Benchmarking Wireless Representations: High-Dimensional vs. Compressed Embeddings for Efficiency and Robustness

2026-05-03

AI Summary

This study systematically evaluates the trade-offs between efficiency and robustness of high-dimensional versus compressed wireless channel representations in practical systems. Focusing on line-of-sight/non-line-of-sight (LoS/NLoS) classification, beam selection, and a newly introduced power allocation task, the work compares end-to-end performance using high-dimensional embeddings generated by a foundation wireless model against low-dimensional representations learned via an autoencoder. The results demonstrate that while high-dimensional embeddings achieve superior performance in few-shot scenarios, they incur substantial computational and transmission overhead. In contrast, compressed representations exhibit significantly better noise resilience, task stability, and resource efficiency. Notably, this work is the first to incorporate power allocation into the representation evaluation framework, clearly revealing the fundamental trade-offs among latency, model parameter count, and robustness across different representation strategies.

Abstract

Building on recent advances in representation learning for wireless channels, this work investigates the cost-benefit trade-offs of high-dimensional channel embeddings in practical systems. We benchmark multiple wireless representations: high-dimensional learned embeddings from a wireless foundation model, compact autoencoder-based representations with significantly lower dimensionality, and raw data baselines, evaluating their performance across diverse downstream tasks. We then systematically analyze data efficiency, noise robustness, and computational complexity, explicitly characterizing the resource overhead associated with high-dimensional embeddings. Beyond standard tasks such as line-of-sight/non-line-of-sight (LoS/NLoS) classification and beam selection, we introduce power allocation as a new downstream task. Our results reveal clear trade-offs: while high-dimensional embeddings can perform well in few-shot regimes for certain tasks, they incur substantial latency and parameter overhead. In contrast, compressed latent representations learned by autoencoders demonstrate improved noise robustness and more stable performance across tasks, while significantly reducing computational and transmission costs.

Problem

• wireless representations
• high-dimensional embeddings
• compressed embeddings
• efficiency
• robustness

Innovation

• wireless representation learning
• high-dimensional embeddings
• compressed embeddings
• autoencoder
• noise robustness