Practical codes for the Gaussian wiretap channel are designed aiming at satisfying information-theoretic metrics to ensure security against a passive eavesdropper (Eve). Specifically, a design criterion is introduced for the coset coding scheme in order to satisfy a strong secrecy condition described with the mutual information between the secret message and Eve's observation. In addition, mutual information neural estimation (MINE) powered from deep learning tools is applied in order to directly compute the information-theoretic security constraint, and verify the proposed solutions. It is shown that finite-length coset codes can indeed ensure secure transmission from an information-theoretic perspective.
Practical codes for the Gaussian wiretap channel are designed aiming at satisfying information-theoretic metrics to ensure security against a passive eavesdropper (Eve). Specifically, a design criterion is introduced for the coset coding scheme in order to satisfy a strong secrecy condition described with the mutual information between the secret message and Eve's observation. In addition, mutual information neural estimation (MINE) powered from deep learning tools is applied in order to directly compute the information-theoretic security constraint, and verify the proposed solutions. It is shown that finite-length coset codes can indeed ensure secure transmission from an information-theoretic perspective.
In this paper, we investigate the application of Reed-Muller (RM) codes for Physical-layer security in a real world wiretap channel scenario. Utilizing software-defined radios (SDRs) in a real indoor environment, we implement a coset coding scheme that leverages the hierarchical structure of RM codes to secure data transmission. The generator matrix of the RM code is used to partition codewords into cosets in the usual way, where each message corresponds to a unique coset, and auxiliary bits select specific codewords within each coset. This approach enables the legitimate receiver (Bob) can decode the transmitted message with minimal information leakage to eavesdropper (Eve) thus protecting the confidentiality of the communication with the help of coset structure. Mutual information neural estimation (MINE) is used to quantify information leakage and validate the effectiveness of the scheme. Experimental results indicate that RM codes can achieve robust security even in practical environments affected by real-world channel impairments. These findings demonstrate the potential of RM codes as an efficient solution for physical-layer security, particularly for applications that require low latency and short blocklengths.
Transmissions over the wiretap channel have been studied for a long time from the information theory standpoint. This has allowed to assess the secrecy performance against eavesdropping while ensuring reliable transmission towards the legitimate receiver. However, most previous studies rely on a number of assumptions which are far from practical wireless communications, like infinite length codewords, random coding, discrete channels or continuous channels with Gaussian signaling. In this paper, we show how the level of security at the physical layer can be assessed from the information theoretic standpoint while taking into account the constraints of practical transmissions over realistic wireless wiretap channels, i.e., by considering practical codes with finite length, discrete modulation formats and continuous channels with fading. For this purpose, we consider the notion of mutual information security, which is provably equivalent to semantic security. Our target is to show that classical and already implemented coding and modulation schemes can be used to achieve some level of security at the physical layer, opposed to approaches resorting to completely new designs tailored to secure transmissions. To corroborate this thesis, we consider some coding and modulation schemes compliant with the IEEE 802.16e (WiMax) standard and show how they can be used to achieve some given security level.
Recently, some researches are devoted to the topic of end-to-end learning a physical layer secure communication system based on autoencoder under Gaussian wiretap channel. However, in those works, the reliability and security of the encoder model were learned through necessary decoding outputs of not only legitimate receiver but also the eavesdropper. In fact, the assumption of known eavesdropper's decoder or its output is not practical. To address this issue, in this paper we propose a dual mutual information neural estimation (MINE) based neural secure communications model. The security constraints of this method is constructed only with the input and output signal samples of the legal and eavesdropper channels and benefit that training the encoder is completely independent of the decoder. Moreover, since the design of secure coding does not rely on the eavesdropper's decoding results, the security performance would not be affected by the eavesdropper's decoding means. Numerical results show that the performance of our model is guaranteed whether the eavesdropper learns the decoder himself or uses the legal decoder.
Physical-layer security uses the noise already present in the channel to keep a message secret without making any assumption about the eavesdropper's computing power. In Wyner's wiretap channel model, coset coding uses randomness to map the message to many codewords and protects against information leakage from an active eavesdropper. The secrecy provided by such codes is measured by their equivocation, which is the eavesdropper's remaining uncertainty about the message after she observes her channel. So computing equivocation is an important aspect of analyzing the wiretap channel model. Many channels of practical interest have no closed-form expression for equivocation, and exact calculation has been limited to a few specific channels. This dissertation studies the equivocation analysis of coset codes along two complementary directions. When a channel offers no closed-form solution, we estimate the information leakage with a neural network called the mutual information neural estimator (MINE). And for a discrete memoryless channel, we derive a general framework for the exact equivocation. The first study builds coset codes from short Reed-Muller codes and tests them over the air in a real indoor environment with software-defined radios. In this study, the secrecy is estimated using MINE, and short coset codes are shown to provide a secrecy advantage. The second study proposes a low-density parity-check (LDPC) code that injects artificial noise by flipping a chosen set of coded bits over a Gaussian wiretap channel. The third study moves from estimation to exact analysis. It examines coset coding over the Ghost Modulation wiretap channel, which is modeled as an asymmetric binary crossover erasure channel (ABCEC). Then it derives exact expressions for both the equivocation and the secrecy capacity. The final study generalizes the exact analysis to any discrete memoryless channel and verifies the framework against well-studied channels. Finally, it applies the developed framework to the asymmetric Z-channel with an efficient formulation that greatly reduces the computational cost.
In a classical wiretap channel setting, Alice communicates with Bob through a main communication channel, while her transmission also reaches an eavesdropper Eve through a wiretap channel. In this paper, we consider a general class of polar secrecy codes for wiretap channels and study their finite-length performance. In particular, bounds on the normalized mutual information security (MIS) leakage, a fundamental measure of secrecy in information-theoretic security frameworks, are presented for polar secrecy codes. The bounds are utilized to characterize the finite-length scaling behavior of polar secrecy codes, where scaling here refers to the non-asymptotic behavior of both the gap to the secrecy capacity as well as the MIS leakage. Furthermore, the bounds are shown to facilitate characterizing numerical bounds on the secrecy guarantees of polar secrecy codes in finite block lengths of practical relevance, where directly calculating the MIS leakage is in general infeasible.
Most of the existing information-theoretic secure key generation schemes have characterized maximum achievable secrecy rate or secrecy capacity over noiseless public broadcast channels. However, in many practical systems even legitimate nodes have to communicate over noisy channels. In this letter, we propose a secret key generation scheme based on secure network coding in Gaussian wiretap channels. We prove that even when the channel from the sender to the receiver is weaker than that to the eavesdropper, secret communication at a positive rate is still possible if the condition σ1<; √σ2/2 + σ 2/3 is satisfied. Moreover, we establish an upper bound on the secrecy leakage rate after privacy amplification using double-layered random coding and joint typicality encoding and decoding over noisy public channels. The proof of achievability and the converse is given to demonstrate that the proposed scheme can achieve information-theoretic security.