Rivaling Transformers: Multi-Scale Structured State-Space Mixtures for Agentic 6G O-RAN

2025-10-06

AI Summary

To address the challenge of short-term, user-level KPI forecasting for near-real-time intelligent control in 6G O-RAN—under multi-timescale dynamics and resource constraints—this paper proposes MS³M, a lightweight Multi-Scale Structured State-Space Mixture model. MS³M innovatively integrates the HiPPO-LegS kernel with bilinear discretization to capture complex temporal dynamics, while incorporating squeeze-and-excitation gating and channel shuffling to enhance expressiveness with minimal overhead. Coupled with depthwise separable convolutions and a sliding-window training strategy, it enables low-latency time-series modeling. Evaluated on a custom-built O-RAN testbed, MS³M achieves only 0.7M parameters and an inference latency of 57 ms—3–10× faster than Transformer baselines—while maintaining competitive prediction accuracy.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In sixth-generation (6G) Open Radio Access Networks (O-RAN), proactive control is preferable. A key open challenge is delivering control-grade predictions within Near-Real-Time (Near-RT) latency and computational constraints under multi-timescale dynamics. We therefore cast RAN Intelligent Controller (RIC) analytics as an agentic perceive-predict xApp that turns noisy, multivariate RAN telemetry into short-horizon per-User Equipment (UE) key performance indicator (KPI) forecasts to drive anticipatory control. In this regard, Transformers are powerful for sequence learning and time-series forecasting, but they are memory-intensive, which limits Near-RT RIC use. Therefore, we need models that maintain accuracy while reducing latency and data movement. To this end, we propose a lightweight Multi-Scale Structured State-Space Mixtures (MS3M) forecaster that mixes HiPPO-LegS kernels to capture multi-timescale radio dynamics. We develop stable discrete state-space models (SSMs) via bilinear (Tustin) discretization and apply their causal impulse responses as per-feature depthwise convolutions. Squeeze-and-Excitation gating dynamically reweights KPI channels as conditions change, and a compact gated channel-mixing layer models cross-feature nonlinearities without Transformer-level cost. The model is KPI-agnostic -- Reference Signal Received Power (RSRP) serves as a canonical use case -- and is trained on sliding windows to predict the immediate next step. Empirical evaluations conducted using our bespoke O-RAN testbed KPI time-series dataset (59,441 windows across 13 KPIs). Crucially for O-RAN constraints, MS3M achieves a 0.057 s per-inference latency with 0.70M parameters, yielding 3-10x lower latency than the Transformer baselines evaluated on the same hardware, while maintaining competitive accuracy.

Problem

• Proactive control in 6G O-RAN requires low-latency multi-timescale predictions
• Transformers are too memory-intensive for Near-RT RAN Intelligent Controllers
• Models must maintain accuracy while reducing latency and computational costs

Innovation

• Multi-scale state-space mixtures capture radio dynamics
• Bilinear discretization enables stable sequence modeling
• Gated mechanisms replace Transformers to reduce latency