The rapid iteration of artificial intelligence (AI) and 5G technologies has created essential foundations for Metaverse, easing bottlenecks in content generation, real-time interaction, and large-scale deployment. Integrating unmanned aerial vehicle (UAV) with semantic communication (SemCom) further provides a lightweight sensing and uplink solution, but semantic compression and bias may degrade fidelity. Generative AI (GAI), with powerful contextual modeling and generation capabilities, helps mitigate these issues. Based on this, this paper proposes a GAI-aided SemCom architecture and evaluates the feasibility of applying it to low-altitude internet of things (IoT) networks. We build evaluation models for representative image compression and GAI-aided SemCom techniques, replacing traditional task assumptions with realistic cases. Then, taking UAV networks as an example, we design a path planning and two task allocation schemes. Specifically, the UAV path planning strategy together with a greedy task allocation scheme improves the Metaverse update frequency and ensures the data freshness, while a parametrized deep Q-network (PDQN) task allocation scheme jointly optimizes Metaverse fidelity and freshness through mixed action selection, i.e., preprocessing methods (discrete) and compression size (continuous). Extensive analysis and numerical results corroborate that GAI-aided SemCom technology has practical significance in low-altitude IoT networks. Through the collaboration of multiple data processing schemes and the rational resource allocation, the fidelity and freshness of Metaverse can be greatly improved.

Mars exploration holds the potential to uncover the origin of life, which requires a reliable and sustainable electrical power supply to support long-duration missions. Wireless Power Transfer (WPT) based on Space Solar Power Satellites (SSPSs) has emerged as a promising option due to its continuous 24 × 7 availability and renewable nature. In particular, Laser-based WPT (LWPT) is especially suitable for space energy transmission due to its narrow beam width and high directivity. However, the extremely long propagation distance, high dynamics, and harsh conditions such as atmospheric attenuation, dust storms, and plasma effects degrade the transmission efficiency. To improve the robustness of energy delivery under such uncertain conditions, we incorporate Intelligent Reflecting Surfaces (IRS) to reconfigure the wireless propagation environment. Specifically, IRSs are exploited to provide alternative reflective transmission paths and enable adaptive wavefront control at the Mars rover, thereby enhancing the reliability and efficiency of power transfer when the direct link is partially blocked or severely attenuated. Additionally, we consider the presence of a malicious rover attempting to steal energy or disrupt legitimate charging. To ensure secure charging, we propose an IRS-assisted Challenge-Response Physical-Layer Authentication (CR-PLA) scheme. The False Alarm (FA) and Miss Detection (MD) probabilities are adopted as key performance metrics to quantify the authentication reliability. Finally, we formulate an optimization problem to maximize the harvested energy at the legitimate rover by jointly optimizing satellite selection, active transmit beamforming, and the passive IRS reflection, subject to the FA and MD probabilities. To address this intractable non-convex problem, we propose an Alternating Optimization (AO) algorithm to solve it iteratively. Numerical results demonstrate that, compared with benchmark methods such as Successive Convex Approximation (SCA) and greedy approach, the proposed method significantly enhances the harvested energy while effectively reducing MD probability.

The autonomous driving, which integrates wireless communication, intelligent computing and environmental perception, not only improves traffic safety and reduces vehicle accidents, but also alleviates traffic congestion by optimizing traffic flow and brings comfortable and convenient travel experience to passengers. However, current autonomous driving technology is generally at the L3-L4 levels and faces many challenges, such as semantic segmentation. Semantic segmentation enables vehicles to correctly distinguish the surrounding environment, such as roads, vehicles, pedestrians, etc. It can assist drivers in perceiving the surrounding environment well and making correct decisions, improving driving safety. However, current semantic segmentation work mostly focuses on improving recognition accuracy and neglects inference speed. Slightly higher latency can easily prevent vehicles from making timely and correct decisions, leading to accidents such as car crashes. Therefore, we propose a Multi-Level Real-time Fusion Semantic Segmentation Network (MLRFNet) that improves inference speed while ensuring high semantic segmentation accuracy for autonomous driving. The MLRFNet utilizes two lightweight branches, achieving effectively extract RGB and depth features with low computational cost. In addition, the Feature Fusion Module (FFM) aggregates complementary features from them, while the Cross-Level Refine Module (CRM) merges high-level semantic features and low-level spatial information. Extensive experiments demonstrate that MLRFNet significantly improves the inference speed while ensuring high accuracy. On the Cityscapes validation set, MLRFNet achieves 251.8 FPS and 71.4% mIoU for 512 × 1024 images inputs.

