In ubiquitous networks, sensing and compute resources are distributed across numerous heterogeneous nodes. Consequently, fixed Intrusion Detection Systems (IDS) deployments either lead to resource wastage or leave security gaps. LAVA-Control addresses this challenge by treating intelligence density, which is the number and identity of nodes executing detection tasks-as a first-class control variable. Operating atop a trained machine learning detector, it employs policy-driven bounds, particle swarm optimization (PSO)-based subset selection, and cooldown timers to adaptively select node subsets, while maintaining constraints on latency, energy consumption, and residual risk. To ensure system robustness, LAVA-Control introduces explicit stability and oscillation metrics that quantify the smoothness of system responses under varying traffic and attack scenarios. Empirical evaluations on vehicular and Distributed Denial-of-Service (DDoS) datasets demonstrate that LAVA-Control achieves near-baseline detection accuracy at a reduced operational cost compared to heuristic baselines, and exhibits substantially greater stability in its adaptations than reinforcement-learning-based control mechanisms.
Kalibbala Jonathan Mukisa, Jae Min Lee, Dong-Seong Kim· International Conference on...· 0 citations
The advancement of military internet of things (IoT) surveillance demands real-time casualty detection across distributed camera networks under diverse environmental conditions. Traditional surveillance systems suffer from high latency, bandwidth inefficiency and unreliable cross-camera identity tracking, indicating the need for advanced detection and tracking systems. This study presents an edge-based multicamera casualty detection and tracking system for military IoT networks. The proposed framework, called CCTV-TrackNet, integrates lightweight AI on distributed camera nodes built on Raspberry Pi Zero2W hardware with high-resolution cameras and sensors such as GPS and IMU. At each node, YOLOv12n performs real-time person and casualty detection, DeepSORT maintains short-term continuity, and OSNet-based Re-ID extracts appearance embeddings for cross-camera association. A central server aggregates metadata for global identification, visualizes trajectories, and generates real-time alerts. Experimental results show 90.77% detection accuracy, 37% and 36% reduction in false cases, 30.3 FPS edge performance, and 84.20% cross-camera ID consistency.
Sium Bin Noor, M. A. Dini, Jaemin Lee et al.· International Conference on...· 0 citations
Accurate forecasting of building energy consumption is a cornerstone of modern smart grid management and sustainable facility operations. However, standard deep learning approaches—such as Long Short-Term Memory (LSTM) networks and Transformers—often function as black boxes, failing to explicitly model the governing physical laws and thermodynamic constraints of building systems. This limitation frequently results in poor generalization and instability when applied to diverse building portfolios. To address this challenge, we propose a novel Rational-Aware Architecture, a heterogeneous Mixtureof-Experts (MoE) framework that decomposes the forecasting task into specialized semantic agents. The architecture comprises a Thermodynamicist (Physics-ResNet) to model enthalpy and heat transfer, a Meteorologist (WeatherCNN) to capture environmental gradients, a Manager (Time2Vec) to encode temporal cyclicities, and an Engineer (Sequence Model) to handle historical load inertia. A context-aware gating mechanism dynamically weighs these experts based on specific building characteristics. Extensive experiments on a large-scale dataset of 800 buildings from the ASHRAE Great Energy Predictor III challenge demonstrate that the proposed framework significantly outperforms standard deep learning baselines. The Rational-Aware Transformer achieved a Mean Absolute Percentage Error (MAPE) of $\mathbf{1 9. 9 7 \%}$, representing a relative error reduction of approximately 40% compared to the standard Transformer baseline (33.03%). Furthermore, the Rational-LSTM variant demonstrated exceptional stability with an $R^{2}$ score of 0.9452, effectively mitigating the volatility often observed in pure data-driven approaches. These results confirm that integrating domain knowledge into deep learning architectures yields superior robustness, precision, and interpretability for energy forecasting tasks.
Ali Aouto, Dong-Seong Kim· International Conference on...· 0 citations
AI-driven Internet of Medical Things (IoMT) services increasingly rely on federated learning (FL) for privacy-preserving remote patient monitoring; however, existing frameworks often neglect the longitudinal battery sustainability required for persistent clinical care. This paper presents BattFL, an energy-sustainable federated AI framework that reframes distributed healthcare intelligence as a battery-budgeted service. By integrating explicit per-client battery evolution modeling with computation-communication-security energy decomposition, BattFL supports role-heterogeneous participation (attack-only, health-only, and hybrid) via masked multi-task learning. Furthermore, we incorporate risk-driven adaptive sensing to regulate workload intensity based on predicted clinical risk. Experimental results reveal a pronounced sustainability asymmetry: high-workload attack clients experience up to 41.2% battery depletion within five FL rounds, while health-only clients remain near initial capacity. Across extended horizons $(R=5-60)$, we identify diminishing energy-accuracy returns and participation decay, motivating the need for energy-aware stopping and orchestration mechanisms. By explicitly coupling battery dynamics, security overhead, and federated optimization, BattFL provides a systems-level foundation for secure, resilient, and long-lived AI-driven healthcare services in next-generation ubiquitous IoMT networks.
C. A. Nnadiekwe, S. Ajakwe, J. Isong et al.· International Conference on...· 0 citations
