Optimal Cooperative Wireless Communication for Mobile User Data Offloading

• Main Authors: Guangsheng Feng, Fumin Xia, Yongmin Zhang, Dongdong Su, Haibin Lv, Huiqiang Wang, Hongwu Lv
• Format: Article
• Language: English
• Published: IEEE 2018-01-01
• Series: IEEE Access
• Subjects: Mobile data offloading; bi-level optimization; utility function; user satisfaction; cooperative transmission
• Online Access: https://ieeexplore.ieee.org/document/8289434/

We study the mobile data offloading problem in cooperative cellular networks, in which the cellular base station can offload a fraction of data traffic to its cooperative Wi-Fi networks. Both user satisfaction and benefit are important for the cellular network. Under the case that the cellular network cannot provide sufficient transmission resources, part of data traffic can be offloaded to its cooperative Wi-Fi networks to guarantee the user satisfaction, even though the benefit loss of the cellular network will be incurred. Meanwhile, considering the users are price sensitive, the operators, including cellular and Wi-Fi, can issue an appreciate unit price to attract more users and more traffic requirements to increase their total benefit. In this paper, we intend to design an optimal pricing scheme to maximize the user satisfaction and minimize the benefit losses of the operators simultaneously. First, we formulate the mobile data offloading problem as a joint optimization of “min–max” problem. Then, we convert the problem into a bi-level optimization problem, in which the lower level is to maximize the user satisfaction and the upper level is to minimize the benefit losses of the operators. Since the lower-level problem is convex, a centralized scheme based on the dynamic coordinate search-based method is employed to find the optimal solution of the bi-level problem. In addition, a distributed method is proposed to obtain the near-optimal solution quickly, where one-third-party agent is used to adjust the data traffic allocation. Simulation results show that the proposed method achieves the near-optimal solutions with limited iterations.