Lightweight Encryption and Authentication based on Physical Layer Security for the Internet of Things

Abstract

Security in the Internet of things (IoT) has been facing more challenges compared to the conventional human type communications (HTC) due to the limited hardware resources of a IoT device. In general, the security in the IoT has solely relied on cryptography based security mechanisms. However, the cryptography based security is not suitable for a small IoT device because it requires high computational complexity and large signaling overhead. Motivated by the problems of existing cryptographic techniques, we have studied physical layer security schemes which are appealing solutions for the problems in this dissertation. This dissertation largely consists of two parts; 1) a lightweight encryption for the IoT, 2) a lightweight authentication for the IoT.

In the first part of this dissertation, we propose physical layer security schemes for a lightweight encryption in the IoT. First, a 1-bit compressive sensing (CS) encryption scheme with a retransmission strategy is investigated as a lightweight encryption with a secret key for the IoT. The retransmission strategy based on interleaving in the proposed scheme can improve the detection performance at a legitimate receiver, while it slightly improves decision performance to estimate a measurement matrix at an eavesdropper. Secondly, we present a secure index and data symbol modulation scheme which does not require a shared secret key for orthogonal frequency division multiplexing with index modulation (OFDM-IM) systems. In the proposed scheme, the randomized mapping rules for index and data symbol modulation are used to make a detection of an eavesdropper difficult by exploiting channel state information (CSI) between a pair of legitimate transceivers. Through the analysis and simulation results, we show that it has a better performance than an existing security scheme for OFDM-IM in terms of bit error rate (BER) and successful attack probability.

In the second part of this dissertation, physical layer authentication schemes which are suitable for the IoT are presented. First, we propose a unified approach for compression and authentication with a secret key by using the notion of CS. The unified approach which enables simultaneous compression and authentication is applied to smart grid that is one of IoT applications. Through the analysis and simulation results, we demonstrate that the proposed approach can achieve reasonable performances in terms of compression and authentication. Secondly, we present a physical layer authentication method using CSI and carrier frequency offset (CFO) without a secret key in the IoT. Existing physical layer authentication schemes using CSI are vulnerable to some channel environments (e.g., correlated channels) because a channel is used as a signature to provide an identity of a IoT device. To address the problem, CFO is used as an additional signature. Through the analysis and simulation results, we show that the proposed authentication method can significantly improve the authentication performance under correlated channel environments.