Alternative Positioning, Navigation, and Timing Enabled by Games in Satisfaction Form and Reconfigurable Intelligent Surfaces
Alternative Positioning, Navigation, and Timing Enabled by Games in Satisfaction Form and Reconfigurable Intelligent Surfaces
复制标题
DOI：
10.1109/jsyst.2023.3268989
复制
发表时间：
2023-09
影响因子：
4.4
通讯作者：
Md. Sadman Siraj;Aisha B. Rahman;Maria Diamanti;E. Tsiropoulou;S. Papavassiliou
中科院分区：
文献类型：
--
作者：
Md. Sadman Siraj;Aisha B. Rahman;Maria Diamanti;E. Tsiropoulou;S. Papavassiliou
全球定位系统和其他全球导航卫星系统在多种情况下的潜在退化导致了替代定位、导航和授时（PNT）技术的发展，旨在维持高效和安全的运行。在本文中，我们利用可重构智能表面 (RIS) 来设计替代 PNT 解决方案，从而提高准确性和效率。通过采用博弈论和强化学习（RL）来处理 RIS 编排和配置的具体问题。最初，在目标之间制定并解决满意度博弈，使他们能够自主确定有助于其 PNT 服务的最佳 RIS 数量，而要使用的特定 RIS 集由新颖的 RL 算法确定。为了进一步最大化来自特定RIS组的反射信号的每个目标处的接收信号强度，执行后者的相移优化。在此基础上，采用迭代最小二乘算法，遵循多边定位技术，以便每个目标估计其位置和定时。该方法的性能评估是通过建模和仿真来实现的。
The potential degradation of the Global Positioning System and other Global Navigation Satellite Systems under several circumstances gives rise to the development of alternative position, navigation, and timing (PNT) technologies aiming at maintaining efficient and safe operations. In this article, we exploit the use of reconfigurable intelligent surfaces (RISs) as enablers for the design of an alternative PNT solution, with improved accuracy and efficiency. The specific problem of RISs' orchestration and configuration is treated via the adoption of game theory and reinforcement learning (RL). Initially, a satisfaction game is formulated and solved among the targets, enabling them to autonomously determine the optimal number of RISs that will contribute to their PNT service, while the specific set of RISs to be used is determined by a novel RL algorithm. In order to further maximize the received signal strength at each target of the reflected signals from the specific set of RISs, the phase-shift optimization of the latter is performed. Based on the above, an iterative least squares algorithm is adopted, following the multilateration technique, in order for each target to estimate its position and timing. The performance evaluation of the proposed approach is achieved via modeling and simulation.