With the development of mobile and smart sensing technologies, the use of consumer electronics in smart cities has garnered widespread attention from academic researchers and industry professionals due to data security issues. However, specific issues, including privacy concerns regarding unauthorized access and data sharing, require resolution. In this paper, a blockchain-based verifiable private data access control scheme is proposed to protect user-shared information for data sharing in consumer electronic systems for smart cities. The scheme develops a hybrid on-chain access control system using attribute-based encryption (ABE), establishes an authorization model for access control, and prevents sensitive information leakage during data sharing. The simulation results demonstrate that the proposed scheme effectively achieves access control and ensures the secure sharing of sensitive data.

With the rapid advancement of artificial intelligence/generative AI (AI/GenAI), various industries have undergone significant transformations, particularly in the realm of consumer devices and services. AI/GenAI plays a pivotal role in product purchase decisions by analyzing vast amounts of user-generated information, such as reviews. This article aims to provide new solutions for new energy vehicle (NEV) purchase decisions through AI/GenAI to better serve consumers. First, a genetic algorithm-guided long short-term memory and gated recurrent unit are integrated for sentiment analysis to accurately express the sentiment tendencies through AI/GenAI, with the results being converted into probabilistic linguistic term sets. Second, an enhanced K-means++ and an adaptive three-way clustering algorithm are utilized to ensure the accuracy of clustering results. Third, a model based on confidence and conflict levels to identify the non-cooperative behaviors is established to facilitate consensus among decision-makers. Fourth, during the selection phase, the three-way decision is combined with multiattributive border approximation area comparison methods to obtain the optimal alternatives. Finally, using NEV purchase examples, the proposed methods demonstrate robustness and reliability, showcasing their effectiveness in enabling GenAI-driven consumer services.

Deep neural networks have witnessed rapid advancements and remarkable success in recent years, leading to their increasingly widespread implementation on edge devices. However, the deployment, execution, and life cycle management of traditional artificial neural networks (ANNs) on resource-constrained edge devices present significant challenges. Spiking neural networks (SNNs) are a class of neuroscience-inspired neural networks that emulate the low-power operational mode of biological neurons. SNNs possess advantages, such as low power consumption, low latency, event-driven processing, and reduced communication overhead, making them particularly well suited for edge devices and intelligent edge computing. As a result, they have garnered significant attention in both research and practical applications. In this article, we present a comprehensive survey of the fundamentals of SNNs and the advancements in SNN research for edge computing, exploring potential applications and future directions in this emerging field. We also present a case study highlighting that SNNs outperform ANNs in distributed learning, achieving a 6% improvement in accuracy and an 80% reduction in data transmission.

Currently, deep learning technology has become a crucial branch of artificial intelligence, particularly in consumer electronics, where it brings numerous innovative applications through classification and object detection techniques. However, research shows that deep neural networks are susceptible to adversarial example. This brings a series of security issues for related applications. Moreover, adversarial example can also directly exist in the real world and affect models. In this study, we investigate the impact of physical adversarial example on car plate recognition systems and explore the future development and challenges related to this topic. Our simulation results show that adding specific patterns to the surroundings of a car can generate a physical adversarial example, causing the car plate recognition system to fail.