Joint Precoding and Reflective Beamforming for RIS-Assisted ISCC Networks: A Cross-Layer Optimization Approach

Abstract

Integrated sensing, communication, and computing (ISCC) technology significantly improves spectrum efficiency and reduces hardware costs by unifying the functionalities of sensing, communication, and computation. However, the degradation of wireless link quality caused by obstacles may lead to severe offloading latency. This paper investigates the application of reconfigurable intelligent surface (RIS) technology in ISCC network to enhance the reliability of wireless links and improve the efficiency of computation offloading. In the proposed network, integrated sensing and communication (ISAC) devices employ the same hardware and signaling mechanisms to perform both sensing and communication tasks, while mobile edge computing (MEC) technology is leveraged to process sensing data. To address the cross-layer resource management problem, we formulate a total latency minimization problem for both communication and computation under sensing accuracy constraints, by jointly optimizing the waveform precoding design matrix of ISAC devices, the beam pattern scaling factor, the RIS’s reflective beamforming, and the computing frequency of edge server (ES). Since this problem is highly non-convex, we propose an iterative algorithm based on block coordinate descent (BCD) framework, which leverages semi-definite relaxation (SDR) method, the Charnes-Cooper transform (CCT) method, and closed-form solution to alternately optimize three variable blocks until convergence is achieved for the final solution. Extensive simulations validate the effectiveness of the proposed scheme, demonstrating that the RIS-assisted joint optimization scheme significantly reduces the total system latency. Moreover, we reveal the trade-off between sensing accuracy and system latency.