal intensity from the event we want to detect decreases while the distance from the sensor node increases. At the same time, the signal is interfered by the noise in 3D space. These factors make the sensor node hard to accurately detect the occurrence of the event due to "false alarm" or "missing alarm", resulting in extra difficulties in 3D wireless sensor network coverage. Here we focus on detection issues by using probabilistic sensing model with five different collaborative detectors based on spatial correlation and signal detection theory. Based on the above analysis, we propose the 3D space detection and coverage growing algorithm, which uses probabilistic unit sensing model and four different polyhedrons, aims at achieving the seamless 3D space coverage while the number of nodes required for a fixed space is minimized. Results from simulations demonstrate that the sensing model with collaborative energy detector (ECD) achieves the widest sensing radius, which is 1.1 times of Collaborative CD; 1.48-1.62 times of Collaborative ED. Results also demonstrate that our algorithm could achieve the seamless 3D space coverage and truncated octahedron is the best to fill the determined space, resulting in the least nodes required. (C) 2015 Elsevier Ltd. All rights reserved.