Green Deployment of Camera-Assisted mmWave Networks via Submodular Optimization

Abstract

This paper discusses a network deployment problem for camera- assisted millimeter-wave (mmWave) access networks with the purpose of optimizing the coverage probability with the constraint of a predefined total energy budget. The camera-assisted mmWave access network in which mmWave communications are controlled using RGB-depth (RGB-D) camera information has been proposed to mitigate the communication outage due to human blockage. The synergistic effect between the mmWave system and the cameras reduces not only the outage probability but also the total energy consumption. In order to utilize the synergistic effect, the deployment of mmWave base stations (BSs) and RGB-D cameras is formulated as a submodular optimization problem, assuming that the deployment of BSs and cameras are modeled as two independent Poisson point processes. An approximate algorithm is presented to solve the deployment problem, and it is proven that a (1-e^(-1) )/2-approximate solution is obtained for submodular optimization using a modified greedy algorithm. The numerical results reveal deployment requirements under which the camera-assisted mmWave system can achieve a higher coverage probability compared to a conventional system that does not employ RGB-D cameras. The results also show that the approximate solution are obtained with a low computation time using the greedy algorithm.