As an emerging application of the Internet of Things (IoT), autonomous vehicles (AVs) has attracted widespread attention from scholars. Accurate vehicle positioning is crucial for AVs to navigate safely and efficiently. Among the key components of positioning systems, odometry plays a vital role in tracking the vehicle’s movement, especially when GPS is unavailable. Traditional single-modal odometry methods, which rely solely on either LiDAR or cameras, often experience limited accuracy in challenging environmental or weather conditions. Therefore, some researchers proposed an idea of combining information from both LiDAR and cameras, called LiDAR-camera odometry (LCO). However, the existing LCO methods face issues like insufficient data integration or complexity in system structure. To address these challenges, we propose a novel LCO method named EVP-LCO, which enhances the sparse features from LiDAR by pseudo-LiDAR. In our method, we use a data augmentation module to enrich the details of LiDAR point cloud. Furthermore, we devise a feature regrouping strategy in two-step Levenberg-Marquardt (LM) optimization process to estimate accurate pose and reconstruct colorful global map. The results on the KITTI Odometry dataset show that EVP-LCO significantly improves vehicle positioning accuracy.

Intelligent vehicles are becoming an essential means of transportation for users, which provide them with high-quality services and convenient travel experiences. It should be noted that while the number of intelligent vehicles is growing rapidly, the incidence rate of traffic accidents is also increasing synchronously. Thus, some researchers try to reduce the risks through trajectory prediction. The current trajectory prediction schemes can be divided into single-vehicle-based scheme and vehicle-road collaboration scheme. However, single-vehicle-based scheme is lack in the states of surrounding vehicles and environment. Vehicle-road collaboration scheme is unable to overcome the inevitably limited field of view. In addition, present trajectory prediction schemes are limited by the computing resources of vehicles. Digital twin has become a hot topic due to its high computing speed and adequate environment information based on its cloud platform. Combining digital twin and trajectory prediction can not only broaden the vehicle’s field of view but also alleviate vehicle computing resources. For this, we first propose an adaptive trajectory prediction scheme based on digital twin platform. The whole computational process is placed in the cloud to alleviate the limited computing power while improving the computing speed. Vehicles can obtain the information of surrounding environment and the prediction results in real time, which improves the accuracy of the data used for prediction. Moreover, we utilize multichannel attention mechanism to learn the trajectory’s relevance weights adaptively, which improves the processing speed of scheme. The experimental results show that the performance of our scheme is better than others, so that drivers can adjust their routes to avoid collisions.

With the rapid growth of Internet of Medical Things (IoMT) devices and the advancement of medical large language models, smart healthcare applications, including regular vital sign monitoring, medical intervention, and medication adherence tracking, play an increasingly important role in the near future. As the data generated by smart healthcare applications usually have strict requirements for latency and privacy, traditional data processing in a centralized manner is not the preferred choice, especially for users in remote areas. In this article, we consider the nonterrestrial networks composed of Low Earth Orbit satellites and autonomous aerial vehicles (AAVs) to realize seamless coverage. A federated reinforcement learning-based hierarchical computation offloading strategy is proposed to make distinct decisions and protect data privacy. Moreover, we optimize the communication efficiency between IoMT devices, AAVs, and satellites, improve task processing capabilities, and reduce network load by reasonably setting task priorities. The simulation results show that the proposed method performs well in reducing energy consumption and shortening task completion time, significantly improving resource utilization and achieving system load balancing.

