Integrated sensing and communication (ISAC) is expected to play a prominent role among emerging technologies in future wireless communications. In particular, a communication radar coexistence system is degraded significantly by mutual interference. In this work, given the advantages of promising reconfigurable intelligent surface (RIS), we propose a simultaneously transmitting and reflecting RIS (STAR-RIS)-assisted radar coexistence system where a STAR-RIS is introduced to improve the communication performance while suppressing the mutual interference and providing full space coverage. Based on the realistic conditions of correlated fading, and the presence of multiple user equipments (UEs) at both sides of the RIS, we derive the achievable rates at the radar and the communication receiver side in closed forms in terms of statistical channel state information (CSI). Next, we perform alternating optimization (AO) for optimizing the STAR-RIS and the radar beamforming. Regarding the former, we optimize the amplitudes and phase shifts of the STAR-RIS through a projected gradient ascent algorithm (PGAM) simultaneously with respect to the amplitudes and phase shifts of the surface for both energy splitting (ES) and mode switching (MS) operation protocols. The proposed optimization saves enough overhead since it can be performed every several coherence intervals. This property is particularly beneficial compared to reflecting-only RIS because a STAR-RIS includes the double number of variables, which require increased overhead. Finally, simulation results illustrate how the proposed architecture outperforms the conventional RIS counterpart, and show how the various parameters affect the performance. Moreover, a benchmark full instantaneous CSI (I-CSI) based design is provided and shown to result in higher sum-rate but also in large overhead associated with complexity.

翻译：集成感知与通信（ISAC）有望在未来无线通信的新兴技术中发挥重要作用。特别地，通信雷达共存系统会受到相互干扰的显著恶化。鉴于可重构智能超表面（RIS）的优势，本文提出了一种同时透射与反射RIS（STAR-RIS）辅助的雷达共存系统，通过引入STAR-RIS在抑制相互干扰并提供全空间覆盖的同时提升通信性能。基于相关衰落以及RIS两侧存在多用户设备（UE）的现实条件，我们推导了雷达与通信接收端可达速率的统计信道状态信息（CSI）闭合表达式。随后，我们采用交替优化（AO）方法优化STAR-RIS与雷达波束赋形。针对前者，我们通过投影梯度上升算法（PGAM）同时对STAR-RIS的幅度与相移进行优化，以适用于能量分裂（ES）与模式切换（MS）两种操作协议。所提优化方案可每隔数个相干间隔执行一次，从而节省了大量开销。相较于仅反射型RIS，这一特性尤为有益，因为STAR-RIS的变量数量翻倍，所需开销更高。最后，仿真结果展示了所提架构相较于传统RIS的性能优势，并揭示了各参数对性能的影响。此外，我们提供了基于全瞬时CSI（I-CSI）的基准设计方案，该方案虽能获得更高的总速率，但伴随较高的复杂性与开销。