The emergence of driverless taxis has made the cybersecurity of these systems a critical issue. Once an attacker gains control of the vehicle, it could lead to serious security risks. Therefore, ensuring the safety of driverless taxi systems is a significant challenge in current technological development. This paper proposes a solution to the problem. It uses secure user authentication and key agreement. These are based on cancelable biometrics and homomorphic encryption. A cancelable biometric scheme is designed. This is based on Gaussian Random Projection (GRP). The aim is to ensure the privacy and security of biometric data is protected. Users only need to provide their facial information to establish a secure connection with driverless taxis, which provides great convenience. We perform security analysis and simulation of the proposed scheme using the Real-Or-Random (ROR) model and the automated validation of Internet security protocols and applications (AVISPA) tool. The experimental results demonstrate the enhanced reliability of our proposed scheme and its inherent simplicity. The proposed scheme has the potential to be a suitable candidate for driverless taxi applications that require remote user authentication.

The popularity of smart homes has been facilitated by the development of information and communication technology and wireless sensors. These technologies allow users to remotely control smart devices in their homes. However, smart home communication primarily relies on the Internet, which exposes smart home networks to multiple security risks, such as smart device capture, simulation attacks, and internal privilege attacks. These threats could result in illegal users stealing data transmitted by smart devices, thereby violating users’ privacy and security. To address these challenges, this article proposes a novel revocable biometric authentication scheme based on secret sharing technology for secure remote user authentication in smart home environments. This solution introduces secret sharing technology to protect users’ revocable biometric templates and enhances the security of user privacy. To verify the effectiveness of the proposed scheme, we conducted experimental verification, including measuring the equal error rate (EER), unlinkability, and revocability. The experimental results demonstrate that this solution can effectively protect the security of user biometric information while maintaining high authentication accuracy. Additionally, it has good revocability and unlinkability, which helps to improve the security and safety of smart homes without compromising user experience.

Wireless body area networks (WBANs) are widely used in the medical field, which makes it easier for doctors to gather patients’ physiological information to diagnose their physical status. Recently, a variety of lightweight authentication key agreement (AKA) protocols have been proposed for WBANs. However, the computational overhead of the authentication process for most of these protocols increases exponentially when the number of patients increases dramatically. Therefore, to improve the authentication efficiency, a many-to-one, lightweight batch AKA protocol based on the Chinese remainder theorem is proposed, which can realize that cloud server (CS) interacts with the gateway nodes once to obtain multiple session keys with different sensor nodes, so as to achieve the effect that multiple sensor nodes participate in the AKA at the same time. In addition, only hash function, XOR operation, and symmetric encryption are used in the proposed protocol. During the key agreement process, CS can compute the session key between it and the sensor node on demand. Sensor nodes can dynamically join and leave the WBAN, which increases the flexibility of the protocol. We verify the protocol’s security using formal security analysis tools, such as the real-or-random model, Burrows-Abadi–Needham logic, and the ProVerif tool. The experiment and comparison results indicate that the proposed protocol outperforms the other related protocols in terms of efficiency.

The fifth-generation (5G) networks can provide high data rates, ultralow latency and huge network capacity. In 5G networks environment, the popularity of the Internet of Things (IoT) has led to a rapid increase in the amount of data. Multiserver distributed cloud computing technology provides an excellent solution to alleviate network pressure caused by the rapid growth of data. However, this technology serves as a two-edged weapon, which not only makes various IoT applications possible but also brings growing concerns for user privacy and ever pressing security challenges. To ensure the high security of 5G network-based applications, we design a secure user anonymity-preserving biometrics and PUFs-based multiserver authentication scheme with key agreement. In our method, we make full use of the inherent security features of user fingerprint and smart device PUF to design a secure multiserver authentication scheme with key agreement in 5G Networks. The proposed scheme is able to resist recognized attacks and its robustness has been verified by security analysis.

Stationary functional brain networks (sFBNs) and dynamic functional brain networks (dFBNs) derived from resting-state functional MRI characterize the complex interactions of the human brain from different aspects and could offer complementary information for brain disease analysis. Most current studies focus on sFBN or dFBN analysis, thus limiting the performance of brain network analysis. A few works have explored integrating sFBN and dFBN to identify brain diseases, and achieved better performance than conventional methods. However, these studies still ignore some valuable discriminative information, such as the distribution information of subjects between and within categories. This paper presents a Double Collaborative Learning Network (DCLNet), which takes advantage of both collaborative encoder and collaborative contrastive learning, to learn complementary information of sFBN and dFBN and distribution information of subjects between inter- and intra-categories for brain disease classification. Specifically, we first construct sFBN and dFBN using traditional correlation-based methods with rs-fMRI data, respectively. Then, we build a collaborative encoder to extract brain network features at different levels (i.e., connectivity-based, brain-region-based, and brain-network-based features), and design a prune-graft transformer module to embed the complementary information of the features at each level between two kinds of FBNs. We also develop a collaborative contrastive learning module to capture the distribution information of subjects between and within different categories, thereby learning the more discriminative features of brain networks. We evaluate the DCLNet on two real brain disease datasets with rs-fMRI data, with experimental results demonstrating the superiority of the proposed method.