With the advantages of seamless coverage and ubiquitous connections, Non-Terrestrial Networks (NTNs) composed of Low Earth Orbit (LEO) satellites and Unmanned Aerial Vehicles (UAVs) can provide content caching services for future Connected Automated Vehicles (CAVs) to satisfy onboard collaborative viewing, traffic sensing, and metaverse entertainments in remote areas. However, the heterogeneous caching hardware, communication environments, and frequent network dynamics make the optimization of content caching policy highly complicated. Firstly, considering all LEO satellites as caching satellites can lead to content duplication and radio interference, causing storage waste and NTN transmission quality deterioration. Secondly, how to provide customized QoS by intra-layer and inter-layer cooperative caching in such complicated environments remains an open issue. Thus, we propose a Delay-Motivated Ant Colony Optimization (DM-ACO) scheme to select caching LEO satellites with reduced system propagation delay. Then, the Multi-Agent Deep Reinforcement Learning-based Hierarchical Caching and Asynchronous Updating (MADRL-HCAU) strategy is designed to manage the caching capacity of LEO satellites and UAVs, providing customized services for CAVs and dispensing the peak traffic. Simulation results illustrate that the proposed scheme can not only effectively accelerate the caching refreshing and content downloading process but also significantly reduce the packet drop and improve the cache hit ratio.

The development of 6G should enable users in remote and harsh areas to enjoy computation-intensive services including metaverse entertainment, intelligent transportation, and immersive communications. Low Earth Orbit (LEO) satellite constellations widely constructed in recent years have been recognized as an efficient solution to complement the terrestrial infrastructure with seamless coverage and decreasing expenses for both communication and computation services. However, the widely studied Federated Reinforcement Learning (FRL) based task offloading strategies neglect the potential trust concerns like malicious satellites and buffer pollution, while 6G service providers may rent the LEO satellites belonging to different companies to minimize the expense. To address these issues, blockchain has been considered in the Zero Trust (ZT) scenario, with the group consensus mechanism through the smart contract. Moreover, we propose a Constrained Correction Voting Mechanism (CCVM) to give punishing correction to the aggregation weight of malicious voting satellites. Furthermore, a Cold Start Reputation Aggregation (CSRA) scheme is adopted to first severely degrade and then gradually recover the weight of Federated Learning (FL) sub-models trained by malicious satellites. Thus, the Blockchain-enabled Cold Start Aggregation FRL (BCSA-FRL) scheme is proposed to make effective and secure offloading decisions in the ZT LEO satellite Networks. The numerical results illustrate the advantages of our proposal.

Roadside units (RSUs) play a vital role in intelligent transportation systems (ITS), working as critical elements in delivering superior Internet of Vehicles (IoV) services. A large service coverage and fast accident information diffusion RSU deployment solution can reliably ensure the ITS’ quality of service. Simultaneously, with the development of the city and the ITS, changes in traffic flow lead to RSU load imbalance, which will reduce the benefit of the original RSU deployment, and it is necessary to adjust RSU locations with minimal cost. Besides, due to the high visibility of the ITS, RSUs are highly susceptible to external attacks, which is commonly overlooked in existing RSU deployment work. Specifically, Sybil attack is one of the most dangerous attacks against ITS, and it can reshape the network state by forging multiple identities, interfering with risk sensing, etc. Motivated by these, we respectively propose the PSO-meme joint heuristic deployment algorithm (PJHDA) and the heuristic RSU multi-objective adaptation adjustment algorithm (HRMA3) to carry out deployment and adaptation adjustment of the city’s RSUs, taking into account the constraint of Sybil attack detection. Numerical results demonstrate that the multi-attribute performance of PJHDA is superior to the existing schemes. Compared with benchmark schemes, the HRMA3 excels in achieving advanced service coverage and load balancing while controlling costs, and both proposed schemes exhibit higher Sybil attack detection rate.

