Generating fuzzy rule bases from data is essential for building interpretable fuzzy systems. Traditional approaches like Wang-Mendel (WM) rely on correlations but often produce large, redundant rule bases, reducing interpretability and increasing computational cost. To address this, recent work has incorporated causal discovery into rule generation. Te Zhang et al. introduced a method using directed graphs within the Markov blanket, but their reliance on DirectLiNGAM limits applicability to linear data. This paper adopts the Causal Additive Model with Unobserved Variables (CAMUV) to identify a target variable’s Markov blanket and extract its direct causal features. These are then used as inputs to a Takagi-Sugeno-Kang (TSK) fuzzy system. Compared to causal learning algorithms like GRaSP, DECI, and BOSS, CAMUV better handles nonlinear and partially unobserved data, enhancing causal discovery and interpretability. Unlike WM, the TSK system generates dynamic causal if-then rules, improving both interpretability and modeling power. Experiments across seven datasets show an average accuracy improvement of approximately 5% over benchmark models. This work offers a novel, causality-driven approach to constructing interpretable fuzzy systems for complex data.

5G messaging services, based on Global System for Mobile Communications Association (GSMA) Rich Communication Service (RCS) and 3rd Generation Partnership Project (3GPP) IP Multimedia Subsystem (IMS), have been deployed globally by more than 90 mobile operators serving over 421 million monthly active users via 1.2 billion devices. Despite the widespread use, security research of 5G messaging remains sparse. In this paper, we present a comprehensive security analysis and measurement of 5G messaging services, assisted by a semi-automated testing tool we developed. We considered both carrier-side deployment and phone-side software implementations by testing against three large operators, each with hundreds of millions of subscribers, and six popular 5G messaging-enabled devices. We uncovered 4 categories of vulnerabilities, allowing for a wide range of attacks, including Man-In-The-Middle (MITM) attacks, zero-click remote information leakage, phone storage exhaustion and mobile data consumption, and Denial-of-Services (DoS) attacks. Our study underscores the need for further security enhancements in security specifications, implementation, and deployment of 5G messaging services.

In this paper, we study the design of neural network based joint source-channel coding (JSCC) for point-to-point communication of Gauss-Markov sources over the additive white Gaussian noise (AWGN) channel with bandwidth compression. Among the existing deep learning (DL) -based JSCC methods for such sources, the long short-term memory (LSTM) based structure has good performance. However, it takes up huge time and space consumption because of its complex structure. In this work, we propose to adopt the causal convolution and dilated convolution to form our encoder and decoder due to their abilities of effectively extracting the temporal information of sources and their superiority in terms of reducing the time and space consumption. Experimental results show that the proposed model outperforms the traditional JSCC schemes and is comparable to the LSTM-based model in terms of source reconstruction quality. Besides, the proposed model shows a great robustness in the case of channel quality mismatch and correlation coefficient mismatch. Furthermore, our model takes lower time in the test phase and much lower space consumption compared to LSTM-based model.