In a mobile edge computing environment, the computing tasks of resource-constrained IoT devices are often offloaded to mobile edge computing servers for processing. In order to ensure the security of the task offloading process, both parties need to perform mutual authentication and negotiate a session key first. The security defenses in the existing authentication schemes are often only aimed at external attackers, while ignoring the possible malicious behaviors of semi-trusted servers. Furthermore, they cannot effectively take into account the device-side lightweight and security, as well as the load problem of a single registry. In this paper, we propose a new anonymous authentication key agreement scheme that fully considers the resource constraints of terminal devices and the security risks of semi-trusted servers. In the scheme, we use the method of generating pairing information during registration to avoid the server-side directly contacting the user’s private information, and support trusted third parties not to participate in the authentication process. In addition, by setting up authentication servers to outsource computing tasks, the device-side can avoid blindly selecting a computing server for task offloading, achieve accurate task assignment and efficient execution of authentication. We use Real-Or-Random model and BAN logic to demonstrate the security of the proposed scheme, and use the ProVerif tool to verify its authenticated reachability and confidentiality. Compared with other schemes with the same structure, this scheme is superior to similar schemes, and has higher security on the basis of ensuring the least amount of computation on the device-side.

This article discusses the application of wireless sensor networks (WSNs) in industrial settings, particularly in the context of the Internet of Things. The advantages of WSNs, such as convenience, low energy consumption, and real-time monitoring, make them highly suitable for industrial applications. However, these networks face several challenges, including signal interference, security concerns, privacy issues, and battery life. This article explores these challenges and proposes potential solutions. Current industrial wireless sensor protocols often prioritize lightweight or real-time aspects, potentially neglecting security considerations. This article proposes a strategy for protecting user biometrics during authentication using cancelable biometrics. The approach utilizes sliding window hashing, fuzzy extractors, and physical unclonable functions to generate a cancelable biometric template. The proposed industrial wireless sensor protocols will be used in this strategy. This protocol aims to safeguard users' biometric information in the event of a data breach while detecting and preventing potential attacks on the sensors.

Analysis of functional connectivity networks (FCNs) derived from resting-state functional magnetic resonance imaging (rs-fMRI) has greatly advanced our understanding of brain diseases, including Alzheimer's disease (AD) and attention deficit hyperactivity disorder (ADHD). Advanced machine learning techniques, such as convolutional neural networks (CNNs), have been used to learn high-level feature representations of FCNs for automated brain disease classification. Even though convolution operations in CNNs are good at extracting local properties of FCNs, they generally cannot well capture global temporal representations of FCNs. Recently, the transformer technique has demonstrated remarkable performance in various tasks, which is attributed to its effective self-attention mechanism in capturing the global temporal feature representations. However, it cannot effectively model the local network characteristics of FCNs. To this end, in this paper, we propose a novel network structure for Local sequential feature Coupling Global representation learning (LCGNet) to take advantage of convolutional operations and self-attention mechanisms for enhanced FCN representation learning. Specifically, we first build a dynamic FCN for each subject using an overlapped sliding window approach. We then construct three sequential components (i.e., edge-to-vertex layer, vertex-to-network layer, and network-to-temporality layer) with a dual backbone branch of CNN and transformer to extract and couple from local to global topological information of brain networks. Experimental results on two real datasets (i.e., ADNI and ADHD-200) with rs-fMRI data show the superiority of our LCGNet.

In mobile edge computing, the computing tasks of IoT terminal devices with limited computing power often need to be offloaded to servers for processing. However, there are malicious attacks by adversaries and malicious behaviors of servers in the network, coupled with the use of insecure network channels for data information transmission. These factors seriously threaten the privacy and data security of terminal devices and users. Therefore, it is urgent to use a safe and efficient anonymous authentication key agreement mechanism to verify the legitimacy of the identities of computing participants and ensure the safe transmission of task data. Recently Jia et al. proposed an identity-based authentication scheme, which combines many advantages of previous work and is resistant to various attacks. However, we found that their scheme has security problems, such as offline key guessing attack, internal attack, and user anonymity problems. We classify them as semi-trusted server attacks. In order to solve these security problems, we propose an improved scheme to better realize the authentication function by using flexible and security-enhanced keys for terminal equipment (TE), while ensuring the anonymity of the TE through implicit ID. Furthermore, we provide formal security proof, formal security verification, and security analysis for the improved protocol. Compared with the previous scheme, the scheme has certain improvements in security and performance.

