🤖 AI Summary
This work addresses the challenge of balancing high pilot overhead, computational complexity, and spectral efficiency in downlink multiuser transmission with dynamic metasurface antennas (DMAs). To this end, a two-timescale transmission design is proposed: at the beginning of each frame, DMA coefficients are optimized using statistical channel state information (CSI), while within each time slot, low-dimensional instantaneous CSI is leveraged to update the digital precoder. The paper introduces a novel optimization framework for DMA coefficients that outperforms random successive convex approximation and derives a closed-form solution for digital precoding in the single-user case. By integrating the weighted minimum mean square error method with the DMA architecture, the proposed scheme significantly reduces pilot overhead and computational complexity while maintaining high spectral efficiency, making it well-suited for future wireless communication systems.
📝 Abstract
Dynamic metasurface antennas (DMAs) promise to relieve massive multiple-input multiple-output architectures from their high energy consumption and hardware costs. This paper proposes a two-timescale design for downlink multiuser transmission via DMAs, a design that balances pilot overhead, complexity, and spectral efficiency. At the onset of each frame, the DMA coefficients are configured based only on statistical channel-state information (CSI), a process for which the paper introduces an optimization framework that is shown to outperform the widely used stochastic successive convex approximation method. Then, within each frame, the digital precoder is updated at each slot, based on the optimized DMA coefficients and the effective lower-dimensional instantaneous CSI. The weighted minimum mean-squared error method is applied for this short-term optimization and, for the special case of single-user transmission, a closed-form solution for the digital precoder is provided. Performance evaluations demonstrate that the proposed two-timescale design can be an attractive ingredient for future wireless networks.