Joint Optimization of Energy and Latency for Hybrid Near-Field and Far-Field Localization with Multi-access Edge Computing

This paper investigates an energy and delay optimization framework for effective integrated near-field and far-field localization within multi-access edge computing (MAEC). In the system under consideration, each user has the option to execute the task locally on their device or to transfer it to the nearest MAEC server or central server, depending on the transmission conditions associated with near-field or far-field propagation. Our objective is to concurrently reduce energy consumption and computational latency within the bounds of practical constraints related to transmission, computation, and offloading strategies. The proposed optimization problem exhibits a firmly non-convex structure due to the intricate interdependence between decision variables such as offloading selection and computational resource allocation among the MAEC server and the central server; consequently, solving this problem presents a significant challenge. To address this issue, we employ the alternating optimization algorithm, which enables the decomposition of the original problem into manageable subproblems and facilitates an iterative process to progressively update the optimal variables. Simulation results demonstrate that the proposed method yields superior performance compared to conventional approaches.