Lifetime-Guaranteed Cost-Minimized Heterogeneous Visual Sensor Networks With Feature Collection for Effective Target Coverage in 3-D Space

In visual sensor networks (VSNs), the generation and transmission of a large quantity of image data lead to huge network traffic. As a result, their lifetime is short and their delay is long. Therefore, reducing their traffic is critical to prolonging their lifetime and shortening their latency. Since the feature data of an image is much smaller than raw data, deploying processing nodes near camera nodes (CNs) to extract the features of their images for transmission can greatly reduce network traffic. Moreover, installing CNs of multiple types can also help since their use can decrease the total size of images. This work comprehensively utilizes these two measures and investigates how to deploy a heterogeneous VSN to satisfy sampling frequency requirements of targets and gather features of their images while minimizing the network deployment cost and meeting the network lifetime requirement. We formulate this important and new problem as a mixed integer nonlinear program and propose a separate node deployment algorithm and a joint one to solve it. Extensive simulation results demonstrate that both success rate and solution quality of the latter are higher than those of the former at the expense of computational time.