Skip to content
Open access

A routing-signal study of confidence-gated conditional computation for lightweight IoT intrusion detection

Mahmoud Abbasi Javier Prieto Tejedor Carolina Villoria Torres Marta Plaza-Hernández
Jul 2026 · Open Research Europe · Vol 6, pp. 261 · 0 citations · 41 references

TL;DR

Confidence-gated routing gives large, statistically significant gains over the tiny model on every dataset, whereas a learned router improves only marginally over plain MSP, so once an escalation budget is fixed a simple MSP gate suffices.

Abstract

Background Intrusion detection at Internet of Things (IoT) edge gateways must run under tight compute and memory budgets, motivating dual-path designs that classify most flows with a cheap model and escalate only uncertain flows to a heavier one. A prior conference study introduced such a system, in which a tiny multilayer perceptron (MLP) on features compressed by principal component analysis (PCA) is the fast path, gated to a heavy MLP by the maximum softmax probability (MSP). That system recovers near-heavy accuracy while escalating few flows. Methods We extend that work into a systematic study of what signal should drive the escalation gate. On three benchmarks (ACI-IoT-2023, TabularIoTAttacks-2024, and the large-scale CICIoT2023), we compare routing signals (MSP, predictive entropy, margin, energy, temperature-scaled confidence, latent norm) against a learned router and an oracle bound at matched cost; study temperature-scaling calibration and a leakage-free threshold rule; test a tiny-ensemble disagreement signal; and evaluate robustness, latency, model size, statistical significance, and cross-dataset transfer. Results Confidence-gated routing gives large, statistically significant gains over the tiny model on every dataset, whereas a learned router improves only marginally over plain MSP, so once an escalation budget is fixed a simple MSP gate suffices. On CICIoT2023 we observe an oracle gap. Routing cannot reach the heavy model even at large budgets, because the tiny model is confidently wrong on a few high-volume attack families such as denial-of-service (DoS) floods that are never escalated. Neither ensembling nor confidence-independent signals close the gap. A fixed threshold does not transfer across datasets, its escalation rate swinging from 1 % to 55 % . Conclusions The efficiency benefit grows with a stronger heavy backbone, including a gradient-boosted-tree heavy. Gates should be set by a budget, and closing the oracle gap requires identifying confident errors within high-confidence classes rather than a better routing signal.

Read PDF

Similar papers

Open access Aug 2026

XAI-Driven Intrusion Detection for Internet of Things Networks

A leakage-aware and explainability-informed framework that can be used to judge when ensemble complexity yields genuine predictive and practical benefits for the detection of IoT intrusion is developed.

A. Alqahtani, Fatimah Alakeel, Lujain Abuhaimed · 0 citations
Conference Open access Aug 2026

An Early-Exit Hybrid Intrusion Detection Architecture for Resource-Constrained Iot Edge Devices

The outcomes of the experimental work with the proposed Lightweight NIDS prove that the proposed system can attain competitive detection accuracy and meeting strict embedded performance constraints, providing an effective and edge computing security solution, which is scalable.

Sushant Kumar, V. Musale · 0 citations
Open access Sep 2026

Operational Internet of Things Intrusion Detection Under Domain Shift with a Lipschitz-Constrained Markov Decision Process Decoder

Threshold transfer yields a deployment-faithful, calibrated, and auditable evaluation of IoT intrusion detection under domain shift, exposing a realistic generalization gap that global ranking metrics alone conceal.

Mustafa İpekbayrak, Zeynep Gurkas-Aydin · 0 citations
Open access Aug 2026

Sequential submodular feature-sample selection for lightweight and update-efficient IoT intrusion detection

The Sequential Submodular Feature-Sample Selection (SSFSS) framework is proposed, a theoretically grounded approach that sequentially optimizes the feature and sample spaces to reduce training cost while preserving detection fidelity, and is positioned as an efficient update-aware preprocessing framework whose measured...

Mohammed Nagah Amr, Ahmed S. Elliethy, T. Mekkawy et al. · 0 citations
Open access 2026

Correlation-Based Leakage Screening is Insufficient for IoT/IIoT Intrusion-Detection Benchmarks: Evidence of Redundant Shortcut Encoding in Edge-IIoTset

Machine-learning intrusion detection for the Internet of Things (IoT) and Industrial IoT (IIoT) routinely reports near-perfect accuracy on public benchmarks, yet the feature-level basis of that performance is rarely interrogated. This study benchmarks four supervised classifiers, random forest, gradient boosting, logis...

F. Dhaou · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.