EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ARIA: Adaptive Radar Intelligence Architecture

Vaibhav Tayde, Hare Ram Sah

International Journal of IoT, Embedded Systems and Industrial Automation · 2026

DOI: 10.66261/jyyfdp64 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of IoT, Embedded Systems and Industrial Automation, 2026. Verfasst von Vaibhav Tayde und 1 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid advancement of Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), and adaptive signal processing technologies has created unprecedented opportunities for the design of next-generation intelligent radar systems. Conventional radar architectures employ static signal processing pipelines that are unable to dynamically adapt to complex, time-varying, and contested electromagnetic environments — resulting in elevated false alarm rates, reduced target discrimination accuracy, and degraded performance under electronic warfare conditions. ARIA (Adaptive Radar Intelligence Architecture) is proposed as a comprehensive, modular, and end-to-end AI-driven framework that integrates Deep Learning-based Automatic Target Recognition (ATR), Reinforcement Learning (RL) for adaptive waveform design and beam scheduling, Transformer-based multi-target tracking, Explainable AI (XAI) for operational transparency, Federated Learning for distributed radar network intelligence, and edge computing for real-time embedded deployment. This systematic literature review and meta-analysis surveys ten representative studies published between 2021 and 2024, synthesizing AI methodologies, benchmark datasets, performance metrics, and comparative strengths across the principal functional domains addressed by ARIA. Meta-analysis of reported performance metrics yields a pooled effect size of 0.878 (95% CI: 0.860–0.896), confirming the overall effectiveness of AI-based approaches while revealing significant heterogeneity across technique categories. Deep Learning approaches demonstrate the highest pooled accuracy (0.930), while Reinforcement Learning exhibits the greatest variance (SD = 0.072). Principal research gaps identified include the sim-to-real gap in RL-based adaptive systems, scarcity of publicly accessible labeled radar datasets, absence of an integrated end-to-end intelligent radar architecture, and insufficient model explainability in safety-critical deployment contexts. ARIA is positioned as a direct architectural response to these gaps.

Zitieren

APA

Tayde, V., & Sah, H. R. (2026). ARIA: Adaptive Radar Intelligence Architecture. International Journal of IoT, Embedded Systems and Industrial Automation. https://doi.org/10.66261/jyyfdp64

Vancouver

Tayde V, Sah HR. ARIA: Adaptive Radar Intelligence Architecture. International Journal of IoT, Embedded Systems and Industrial Automation. 2026. doi:10.66261/jyyfdp64

BibTeX
@article{Tayde2026,
  author = {Tayde, Vaibhav and Sah, Hare Ram},
  title = {ARIA: Adaptive Radar Intelligence Architecture},
  journal = {International Journal of IoT, Embedded Systems and Industrial Automation},
  year = {2026},
  doi = {10.66261/jyyfdp64}
}
Kostenlos starten →