Spatial PDE-aware Selective State-space with Nested Memory for Mobile Traffic Grid Forecasting

• URL: http://arxiv.org/abs/2603.12353v1
• Date: Thu, 12 Mar 2026 18:17:57 GMT
• Title: Spatial PDE-aware Selective State-space with Nested Memory for Mobile Traffic Grid Forecasting
• Abstract summary: We propose NeST-S6, a convolutional selective state-space model (SSM) with a spatial PDE-aware core, implemented in a nested learning paradigm.<n>NeST-S6 speeds full-grid reconstruction by 32 times and reduces MAC resolutions by 4.3 times compared to competitive per-pixel scanning models.
• Score: 0.81996963503528
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Traffic forecasting in cellular networks is a challenging spatiotemporal prediction problem due to strong temporal dependencies, spatial heterogeneity across cells, and the need for scalability to large network deployments. Traditional cell-specific models incur prohibitive training and maintenance costs, while global models often fail to capture heterogeneous spatial dynamics. Recent spatiotemporal architectures based on attention or graph neural networks improve accuracy but introduce high computational overhead, limiting their applicability in large-scale or real-time settings. We study spatiotemporal grid forecasting, where each time step is a 2D lattice of traffic values, and predict the next grid patch using previous patches. We propose NeST-S6, a convolutional selective state-space model (SSM) with a spatial PDE-aware core, implemented in a nested learning paradigm: convolutional local spatial mixing feeds a spatial PDE-aware SSM core, while a nested-learning long-term memory is updated by a learned optimizer when one-step prediction errors indicate unmodeled dynamics. On the mobile-traffic grid (Milan dataset) at three resolutions (202, 502, 1002), NeST-S6 attains lower errors than a strong Mamba-family baseline in both single-step and 6-step autoregressive rollouts. Under drift stress tests, our model's nested memory lowers MAE by 48-65% over a no-memory ablation. NeST-S6 also speeds full-grid reconstruction by 32 times and reduces MACs by 4.3 times compared to competitive per-pixel scanning models, while achieving 61% lower per-pixel RMSE.

Related papers

• GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis Mamba [6.3881766286624115]
  We introduce GAMMA-Net, a novel approach that integrates Graph Attention Networks (GAT) with Selective State Space Models (Mamba)<n>GAT uses a self-attention mechanism to adjust the influence of nodes within the traffic network, enabling adaptive spatial dependency modeling.<n>Mamba module efficiently models long-term temporal and spatial dynamics without the heavy computational cost.
  arXiv  Detail & Related papers  (2026-04-18T06:14:34Z)
• LiQSS: Post-Transformer Linear Quantum-Inspired State-Space Tensor Networks for Real-Time 6G [85.58816960936069]
  Proactive and agentic control in Sixth-Generation (6G) Open Radio Access Networks (O-RAN) requires control-grade prediction under stringent Near-Time (Near-RT) latency and computational constraints.<n>This paper investigates a post-Transformer paradigm for efficient radio telemetry forecasting.<n>We propose a quantum-inspired state-space tensor network that replaces self-attention with stable structured state-space dynamics kernels.
  arXiv  Detail & Related papers  (2026-01-18T12:08:38Z)
• SpikySpace: A Spiking State Space Model for Energy-Efficient Time Series Forecasting [9.976522013586244]
  SpikySpace is a spiking state-space model that reduces the quadratic cost in the attention block to linear time via selective scanning.<n>Because complex operations such as exponentials and divisions are costly on neuromorphic chips, we introduce simplified approximations of SiLU and Softplus.<n>In matched settings, SpikySpace reduces estimated energy consumption by 98.73% and 96.24% compared to two state-of-the-art transformer based approaches.
  arXiv  Detail & Related papers  (2026-01-02T13:10:53Z)
• A lightweight Spatial-Temporal Graph Neural Network for Long-term Time Series Forecasting [0.0]
  Lite-STGNN is a lightweight spatial-temporal graph neural network for long-term multivariate forecasting.<n>It integrates decomposition-based temporal modeling with learnable sparse graph structure.<n> Lite-STGNN achieves state-of-the-art accuracy on four benchmark datasets for horizons up to 720 steps.
  arXiv  Detail & Related papers  (2025-12-19T11:12:20Z)
• Wireless Traffic Prediction with Large Language Model [54.07581399989292]
  TIDES is a novel framework that captures spatial-temporal correlations for wireless traffic prediction.<n> TIDES achieves efficient adaptation to domain-specific patterns without incurring excessive training overhead.<n>Our results indicate that integrating spatial awareness into LLM-based predictors is the key to unlocking scalable and intelligent network management in future 6G systems.
  arXiv  Detail & Related papers  (2025-12-19T04:47:40Z)
• Agentic World Modeling for 6G: Near-Real-Time Generative State-Space Reasoning [70.56067503630486]
  We argue that sixth-generation (6G) intelligence is not fluent token prediction but calibrated the capacity to imagine and choose.<n>We show that WM-MS3M cuts mean absolute error (MAE) by 1.69% versus MS3M with 32% fewer parameters and similar latency, and achieves 35-80% lower root mean squared error (RMSE) than attention/hybrid baselines with 2.3-4.1x faster inference.
  arXiv  Detail & Related papers  (2025-11-04T17:22:22Z)
• HiSTM: Hierarchical Spatiotemporal Mamba for Cellular Traffic Forecasting [0.7111641404908191]
  We present Hierarchical SpatioTemporal Mamba (HiSTM), which combines a dual spatial encoder with a Mamba-based temporal module and attention mechanism.<n>HiSTM employs selective state space methods to capture spatial and temporal patterns in network traffic.<n>We show that the HiSTM generalizes well across different datasets and improves in accuracy over longer time-horizons.
  arXiv  Detail & Related papers  (2025-08-07T14:18:18Z)
• Enhancing Spatiotemporal Networks with xLSTM: A Scalar LSTM Approach for Cellular Traffic Forecasting [0.7111641404908191]
  We introduce a lightweight, dual-path Spatiotemporal Network that leverages a gradientr LSTM for efficient modeling and a three-layer Conv3D module for spatial feature extraction.<n>We show superior forecast performance over ConvLSTM baselines and strong generalization to unseen regions, making it well-suited for large-scale next-generation network deployments.
  arXiv  Detail & Related papers  (2025-07-17T22:48:46Z)
• Improving Traffic Flow Predictions with SGCN-LSTM: A Hybrid Model for Spatial and Temporal Dependencies [55.2480439325792]
  This paper introduces the Signal-Enhanced Graph Convolutional Network Long Short Term Memory (SGCN-LSTM) model for predicting traffic speeds across road networks. Experiments on the PEMS-BAY road network traffic dataset demonstrate the SGCN-LSTM model's effectiveness.
  arXiv  Detail & Related papers  (2024-11-01T00:37:00Z)
• Hyperdimensional Computing Empowered Federated Foundation Model over Wireless Networks for Metaverse [56.384390765357004]
  We propose an integrated federated split learning and hyperdimensional computing framework for emerging foundation models. This novel approach reduces communication costs, computation load, and privacy risks, making it suitable for resource-constrained edge devices in the Metaverse.
  arXiv  Detail & Related papers  (2024-08-26T17:03:14Z)
• Multiscale Spatio-Temporal Graph Neural Networks for 3D Skeleton-Based Motion Prediction [92.16318571149553]
  We propose a multiscale-temporal graph neural network (MST-GNN) to predict the future 3D-based skeleton human poses. The MST-GNN outperforms state-of-the-art methods in both short and long-term motion prediction.
  arXiv  Detail & Related papers  (2021-08-25T14:05:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.