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 propose HiSTM (Hierarchical SpatioTemporal Mamba), a spatiotemporal forecasting architecture built on state-space modeling. HiSTM combines spatial convolutional encoding for local neighborhood interactions with Mamba-based temporal modeling to capture long-range dependencies, followed by attention-based temporal aggregation for prediction. The hierarchical design enables representation learning with linear computational complexity in sequence length and supports both grid-based and correlation-defined spatial structures. Cluster-aware extensions incorporate spatial regime information to handle heterogeneous traffic patterns. Experimental evaluation on large-scale real-world cellular datasets demonstrates that HiSTM achieves better accuracy, outperforming strong baselines. On the Milan dataset, HiSTM reduces MAE by 29.4% compared to STN, while achieving the lowest RMSE and highest R2 score among all evaluated models. In multi-step autoregressive forecasting, HiSTM maintains 36.8% lower MAE than STN and 11.3% lower than STTRE at the 6-step horizon, with a 58% slower error accumulation rate compared to STN. On the unseen Trentino dataset, HiSTM achieves 47.3% MAE reduction over STN and demonstrates better cross-dataset generalization. A single HiSTM model outperforms 10,000 independently trained cell-specific LSTMs, demonstrating the advantage of joint spatiotemporal learning. HiSTM maintains best-in-class performance with up to 30% missing data, outperforming all baselines under various missing data scenarios. The model achieves these results while being 45 & times; smaller than PredRNNpp, 18 & times; smaller than xLSTM, and maintaining competitive inference latency of 1.19 ms, showcasing its effectiveness for scalable 5/6G traffic prediction in resource-constrained environments.

Modern vehicles increasingly rely on Controller Area Networks (CAN), making them vulnerable to cyber-physical attacks such as denial of service (DoS) and fuzzy attacks. Real-time detection of such attacks is challenging due to the high message frequency, strict timing constraints, and lack of message authentication in CAN. This paper presents a cost-effective, two-stage statistical learning method for real-time intrusion detection in CAN networks. During training, times-tamp differences between consecutive message IDs are stored in a unidirectional dictionary, while bidirectional relationships are captured in an adjacency matrix. In the detection phase, incoming messages are checked against learned time bounds or, if missing, classified using a backward adjacency scan. The method is evaluated with two realistic datasets, Car-Hacking and Survival Analysis, containing both DoS and fuzzy scenarios. Results show high accuracy, precision, recall, and F1-score, with sub-millisecond per-message latency, meeting the stringent real-time requirements of CAN systems. These results confirm the effectiveness and efficiency of the proposed method for in-vehicle intrusion detection. © 2025 IEEE.

Time-Sensitive Networking (TSN) is a toolbox of technologies that enable deterministic communication over Ethernet. A key area has been TSN's time-aware traffic shaping (TAS), which supports stringent end-to-end latency and reliability requirements. Configuration of TAS requires the computation of a network-wide traffic schedule, which is particularly challenging with integrated wireless networks (e.g., 5G, Wi-Fi) due to the stochastic nature of wireless links. This paper introduces a novel method for configuring TAS, focusing on cyclic traffic patterns and jitter of wireless links. We formulate a linear program that computes a network-wide time-aware schedule, robust to wireless performance uncertainties. The given method enables robust scheduling of multiple TSN frames per transmission window using a tunable robustness parameter ({\Gamma}). To reduce computational complexity, we also propose a sequential batch-scheduling heuristic that runs in polynomial time. Our approach is evaluated by using different network topologies and wireless link characteristics, demonstrating that the heuristic can schedule 90% of 6500 requested TSN streams in a large topology. 

As Fifth Generation Networks (5G) networks evolve, operators have an opportunity to create high-quality customer experiences by ensuring seamless and personalized connectivity through consumer-oriented network slicing. However, effectively managing consumer slices remains a challenge due to dynamic user demands, mobility patterns, and the need for real-time Quality of Service (QoS) assurance. This paper presents an Artificial Intelligence (AI)-driven framework for intelligent service feasibility, qualification, and provisioning in 5G networks, to automate consumer slicing. By leveraging real-time data from the Radio Access Network (RAN) and Core Network (CN), along with Service Quality Indicator (SQI), the framework enables dynamic resource allocation, predictive service qualification, and proactive slice optimization. Key innovations of the frame-work include the coherent integration of AI-driven slice load prediction, mobility-aware service provisioning, and automated QoS assurance via Application Programming Interfaces (APIs), ensuring optimal performance for consumer applications while maintaining compliance with Service Level Agreement (SLA). Experimental results demonstrate the framework's capabilities in enhancing user experiences by enabling personalized slice selection, reducing service disruptions, and optimizing network resource utilization. 

