UCAA: User-Centric User Association and Resource Allocation in Fog Computing Networks

In recent years, with the eruptive popularity of mobile Internet and the emergence of various new IoT applications, fog computing is proposed to shift the cloud computing services towards the edge, making up for its lack of mobility support and high delay. Fog computing is customized for scenarios with scarce resources and unpredictable environments, but there is no user-centric joint optimization fog computing models designed for such scenarios. In this paper, we aim to maximize the user experience and overall system performance by jointly optimizing user association and resource allocation in the scenarios mentioned above, which can be formulated as a mix-integer non-linear programming problem. To solve the NP-hard problem, we propose a low-complexity two-step interactive optimal algorithm, named UCAA algorithm. For the user association problem, we propose a semi-definite programming based algorithm, and then further propose a Kuhn-Munkres algorithm based user association decision approximation algorithm. For the resource allocation problem, we first prove that it can be decoupled into two sub-problems, ie., transmission power selection problem and computing resource allocation problem, and solve them individually, in addition, we have given a rigorous proof that the optimal solution of the two sub-problems is the optimal solution to the original problem as well. The numerical results show that the proposed UCAA algorithm achieves better performance than conventional algorithms in terms of the value of average user-centric utility, especially in case of more user equipments (UEs), fewer fog nodes, limited computing capacity of fog nodes, lower delay tolerance, lower local computation capacity, etc., which presented to illustrate that the UCAA algorithm can significantly improve user experience and system performance in the considering fog computing scenarios.