【摘要】 针对室内可见光通信（VLC）视距链路存在遮挡场景，提出一种基于智能反射面（IRS）辅助的VLC系统设计方案。首先，基于室内VLC系统模型，引入IRS改善的非视距链路，有效补偿被遮挡链路的信道衰减；随后，对单/多墙面分布式部署场景进行优化设计。在单墙面场景下，以信道增益最大化为目标，构建“反射单元尺寸+数量”的联合优化目标函数，并利用序列二次规划算法求解单墙面场景下的IRS关键参数；在多墙面分布式部署场景下，以可达数据速率最大化为目标，利用混合粒子群算法对IRS阵列数量及部署墙面数等参数配置进行优化。仿真结果表明，针对单墙面场景，在保持1×10^-3误符号率的条件下，联合优化后的方案相较于未优化和单独优化的方案分别节省了2.3 dB和1.1 dB信噪比；针对多墙面场景，在相同发送功率下，联合优化方案的可达速率较未优化方案提升23%，并且当目标速率为1.5×10⁴ bps时联合优化方案可获得8 dB的信噪比增益。所设计系统能够显著提升遮挡场景下的通信性能，为IRS辅助VLC的工程部署提供理论依据。更多还原

【Abstract】 To address the occlusion scenario of the line-of-sight link in indoor visible light communication（VLC） systems, this study proposes a system design scheme for VLC assisted by an intelligent reflecting surface（IRS）. First, based on an indoor VLC system model, an IRS is introduced to improve the non-line-of-sight link, which effectively compensates for the channel attenuation of the occluded link. Subsequently, both single-wall and multiwall distributed deployment scenarios are designed. For the single-wall scenario, a joint optimization objective function of “reflective unit size+number of units” is constructed to maximize channel gain, and the sequential quadratic programming algorithm is used to solve the key parameters of the IRS. For the multiwall distributed deployment scenario, to maximize the achievable data rate, the hybrid particle swarm algorithm is adopted to optimize the parameter configuration, such as the numbers of IRS arrays and deployed walls. Simulation results show that in the single-wall scenario, to maintain a symbol error rate of 1×10^-3, the proposed joint optimization scheme reduces the required signal-to-noise ratio（SNR） by 2. 3 dB and 1. 1 dB compared with the unoptimized and separately optimized schemes, respectively. In the multiwall scenario, under the same transmission power, the achievable rate of the joint optimization scheme is 23% higher than that of the unoptimized scheme, and when the target rate is 1. 5×10⁴ bps, the joint optimization scheme can achieve an SNR gain of 8 dB. The proposed scheme significantly improves the communication performance of the VLC system in the occlusion scenario and provides a theoretical basis for the engineering deployment of IRS-assisted VLC.更多还原