Study on Energy Conservation for Cellular Systems: A Global Optimization Approach

With the increasing popularity of wireless devices and multimedia services, data transmission in cellular systems has increased dramatically, resulting in huge energy consumption. This impacts negatively on the environment (e.g., high textCO-2 emissions and the depletion of nonrenewable energy sources) and increases the operation cost for telecommunications. Effectively solving this issue has become a significant research topic. In this study, we aim to minimize the total power consumption for cellular systems by jointly considering base station (BS) deployment and power allocation, while providing user transmission rate and quality of experience (QoE) guarantees. We construct a mathematical model which is a nonconvex mixed integer nonlinear programming (MINLP) problem. To solve it with global optimality, we design a novel branch-and-bound (B&B) algorithm based on the characteristics of the MINLP problem. We compare our proposed B&B algorithm with a reference B&B algorithm using the relaxation method published in previous literatures. Numerical results show that our algorithm substantially surpassed the reference algorithm in terms of computational complexity. Moreover, we investigate in detail the effect of distinct factor on the energy consumption of cellular systems. Our results not only provide insights into the design of green cellular systems but also serve as performance benchmarks for heuristic and distributed algorithms developed for practical implementations.