MambaCSP: Hybrid-Attention State Space Models for Hardware-Efficient Channel State Prediction

2026-04-23

AI Summary

This work addresses the high computational overhead, substantial memory consumption, and significant inference latency of conventional Transformers and large language models (LLMs) in channel state information (CSI) prediction, which hinder their applicability in real-time wireless communication systems. To overcome these limitations, the authors propose MambaCSP, the first framework to introduce selective state space models into CSI prediction. Built upon the Mamba backbone, MambaCSP periodically integrates lightweight Patch-Mixer attention layers, enabling effective modeling of long-range dependencies while preserving linear time complexity. Experimental results in MISO-OFDM scenarios demonstrate that, compared to LLM-based approaches, MambaCSP improves prediction accuracy by 9–12%, achieves a 3.0× higher throughput, reduces GPU memory usage by 2.6×, and accelerates inference by 2.9×.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Recent works have demonstrated that attention-based transformer and large language model (LLM) architectures can achieve strong channel state prediction (CSP) performance by capturing long-range temporal dependencies across channel state information (CSI) sequences. However, these models suffer from quadratic scaling in sequence length, leading to substantial computational cost, memory consumption, and inference latency, which limits their applicability in real-time and resource-constrained wireless deployments. In this paper, we investigate whether selective state space models (SSMs) can serve as a hardware-efficient alternative for CSI prediction. We propose MambaCSP, a hybrid-attention SSM architecture that replaces LLM-based prediction backbones with a linear-time Mamba model. To overcome the local-only dependencies of pure SSMs, we introduce lightweight patch-mixer attention layers that periodically inject cross-token attentions, helping with long-context CSI prediction. Extensive MISO-OFDM simulations show that MambaCSP improves prediction accuracy over LLM-based approaches by 9-12%, while delivering up to 3.0x higher throughput, 2.6x lower VRAM usage, and 2.9x faster inference. Our results demonstrate that hybrid state space architectures provide a promising direction for scalable and hardware-efficient AI-native CSI prediction in future wireless networks.

Problem

• channel state prediction
• computational efficiency
• hardware efficiency
• sequence modeling
• wireless communications

Innovation

• State Space Models
• Channel State Prediction
• Mamba
• Hybrid Attention
• Hardware Efficiency