EFFICIENT PHYSICAL KEY GENERATION TECHNIQUE BASED ON CHANNEL CORRELATION AND TRANSFORM FUNCTION FOR THE INTERNET OF THINGS.

Abstract

In the context of the Internet of Things (IoT), devices often operate with constrained computing resources and limited energy capacity. This limitation poses challenges for employing encryption schemes with high computational complexity to ensure secure communication. In response to this challenge, physical layer key generation emerges as a promising solution. Physical layer key generation leverages the characteristics of the wireless channel to generate shared secret keys. By exploiting the inherent randomness and unpredictability of wireless channel measurements, such as signal strength variations and noise patterns, secure keys can be derived without the need for complex cryptographic operations. This approach offers several advantages for IoT devices. Firstly, it utilizes existing wireless communication mechanisms, eliminating the need for additional resources dedicated solely to key management. Secondly, it can be implemented with minimal computational overhead, making it suitable for resource-constrained IoT devices. Thirdly, since the keys are derived directly from physical channel properties, they can be highly resistant to certain types of attacks, such as eavesdropping and man-in-the-middle attacks. This paper proposed a novel physical layer key generation method for authenticating IoT-enabled communication devices. By integrating transform functions and channel parameters into the key generation process, the proposed algorithm aims to minimize key disagreement rates and enhance the capacity of physical layer security, contributing to the overall robustness and reliability of wireless communication systems. The proposed algorithm can generate equal 128-bit keys without preprocessing or error correction, achieving KSR values of 1.05 bps. The performance of the proposed method is evaluated through numerical simulations, with computing time reduced by up to 25.77 and 26.08 times, and communication/synchronization time reduced up to 1.55 and 1.52 times compared to scenario-1.