Skip to content
Open access

Joint Movable-Antenna and Movable-Signal Design for Downlink Multiuser MISO Communications

2026 · IEEE Transactions on Wireless Communications · Vol 25, pp. 25827-25843 · 0 citations · 38 references

Abstract

We investigate a downlink multiuser MISO system where a base station controls the wireless channel through movable antennas (MAs) sliding on a one-dimensional rail and a movable signal (MS) realized by block-level carrier-frequency tuning. The MS is not used as simultaneous carrier aggregation; instead, one licensed carrier state is selected for a transmission block after RF settling and channel acquisition. Under a two-path channel model, each per-antenna coefficient is a sum of two complex exponentials whose phases depend jointly on antenna positions and carrier frequency through space-frequency products. We formulate a joint max-min fair (MMF) design that optimizes MA positions, carrier frequency, and linear precoding under rail-length, spacing, frequency, and power constraints. For the single-user case, we derive an MRT SNR upper bound and show that the carrier-position visibility of near-coherent two-path combining grows with the product of rail length and tuning range, yielding a space-frequency coverage law. For the multiuser case, we analyze the Gram matrix of normalized steering vectors, obtain a Gershgorin-based lower bound on its minimum eigenvalue in terms of inter-user correlation, and translate it into an explicit ZF-MMF rate lower bound governed by worst-user effective gain and inter-user correlation. Guided by these properties, we propose theory-driven alternating optimization and proximal block-SCA algorithms. Each geometry trial is re-evaluated after the corresponding digital-precoder update and is accepted only when an exact-MMF sufficient-increase test is satisfied. We prove that the accepted exact-MMF sequence is monotone non-decreasing, converges, and has a vanishing first-order stationarity gap up to the inner-solver accuracy. We further clarify the practical time scales, multipath extension, channel-estimation robustness, and resource-matched baseline settings. Simulations corroborate the predicted coverage/orthogonality trends and demonstrate MMF gains over fixed-ULA, MA-only, and MS-only baselines under the same antenna, power, rail-length, and frequency-window budgets. Additional convergence and CSI-perturbation experiments verify the monotone accepted-MMF behavior and the predicted robustness gap.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.