FuseFi framework uses diverse Wi-Fi signals for sensing without packet injection

Researchers have developed FuseFi, a novel Wi-Fi sensing framework that leverages irregularly sampled channel state information (CSI) from diverse communication packets and frequency bands. This approach eliminates the need for intrusive packet injection, thereby preserving communication throughput and offering greater deployment flexibility. FuseFi includes a pipeline to harmonize heterogeneous packets and a time-aware attention model that learns from non-uniform CSI sequences. Evaluations on a new dataset, CommCSI-HAR, and existing benchmarks demonstrate that FuseFi achieves state-of-the-art accuracy with a compact model size. AI

IMPACT This research could enable more efficient and flexible Wi-Fi sensing applications by leveraging existing communication data.

RANK_REASON Academic paper detailing a new technical framework for Wi-Fi sensing. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Channel state information
• CommCSI-HAR
• Gaofeng Dong
• Wi-Fi

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →