AI-Assisted Multi-AP Sensing Signal Reuse
IEEE Internet of Things Magazine, 2026 (ESCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1109/miot.2026.3729001
- Dergi Adı: IEEE Internet of Things Magazine
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
- Anahtar Kelimeler: AI-enabled coexistence, IoT spectrum sharing, ISAC, large language models, multi-AP coordination
- İstanbul Medipol Üniversitesi Adresli: Evet
Özet
Dense Internet of Things (IoT) deployments generate a growing spectrum inefficiency problem where co-located sensing applications independently transmit highly similar sensing signals, flooding unlicensed bands with redundant wireless activity. This paper proposes multi-access point (AP) sensing signal reuse (MAP-SSR), a cooperative framework in which artificial intelligence (AI)-enabled APs group IoT devices by physical-layer requirement similarity and steer them toward compatible existing transmissions rather than generating new ones. The framework further integrates large language models (LLMs) to enable semantic requirement parsing, adaptive threshold management, and ambient signal interpretation, extending MAP-SSR to heterogeneous multi-vendor deployments where devices describe sensing needs using incompatible formats. Monte Carlo evaluation across representative deployment scenarios indicates that reuse opportunities scale with deployment density and that LLM-assisted semantic harmonization provides increasing management-layer reuse opportunities as device heterogeneity grows, collectively reframing IoT density from a spectrum liability into a self-reinforcing spectrum asset.