The development of Autonomous Vehicles (AVs) provides users with high-quality services and convenient travel experiences. As one of the most important functions in the automotive field, mobile positioning has attracted widespread attention from scholars. However, using a single-modal sensor (LiDAR or camera) poses challenges for precise localization due to their measurement flaws. Therefore, some scholars have proposed Visual-LiDAR Odometry (VLO). Nevertheless, most of the existing VLO solely use a single-modal sensor as their main framework and utilize another sensor for optimization, which does not fully leverage the complementary behavior of sensors in different environments. Thus, this paper presents a novel LiDAR-camera fusion method for improving the odometry estimation. Firstly, we employ a depth completion network to convert the image into pseudo-LiDAR to compensate for the missing depth values in the LiDAR point clouds. Then, we adopt Bayesian inference to enhance the robustness of the fusion method in different environments. Finally, evaluations on the public KITTI odometry show that the proposed method outperforms several state-of-the-art methods.

The development of 5G-vehicle-to-everything (5G-V2X) technology makes vehicle-road-cloud collaboration possible. Vehicles and roads transmit sensor data to the cloud via 5G-V2X technology and then the cloud sends the data to the target vehicle. The target vehicle utilizes dynamic environmental data from surrounding vehicles and roadside units to predict the driving trajectory of surrounding vehicles in order to ensure its own safety. However, many existing trajectory prediction schemes are based on incomplete single-vehicle perception and ignore surrounding road conditions, which will greatly limit their value in real-world scenarios. Therefore, this paper proposes a flexible multi-channel vehicle trajectory prediction scheme based on vehicle-road collaboration. Specifically, we first design a flexible multi-channel vehicle trajectory prediction scheme that can extract different vehicle and map features from various information sources. Then, we use the Transformer model to generate predicted trajectories of surrounding vehicles by fusing features from different sources, and achieve parallel computing effects. The most popular dataset, INTERACTION, is used to evaluate the proposed scheme. The results show that our scheme is robust across different scenarios and possesses better accuracy.

As a critical element in intelligent transportation systems (ITS), roadside units (RSUs) are pivotal in delivering superior Internet of Vehicles (IoV) services encompassing intelligent traffic management, accident prevention, and emergency rescue. Considering the high deployment and maintenance costs of RSUs, many studies focus on the efficient RSU deployment issues. However, due to the high visibility of the ITS system, RSUs are highly susceptible to external attacks, which is commonly overlooked in existing RSU deployment researches. Specifically, the Sybil attack is one of the most dangerous attacks against ITS, it can reshape the network state by forging multiple identities, interfering with the operator’s reputation assessment or causing severe DDoS. Therefore, we propose a joint heuristic scheme that combines the advantages of particle swarm optimization and double local-search memetic algorithm to solve the city RSU deployment problem in Sybil attack environments. It can find solutions with higher fitness values and guarantees that the IoV has the capability to detect Sybil attacks. Numerical results show that our proposed scheme not only outperforms other traditional solutions regarding signal validity coverage, overlap rate, and initial propagation speed of accident information, but also performs satisfactorily in Sybil attack detection.

With the increase of the near field communication (NFC) function deployment on smartphones, the mobile NFC card system becomes a solution to inadequacies of conventional physical authentication in accommodating modern Internet of Things (IoT) scenarios involving shared access, like shared family bank cards and household vehicles. While NFC technology brings convenience to these scenarios, it also introduces security vulnerabilities, including relay attacks and man-in-the-middle attacks. Given the multi-user context, these vulnerabilities are further amplified. Therefore, it is urgently needed to design an authentication for securing the mobile NFC card system. However, existing security approaches, such as protocol authen-tication and supervised radio frequency (RF) authentication, face challenges in computational resource allocation and data labeling, making them inadequate for the multi-user mobile NFC card system. To address these issues, this paper proposes NFC-RFAE, a semi-supervised authentication system based on RF signals, demonstrating high accuracy, high recall rate, low latency, and minimal space occupancy in real-life scenarios, including a vehicle keyless system and a mobile bank system.

