Meta-Learning Based Optimization for Large Scale Wireless Systems

TL;DR

The paper tackles the intractable optimization of sum-rate in large-scale wireless systems where conventional non-convex algorithms scale poorly with the number of antennas and users. It introduces an unsupervised meta-learning framework that uses multiple dedicated DNNs to predict gradient updates, enabling direct non-convex optimization with significantly reduced complexity. The approach is validated on three challenging 6G use cases—H-RSMA with imperfect CSIT, ISAC, and BD-RIS—demonstrating substantial performance gains over traditional methods and offering new insights into scalable operation. The results suggest that meta-learning can unlock efficient optimization for large-scale deployments, enabling realistic evaluation and design of next-generation wireless networks.

Abstract

Optimization algorithms for wireless systems play a fundamental role in improving their performance and efficiency. However, it is known that the complexity of conventional optimization algorithms in the literature often exponentially increases with the number of transmit antennas and communication users in the wireless system. Therefore, in the large scale regime, the astronomically large complexity of these optimization algorithms prohibits their use and prevents assessing large scale wireless systems performance under optimized conditions. To overcome this limitation, this work proposes instead the use of an unsupervised meta-learning based approach to directly perform non-convex optimization at significantly reduced complexity. To demonstrate the effectiveness of the proposed meta-learning based solution, the sum-rate (SR) maximization problem for the following three emerging 6G technologies is contemplated: hierarchical rate-splitting multiple access (H-RSMA), integrated sensing and communication (ISAC), and beyond-diagonal reconfigurable intelligent surfaces (BD-RIS). Through numerical results, it is demonstrated that the proposed meta-learning based optimization framework is able to successfully optimize the performance and also reveal unknown aspects of the operation in the large scale regime for the considered three 6G technologies.

Figures (11)

• Figure 1: General structure of the meta-learning framework.
• Figure 2: Proposed meta-learning based optimization algorithm structure.
• Figure 3: H-RSMA system model.
• Figure 4: ESR vs. SNR ($N_t = 100, K=90$): (a) disjoint user groups ($\Delta_k=\frac{\pi}{36}$), and (b) overlapping user groups ($\Delta_k=\frac{\pi}{8}$).
• Figure 5: ISAC system model.