Symbol-Level Channel Pattern Optimization for Reconfigurable MU-MISO Communication Systems

In this paper, we propose a symbol-level channel pattern optimization (SLCPO) scheme for pattern reconfigurable (PR) multi-user multiple-input single-output (MU-MISO) communication systems. Unlike existing transmission schemes that treat the channel as invariant, this work proposes to actively adjust the radiation power of each scattering path of the wireless channel by adjusting the radiation pattern of the transmit array, thereby removing the limitations that communications can only adapt to the wireless channel passively. We separate the optimization into two stages. In the first stage, an optimization problem on the channel matrix aimed at maximizing the worst user's quality-of service (QoS) without investing additional signal power is formulated. Given the non-convex nature of the formulated problem, we design an alternating optimization algorithm, and obtain the channel covariance matrix that achieves optimal performance for each transmission symbol slot. This structure, proven to be optimal, provides key theoretical insights into the ideal channel conditions for symbol-level precoding (SLP). In the second stage, we propose the radiation pattern design in reconfigurable MU-MISO communication systems, where communication channels are jointly decided by the radiation pattern of the transmit array and the physical propagation channel, aligning the practical pattern channel with the theoretical optimal channel. Numerical results verify the superiority of the proposed SLCPO scheme in PR MU-MISO systems in terms of error rate over state-of-the-art.