Energy-Efficient Resource Allocation in Multi-cell Virtualized Wireless Networks

This paper considers the down-link transmission of an OFDMA-based multi-cell virtualized wireless network (VWN) to serve users belonging to different service providers (slices) with the quality-of-service (QoS) requirements in terms of each slice's minimum reserved rate. In order to improve energy efficiency, we formulate a joint base station (BS) assignment, sub-carrier and power allocation problem to minimize the total transmit power of all the BSs subject to the QoS constraints. This problem is inherently a non-convex optimization problem. To tackle its computational complexity, we apply successive convex approximation (SCA) and complementary geometric programming (CGP) to convert the problem into a computationally efficient formulation and propose an iterative algorithm for solving the problem. We introduce a variable called user association factor (UAF) for jointly assigning users to both BSs and sub-carriers. Simulation results illustrate the performance enhancement of the VWN achieved through our formulation for different network scenarios.