Work

Adaptive Secret Key Generation from Sparse RSSI Observations for LoRa-based Vehicular Communications

Year: 2026
Type: article
Abstract: Physical-layer secret key generation is attractive for LoRa (long range) vehicular nodes, but mobility limits sessions to a few dozen received signal strength indicator (RSSI) samples, and reported key rates seldom subtract reconciliation leakage.Conditional signal partitioning lets an existing complexity-aware quantizer and Bose-Chaudhuri-Hocquenghem reconciliation run on sessions under 40 samples, with every RSSI-dependent public message except the acceptance bitmaps charged as leakage.The pipeline runs on a Raspberry Pi 5 and was evaluated on 22 vehicle-to-vehicle and vehicle-to-infrastructure field sessions.The pre-reconciliation key disagreement rate ranged from 16.7% to 46.7%, and 25 of 63 blocks were reconciled.Even with fully random reconciled bits, the extractable length is negative in every session, so the secret-key rate is zero.The public messages alone identify every reconciled block.Assuming full-entropy bits and an ideal code, a model-based estimate puts a 128-bit key at about 714 paired observations with this quantizer.
Language: English
Open Access status: hybrid
OpenAlex ID: W7234850005
FWCI:
Cites: 44
Cited by:
Related to: 10