A Q-learning-based QoS-aware multipath routing protocol in IoMT-based wireless body area network

Collection of recent networking and IoT papers with multiple security-relevant implications. Highlights include: (1) IoMT/WBAN routing (Q-learning multipath) — safety- and privacy-sensitive medical telemetry that could be disrupted or manipulated via routing- and learning-stage attacks; (2) protocol-agnostic backscatter authentication for SWIPT/LoRaWAN — introduces a low-power identification layer but also surfaces replay, relay, and physical-layer cloning risks; (3) federated PEFT for wireless edge (LoRA/Fed-LoRA) — reduces communications overhead but raises model-poisoning and privacy-leak (