Abstract One of the most common solutions for the prevention of fire accidents is to conduct extensive fire evacuation drills in crowded places. However, there are multiple salient advantages to using virtual reality technology to simulate emergency solutions, for instance, saving costs and greatly decreasing uncertain risks or accidents. Therefore, in this article, a multiagent evacuation framework for complex virtual fire scenarios is proposed and effectively used to simulate a multiagent evacuation procedure to approximate the goal of fire drills in a less costly manner. Specifically, the concept of a multihierarchy agent group model is proposed; that is, the evacuation of multiple agents is separated into leader‐follower and freedom modes. Additionally, several complex actions of individual humans in actual fire drills are fully considered, and a multiaction agent schema is presented to characterize the associated real effects. In addition, generative adversarial imitation learning is adopted to obtain the evacuation path of the leader‐agent by training numerous learning epochs. Finally, extensive experiments are conducted to validate the feasibility of our proposed method. The results show that the proposed method is superior to other methods and that it realistically and reasonably shows the procedure of multiagent evacuation in complex fire emergency scenarios.

Functional connectivity (FC) networks derived from resting-state functional magnetic resonance imaging (rs-fMRI) have been widely used in automated identification of brain disorders, such as Alzheimer's disease (AD) and attention deficit hyperactivity disorder (ADHD). To generate compact representations of FC networks, various thresholding methods have been designed for FC network analysis. However, these studies usually use a pre-defined threshold or connection percentage to threshold whole FC networks, thus ignoring the diversity of temporal correlation (e.g., strong associations) between brain regions in subject groups. In this work, we propose a distribution-guided network thresholding learning (DNTL) method for FC network analysis in brain disorder identification with rs-fMRI. Specifically, for each connection of a pair of brain regions, we propose to determine its specific threshold based on the distribution of connection strength (i.e., temporal correlation) between subject groups (e.g., patients and normal controls). The proposed DNTL can adaptively yield an FC-specific threshold for each connection in an FC network, thus preserving diversity of temporal correlation among different brain regions. Experiment results on 365 subjects from two datasets (i.e., ADNI and ADHD-200) suggest that the DNT method outperforms state-of-the-art methods in brain disorder identification with rs-fMRI data.

With the current development and popularization of biometrics recognition technology, our biometrics and other identity information may be illegal bulk scalping, and there is the possibility of being used for false enrolment, network fraud and other illegal criminal activities. Although some network platforms based on biometrics recognition adopt multi-identity authentication, network hacking technology is also improving constantly. Therefore, we must not ignore the importance of biometrics data protection. To this end, we propose a complete user authentication protocol and key agreement scheme based on cancelable biometrics and physical unclonable function (PUF). Firstly, cancelable biometrics are generated by efficient biometrics fusion processing which called “PUF-TTM” (Template Transformation Method) using a PUF embedded into the device. Then based on Biometrics-as-a-Service (BaaS) model and secret sharing technology, a complete authentication protocol in multi-server environment is designed, and the robustness, effectiveness and security of our proposed scheme are ensured from the perspective of performance and security analysis.

Fingerprint enhancement plays a very important role in automatic fingerprint identification system. In order to ensure reliable fingerprint identification and improve fingerprint ridge structure, a novel method based on the collaborative filtering model for fingerprint enhancement is proposed. The proposed method consists of two stages. First, the original fingerprint is pre‐enhanced by using Gabor filter and linear contrast stretching. Next, the pre‐enhanced fingerprint is partitioned into patches in spatial domain, and then the patches are enhanced based on spectra diffusion by using the two‐dimensional (2D) angular‐pass filter and the 2D Butterworth band‐pass filter. The proposed method takes full advantage of the ridge information and spectra diffusion with higher quality to recover the lost ridge information. To evaluate proposed method, the databases FVC2004 are employed, and the comparison experiments are carried out using various methods. Comparative experimental results show that the proposed algorithm outperforms the existing state‐of‐the‐art methods on fingerprint enhancement.

Outlier detection is an interesting issue in data mining and machine learning. In this paper, to detect outliers, an information-entropy-based k-nearest neighborhood relevant outlier factor algorithm is proposed that is combined with Shannon information theory and the triangle pruning strategy. The algorithm accounts for the data points whose k-nearest neighbors are distributed on the edge of the range within the designated radius. In particular, the neighborhood influence on each point is considered to address the problem of information concealment and submergence. Information entropy is used to calculate the weights to distinguish the importance of each attribute. Then, based on the attribute weights, the improved pruning strategy reduces the computational complexity of the subsequent procedures by removing some inliers and obtaining the outlier candidate dataset. Finally, according to the weighted distance between the objects in the candidate dataset and those in the original dataset, the algorithm calculates the dissimilarity between each object and its k-nearest neighbors. The data points with the top $r$ dissimilarity are regarded as the outliers. Experimental results show that, compared to existing methods, the proposed approach improves pruning and detection rates while maintaining the coverage rate.