Synthetic Aperture Radar (SAR) automatic target recognition for defence applications requires models that are accurate, compact, and computationally efficient. This paper proposes EffiSARNet, a lightweight gated multi-stream network for SAR target classification. The architecture combines a shallow convolutional stem with hierarchical Doubly-Stochastic Gated Multi-Stream (DSGM) blocks to enable controlled inter-stream interaction, depthwise spatial filtering, contextual feature gating, and residual information flow. This design preserves target-discriminative scattering cues while maintaining a small computational footprint. Experiments on the ten-class MSTAR benchmark were conducted using validation-based checkpoint selection and independent test-set evaluation. EffiSARNet achieved 99.67 % test accuracy and a 99.67 % weighted F1-score while using only 29,868 trainable parameters, requiring 18.855 million MACs and 37.710 million FLOPs, with an estimated single-precision storage footprint of 0.123 MB for a 128 × 128 input chip. Under the same evaluation protocol, EffiSARNet outperformed several lightweight and deeper convolutional baselines while requiring substantially fewer parameters. These results indicate that EffiSARNet is well suited for deployment-oriented SAR target recognition in resource-constrained environments.
This study examines electroencephalography (EEG)-based mission-type recognition during high-fidelity fighter aircraft simulation, with emphasis on subject-independent generalisation and operational interpretability. Wireless EEG data were collected from seven experienced pilots during three realistic F-16 simulator scenarios: air-to-ground strike under air-defence threat, long-range air-to-air engagement, and close-range air-to-air dogfight. Multiband covariance matrices were extracted from sliding windows and processed using Euclidean alignment and tangent-space mapping. A linear logistic-regression classifier was evaluated with leave-one-pilot-out validation at window and run levels. In addition to the three-class task, a binary air-to-air versus air-to-ground detector was assessed to reflect asymmetric operational risk. The binary model achieved 0.963 run-level accuracy and perfect air-to-air sensitivity at a fixed threshold. Bootstrap resampling supported statistical reliability, while coefficient-based interpretability indicated major beta and low-gamma contributions over fronto-parietal and occipital regions. Although the sample size was limited, the subject-independent validation design provides initial evidence for the feasibility of the proposed approach. The findings support covariance-based EEG decoding as a compact and interpretable approach for mission-context awareness in neuroadaptive aerospace systems. Therefore, these findings are interpreted as preliminary offline estimates from a small subject-independent cohort rather than as evidence of deployment-ready generalisation.
The fundamental concepts, safety guarantees, efficiency, and applicability to real-world cryptographic applications of both classical and quantum secret sharing techniques are carefully discussed in this work. Shamir’s and Blakley’s schemes are examples of CSS schemes that for long have provided information-theoretic security for sharing secrets among multiple parties such that only authorized subsets are able to retrieve the secret. However, in the post-quantum era, these systems inherently have a vulnerability to quantum computational attacks due to the fact that they rely exclusively on classical communication. This paper assesses trade-offs between classical efficiency and quantum-level security by closely examining their methodologies, security models, and resource requirements. Furthermore, it discusses practical issues, such as decoherence, quantum memory constraints, and channel authentication, that present significant challenges in the implementation of QSS. By making a comparison between these two paradigms, we identify the contexts in which each approach is particularly suitable, and thus point to possible hybrid frameworks combining the robustness of classical cryptography with quantum security. These comparisons provide new insights for designing next-generation cryptographic protocols in emerging quantum technologies.
The article provides an analysis of a wideband planar antenna integrated with multiple input multiple output (MIMO) capabilities, designed for radio astronomy and defence grade applications. The suggested antenna comprises of two symmetrically arranged planar radiating elements within a compact area of 45 × 30 mm². A T-shaped defect introduced in the ground plane enhances isolation, achieving values better than -30 dB at 4.4 GHz, 10.8 GHz, 13.7 GHz, 16.6 GHz, and 19.4 GHz. The calculated -10 dB bandwidths are 1.4 GHz (3.21 - 4.61 GHz) and 9.5 GHz (10.5 - 20 GHz), indicating efficient impedance matching which is achieved through the combined effect of the stub loaded rectangular radiators and the T shaped defected ground structure. Across the working frequencies, the antenna achieves an efficiency above 72 % and a gain exceeding 2.1 dBi, with stable impedance characteristics and controlled group delay. The structure demonstrates an extremely low envelope correlation coefficient close to zero and a diversity gain around 10 dB. Additionally, the antenna exhibits a channel capacity loss below 0.4 bps/Hz, a total active reflection coefficient value below -10 dB, and a mean effective gain close to -3 dB, confirming its suitability for MIMO operation. The simulated outcomes show strong agreement with experimental measurements.
The increasing use of swarm drones in recent conflicts has highlighted the vulnerability of critical infrastructures, particularly ports, to low-altitude saturation attacks. This study examines how distance-based defence priorities affect land-based close-in weapon system (CIWS) deployment and simulated penetration outcomes under spatial constraints, permitted coverage conditions, and a limited number of available units. The deployment problem is formulated as a discrete combinatorial optimisation problem on a two-dimensional grid representing protected, deployable, and outer zones, and is solved using the discrete fireworks algorithm (DFWA). The comparative effectiveness is assessed based on swarm-drone penetration simulations conducted using A*-based path-planning and probabilistic engagement models. The results demonstrate that, of the three defence-priority schemes, fine-grained inner-zone-priority deployment substantially reduces penetration success rates compared with uniform coverage-oriented deployment. For six CIWS units and 30 drones, the simulated penetration success rate is 10.70 % under fine-grained inner-zone-priority weighting, compared with 48.71 % under uniform weighting. Within the tested scenarios, the results further indicate that effectively blocking vulnerable approach corridors towards the protected zone can be more important than maximising overall coverage alone. This study provides a concept-level decision-support framework for comparing land-based CIWS deployment schemes under limited-resource conditions and the stated modelling assumptions.
