User provided network (UPN) allows a user with high channel quality to share the network access for users with poor channel quality. As a result, both the quality of experience and efficiency of network resource can be improved by UPN. However, the success of UPN relies heavily on the willingness of users to participate in sharing, so the design of incentive mechanisms is critical for UPN. In this paper, the UPN formed by D2D links under 5G integrated access and backhaul (IAB) networks is considered. In IAB networks, both access and backhaul links use wireless transmissions and dynamically share all the spectrum resources. Thus, the resource allocation for all the links also has a great impact on the UPN efficiency. To this end, a novel joint incentive and resource allocation design is explored. More specifically, considering the fairness between users, a Nash bargaining problem as a cooperative game is formulated by considering the user utility, the sensitivity of battery energy, the incentive compensation, and the limitation of network resources. To achieve the optimal Nash bargaining solution, a centralized algorithm is first designed, in which all the user information is collected by the operator for conducting centralized optimization. Thus, the centralized algorithm leads to a privacy problem. To this end, a distributed algorithm is developed to decompose the primal problem into subproblems for the operator and each user. By passing intermediate parameters between users and iterative execution of subproblems, the solution of the distributed algorithm is proved to converge to the optimal solution of the centralized algorithm. Extensive numerical results have been conducted to show that our design can effectively improve both the user experience and network throughput, i.e., operator's revenue.

中文翻译：

---

5G综合接入回传网络下用户自备网络联合激励与资源分配设计

用户提供网络（UPN）允许信道质量高的用户共享信道质量差的用户的网络接入。因此，UPN 可以同时提高体验质量和网络资源效率。但是，UPN的成功很大程度上依赖于用户参与分享的意愿，因此激励机制的设计对UPN来说至关重要。本文考虑了5G综合接入回传（IAB）网络下D2D链路形成的UPN。在 IAB 网络中，接入和回程链路都使用无线传输并动态共享所有频谱资源。因此，所有链路的资源分配对UPN效率也有很大影响。为此，探索了一种新颖的联合激励和资源分配设计。进一步来说，考虑用户之间的公平性，考虑用户效用、电池能量的敏感性、激励补偿和网络资源的限制，将纳什讨价还价问题作为合作博弈制定。为了实现最优的纳什讨价还价方案，首先设计了集中式算法，由运营商收集所有用户信息进行集中式优化。因此，集中式算法会导致隐私问题。为此，开发了一种分布式算法来将原始问题分解为操作员和每个用户的子问题。通过在用户之间传递中间参数和子问题的迭代执行，证明了分布式算法的解收敛于集中式算法的最优解。