Integrated sensing and communication (ISAC) base stations can provide communication and wide range sensing information for vehicles via downlink (DL) transmission, thus enhancing vehicle driving safety. One major challenge for realizing high performance communication and sensing is how to deal with the DL mutual interference among adjacent ISAC base stations, which includes not only communication related interference, but also radar sensing related interference. In this paper, we establish a DL mutual interference model of adjacent ISAC base stations, and analyze the relationship for mutual interference channels between communications and radar sensing. To improve the sensing and communication performance, we propose a collaborative precoding design for coordinated adjacent base stations to mitigate the mutual interference under the transmit power constraint and constant modulus constraint, which is formulated as a non-convex optimization problem. We first relax the problem into a convex programming by omitting the rank constraint, and propose a joint optimization algorithm to solve the problem. We furthermore propose a sequential optimization algorithm, which divides the collaborative precoding design problem into four subproblems and finds the optimum via a gradient descent algorithm. Finally, we evaluate the collaborative precoding design algorithms by considering sensing and communication performance via numerical results.

翻译：集成感知与通信（ISAC）基站可通过下行链路（DL）传输为车辆提供通信及广域感知信息，从而提升车辆驾驶安全性。实现高性能通信与感知的主要挑战之一是如何处理相邻ISAC基站间的下行互干扰，该干扰不仅包含通信相关干扰，还涉及雷达感知相关干扰。本文建立了相邻ISAC基站的下行互干扰模型，分析了通信与雷达感知互干扰信道之间的关系。为提升感知与通信性能，我们提出了一种面向协调相邻基站的协同预编码设计方案，在发射功率约束与恒模约束下抑制互干扰，并将其建模为非凸优化问题。我们首先通过省略秩约束将问题松弛为凸规划，并提出联合优化算法进行求解；进一步提出一种序贯优化算法，将协同预编码设计问题分解为四个子问题，通过梯度下降算法寻求最优解。最后，通过数值结果从感知与通信性能角度对所提协同预编码设计算法进行评估。