Real-time traffic monitoring is critical for network operators to ensure performance, security, and visibility - especially as encryption becomes the norm. AI and ML have emerged as powerful tools to create deeper insights from network traffic, but collecting the fine-grained features needed at terabit speeds remains a major bottleneck. We introduce Direct Feature Access (DFA): a high-speed telemetry system that extracts flow features at line rate using P4-programmable data planes, and delivers them directly to GPUs via RDMA and GPUDirect - completely bypassing the ML server's CPU. DFA enables feature enrichment and immediate inference on GPUs, eliminating traditional control plane bottlenecks and dramatically reducing latency. We implement DFA on Intel Tofino switches and NVIDIA A100 GPUs, achieving extraction and delivery of over 31 million feature vectors per second - supporting 524,000 flows within sub-20 ms monitoring periods - on a single port. DFA unlocks scalable, real-time, ML-driven traffic analysis at terabit speeds, pushing the frontier of what is possible for next-generation network monitoring. 

Accurate spatiotemporal traffic forecasting is vital for intelligent resource management in 5G and beyond. However, conventional AI approaches often fail to capture the intricate spatial and temporal patterns that exist, due to e.g., the mobility of users. We introduce a lightweight, dual-path Spatiotemporal Network that leverages a Scalar LSTM (sLSTM) for efficient temporal modeling and a three-layer Conv3D module for spatial feature extraction. A fusion layer integrates both streams into a cohesive representation, enabling robust forecasting. Our design improves gradient stability and convergence speed while reducing prediction error. Evaluations on real-world datasets show superior forecast performance over ConvLSTM baselines and strong generalization to unseen regions, making it well-suited for large-scale, next-generation network deployments. Experimental evaluation shows a 23% MAE reduction over ConvLSTM, with a 30% improvement in model generalization. 

Accurate spatiotemporal traffic forecasting is vital for optimizing 5G networks. Traditional LSTM models struggle with capturing complex spatiotemporal dependencies, limiting predictive performance. To address this, we propose an enhanced Spatiotemporal Network (STN) integrating Scalar LSTM (sLSTM), a more efficient variant designed to improve temporal modeling while reducing computational complexity. Our dualpath STN processes the input through an sLSTM for sequential feature extraction and a three-layer Conv3D path for spatial feature learning, with both outputs fused in a dedicated fusion layer for enhanced spatiotemporal representation. By incorporating sLSTM, our model stabilizes gradients, accelerates convergence, and enhances accuracy. Experiments on real-world mobile traffic datasets show a 23% MAE reduction over ConvLSTM, with a 30% improvement on unseen data, demonstrating superior generalization for 5G traffic prediction. 

Cellular traffic forecasting is essential for networkplanning, resource allocation, or load-balancing traffic acrosscells. However, accurate forecasting is difficult due to intricatespatial and temporal patterns that exist due to the mobility ofusers. Existing AI-based traffic forecasting models often trade-offaccuracy and computational efficiency. We present HierarchicalSpatioTemporal Mamba (HiSTM), which combines a dual spatialencoder with a Mamba-based temporal module and attentionmechanism. HiSTM employs selective state space methods tocapture spatial and temporal patterns in network traffic. Inour evaluation, we use a real-world dataset to compare HiSTMagainst several baselines, showing a 29.4% MAE improvementover the STN baseline while using 94% fewer parameters. Weshow that the HiSTM generalizes well across different datasetsand improves in accuracy over longer time-horizons.Index Terms—Time series forecasting, spatiotemporal model-ing, 5G network traffic prediction, deep learning, state spacemodels, Mamba architecture, attention mechanisms, convolu-tional neural networks (CNNs), hierarchical modeling, AI fortelecommunications.