Network automation increasingly requires reproducible workflows that connect deployment with objective post-deployment assurance. This study proposes a six-phase NetDevOps framework integrating Planning, Infrastructure Modeling, Validation, Automated Deployment, Operational Testing, and Operational State Verification. The framework was instantiated using YAML, Jinja2, Ansible, and pyATS and evaluated in two emulated MPLS L3VPN configurations comprising five and ten Cisco devices. Fifteen measured runs per workflow and configuration characterized execution time as well as automation-host processor and memory consumption, while the Operational Compliance Index (OCI) and Global Compliance Index (GCI) quantified operational acceptance. Controlled perturbations examined compliance behavior and baseline-driven state verification. The results showed repeatable execution and stable automation-host resource consumption under the evaluated conditions. OCI and GCI differentiated conformity among evaluable outcomes from incomplete evaluability, while qualified-baseline comparisons exposed structured state differences associated with known interface- and protocol-level perturbations. These findings support the feasibility of integrating automated deployment, quantitative operational acceptance, and baseline-driven state verification within a repeatable NetDevOps lifecycle, without implying comparative superiority, scalability, or multi-vendor portability.
Deep learning-based IoT botnet detection has reached high predictive performance; however, its practical deployment remains constrained by the lack of operationally actionable explainability. This paper introduces TimeAnchor, a hybrid explainability framework that integrates TimeSHAP and Anchors to enhance the temporal stability and internal consistency of explanations while preserving acceptable fidelity. The approach is evaluated on two complementary datasets: CICIDS2017, reflecting controlled and structured traffic conditions, and IoT-23, characterised by real-world, noisy, and highly heterogeneous attack scenarios. The results indicate that TimeAnchor substantially improves explanation consistency and stability across both corpora, achieving high coherence on CICIDS2017 and yielding average gains in stability and consistency on IoT-23 compared with the individual methods. These improvements are accompanied by a moderate trade-off in local fidelity, which appears acceptable given the overall enhancement in explanatory robustness. These findings support the relevance of TimeAnchor for explainable IoT botnet detection in dynamic and realistic operational environments.
Mobile applications that must commit to a data rate—video streaming, guaranteed-rate admission, uplink scheduling—can stall or breach a service-level agreement whenever they promise more than the radio link can deliver. What protects them is not an accurate average-throughput forecast but a trustworthy lower bound on the rate that will actually be available. Conformal prediction (CP) can produce such bounds without distributional assumptions, but mobile 5G throughput is piecewise non-stationary: it jumps at cell handovers and radio-access-technology (RAT) transitions, exactly where a rate decision is most fragile. A natural fix is to condition the CP calibration on the mobility-regime state (RAT and time since the last handover). Using 188,711 s of real commercial 5G traces, we show that whether this conditioning helps is task-dependent. It helps two-sided prediction intervals, whose error scale varies by an order of magnitude across RATs, and it removes the post-handover under-coverage of online adaptive CP at no width cost. But for the one-sided lower bound that actually governs rate selection, a single global CP bound is better: it delivers higher link utilization at the same stall budget, and this advantage is robust to the data’s zero-inflation. The practical lesson for rate-control designers is to match the conditioning to the tail the decision uses, and to judge uncertainty methods by the operational outcome rather than by coverage alone. All results are reproducible from a released pipeline.
A high-altitude platform station (HAPS) can serve an extended motorway corridor from a single aperture, but stratospheric attitude disturbance introduces a pointing error that degrades the isolation between closely spaced beams. Robust user selection addresses this by requiring an explicit angular separation between co-scheduled beams, yet existing formulations are evaluated at a single scheduling instant over compact footprints. This paper asks how long a scheduling decision remains valid once the served vehicles move, and how footprint geometry bounds the number of vehicles that can be co-scheduled. A corridor mobility model is developed; the pairwise separation requirement is enforced by max–min angular dispersion, for which the chance-constrained separation margin acts as a feasibility certificate; and the margin is extended to the age of the decision. Monte Carlo evaluation at 28 GHz with zero-forcing precoding shows that the scheduling geometry evolves about 5×104 times more slowly than the small-scale fading, so that outage is statistically unchanged at a decision age of 100 ms and rises by 30% at 1 s. On the corridor, max–min selection yields lower outage than capped-sum angle-aware selection, channel-aware selection and random selection at every load from K=4 to K=12, at the cost of a 5% lower cell-edge spectral efficiency at intermediate load. Widening the margin to absorb angular drift gives no significant benefit at a realistic candidate-pool size and increases outage 2.7-fold at K=10, so decision age is better bounded through the scheduling interval than compensated through margin inflation.
This study presents a low-cost Internet of Things (IoT)-based antenna-measurement system that integrates automated dual-axis positioning, mobile data visualization, antenna-related calculations, and data export within a unified platform. The system employs a NodeMCU ESP8266, Arduino Uno R3, servo-driven θ- and ϕ-axis positioning mechanism, NETPIE IoT communication, and a mobile application, while a FieldFox vector network analyzer is used for RF measurements. Positioning performance was evaluated at angular steps of 4°, 5°, 6°, and 10° under unloaded and antenna-loaded conditions. Across all 16 test conditions, the maximum positioning error did not exceed 0.3°, remaining below the predefined ±1° requirement. The far-field distance and antenna-gain calculation functions were also validated against independently calculated theoretical reference values, with gain differences arising only from two-decimal-place display rounding. Radiation-pattern measurements at 2.4 GHz using dipole and Yagi–Uda antennas demonstrated the expected approximately omnidirectional H-plane response of the dipole and the more directional response of the Yagi–Uda antenna. Comparison with CST simulations and ME1300 reference measurements showed agreement in the principal radiation characteristics, including main-lobe direction and overall pattern behavior. The developed system provides an accessible platform for antenna radiation-pattern measurement, with future development focused on automated RF data acquisition.
