physical layer security

The dataset contains thousands of signal frames from 60 ZigBee devices. The devices and USRP N210 were placed on two movable cabinets on wheels with a maximum distance of 3 meters between them. Due to distance constraints, a coaxial attenuator with an attenuation of 60 dB was fitted at the receiver in order to obtain signals at low SNR. The SNR of the received signal is changed by adjusting the distance between the two cabinets, which can be reduced to about 0 dB when the distance is about 3 metres.

The VNA dataset has three features: frequency, S21, and phase, while the MIMO dataset has an additional 'Channel' feature. The VNA dataset is larger than the MIMO dataset, with 507,709 rows compared to 164,161 rows in the MIMO dataset. This is because the VNA dataset was sampled at a 1 MHz resolution, while the MIMO dataset was sampled at a 25 MHz resolution, which is the limit set by the MATLAB API. As a result, the VNA dataset provides 4,701 samples per tag, while the MIMO dataset provides 190 samples per tag per channel for each reading.

In communication and networking research, obtaining large, real-world datasets related to the physical layer has always been challenging, especially in IoT and Health IoT. In particular, there is a significant interest in datasets that help characterize physical layer properties without interferences from any other communications signals.

Over 34,000 frames from 60 commercial-off-the-shelf ZigBee devices were collected in various scenarios including indoor/outdoor and line-of-sight/non-line-of-sight (LOS/NLOS). The ZigBee devices are hybrid, with 36 equipped with power amplifiers and the other 24 not. The ZigBee device uses the CC2530 chip, while the power amplifier is the RFX2401C chip. The signal frames in each scenario are placed in a separate folder, where all device numbers are fixed. Each frame reaches its maximum length, which includes 266 symbols.

Physical layer security (PLS) is seen as the means to enhance physical layer trustworthiness in 6G. This work provides a proof-of-concept for one of the most mature PLS technologies, i.e., secret key generation (SKG) from wireless fading coefficients during the channel’s coherence time. As opposed to other works, where only specific parts of the protocol are typically investigated, here, we implement the full SKG chain in four indoor experimental campaigns.