Channel Prediction under Network Distribution Shift Using Continual Learning-based Loss Regularization

2025-09-18

AI Summary

Channel prediction performance for mobile users traversing heterogeneous wireless networks—characterized by varying antenna layouts, carrier frequencies, and scattering properties—degrades significantly due to distribution shift, causing a 37.5% increase in normalized mean square error (NMSE) during handover. Method: This paper proposes a continual learning-based loss regularization framework to mitigate catastrophic forgetting. It innovatively integrates Elastic Weight Consolidation (EWC) and Synaptic Intelligence (SI) for joint parameter importance estimation and regularization, enabling historical knowledge retention and adaptation to new network configurations under constant memory overhead. Results: Evaluated on 3GPP-standardized scenarios and multiple neural network architectures, the approach demonstrates substantial gains: under high SNR conditions, SI reduces NMSE by 1.8 dB (32%–34%), while EWC achieves a 1.4 dB reduction (17%–28%), both markedly outperforming conventional methods.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning/Scheduling and Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Modern wireless networks face critical challenges when mobile users traverse heterogeneous network configurations with varying antenna layouts, carrier frequencies, and scattering statistics. Traditional predictors degrade under distribution shift, with NMSE rising by 37.5% during cross-configuration handovers. This work addresses catastrophic forgetting in channel prediction by proposing a continual learning framework based on loss regularization. The approach augments standard training objectives with penalty terms that selectively preserve network parameters essential for previous configurations while enabling adaptation to new environments. Two prominent regularization strategies are investigated: Elastic Weight Consolidation (EWC) and Synaptic Intelligence (SI). Across 3GPP scenarios and multiple architectures, SI lowers the high-SNR NMSE floor by up to 1.8 dB ($approx$32--34%), while EWC achieves up to 1.4 dB ($approx$17--28%). Notably, standard EWC incurs $mathcal{O}(MK)$ complexity (storing $M$ Fisher diagonal entries and corresponding parameter snapshots across $K$ tasks) unless consolidated, whereas SI maintains $mathcal{O}(M)$ memory complexity (storing $M$ model parameters), independent of task sequence length, making it suitable for resource-constrained wireless infrastructure

Problem

• Predicting wireless channels under network distribution shifts
• Addressing catastrophic forgetting in channel prediction models
• Reducing performance degradation during cross-configuration handovers

Innovation

• Continual learning framework for channel prediction
• Loss regularization to prevent catastrophic forgetting
• Synaptic Intelligence reduces memory complexity