As the evolution of 3GPP specification and the deployment of 5G network, Internet of Vehicles (IoVs) boom fireworks. However, its attack surface is expanded with the increased fusion of various functional interfaces, leading to easier penetration of vehicles. To deal with endless vehicle attacks, scholars propose many methods, where intrusion detection system (IDS) is an important branch. However, many IDSs are based on characteristics of single or specific types of vehicles, which limits model transplantation. Besides, 1-D features are usually utilized in existing IDSs, such as time, traffic, or voltage, etc., limiting the ability to detect attacks related to other dimensions. What is more, many IDSs harness machine learning algorithms and are deployed in vehicles simultaneously, which aggravates the computational burden. Therefore, we devise a cloud-vehicle collaborative IDS based on multidimensional features (CVMIDS) for IoV, called CVMIDS. It solves the problem of data heterogeneity by abstracting different vehicle data to the same feature space. Thus, data sets from different vehicles can be fed into one model for multiclassification, which naturally solves the problem of model transplantation. The feature space is established by combining features in dimensions of time, traffic, and voltage, thereby extending the types of attacks that CVMIDS can detect. Due to the deviated location of abnormal data in feature space compared with normal data, CVMIDS will misclassify vehicle data. Hence, CVMIDS can detect intrusions based on multiclassifying vehicles. Extensive experiments are conducted on three vehicles with different brands and numerical results corroborate the robustness and efficiency of CVMIDS.

As innovative technologies such as autonomous driving, over-the-air technology, and vehicle-to-everything are widely applied to intelligent connected vehicles, people can gain a more convenient and safer driving experience. Although the application of these technologies facilitates our lives, they also bring a series of vulnerable interfaces (such as 5G, Bluetooth, and WiFi), which pose a significant security threat to existing in-vehicle networks. To address these threats, researchers have proposed two mainstream schemes, including message authentication and intrusion detection system (IDS), where the scheme of message authentication needs to occupy the limited bandwidth of controller area network (CAN) bus. Furthermore, most IDSs either cannot locate the sender of the attack, fail to detect aperiodic malicious frames, or require prior knowledge of which CAN identifiers (IDs) belong to which electronic control units (ECUs). To address these weaknesses, we propose a practical voltage-based IDS named IdentifierIDS for real in-vehicle networks. To the best of our knowledge, it is the first scheme to detect intrusions by establishing a voltage fingerprint for each ID without the need for prior knowledge. This allows IdentifierIDS to detect both periodic and aperiodic malicious frames without occupying the limited bandwidth of the CAN bus. As a self-learning IDS, it can adapt to different in-vehicle networks without the need for customization for them. Experiments on three real vehicles demonstrate the robustness of our scheme in different in-vehicle networks.

Intelligent vehicles have become a typical representative of the next generation of technological developments, which provide users with convenient and comfortable driving experiences. However, owing to the factors in weak access control of communication interfaces, lack of authentication for data interaction, and no source/destination address for messages, the controller area network (CAN) is vulnerable to malicious attacks. Due to the resource constraints of bandwidth and the high requirement of real-time data transmission in CAN bus, security solutions, such as in-vehicle gateways and firewalls, cannot be endowed with complex encryption authentication algorithms, contributing to the limited capabilities in security protection. Therefore, some scholars proposed intrusion detection systems (IDSs) based on side-channel analysis. However, most of the existing detection models are based on single-class side-channel features to detect limited types of attacks. How to design a robust, lightweight, highly real-time, and traceable IDS is a major problem in the field of vehicle security. In view of this, we design a new IDS based on multiple features, called MIDS. We have conducted numerous experiments on two different brands of vehicles for three typical attacks, and MIDS has a detection accuracy of more than 98% and detects per frame within 0.12ms.

The near field communication (NFC), as one of the most widely used radio frequency identification (RFID) technologies, has been applied to intelligent devices to replace the traditional key, bringing convenience to people’s lives. While the appearance of NFC keys facilitates users’ lifestyles, it also increases the risk of being stolen for intelligent devices. So, it is urgently needed to take action to protect the NFC security of equipment. There are abundant researches on NFC security, which can be divided to protocols authentication and data analysis two main defense methods. However, the way of protocols authentication is limited by the space available and real-time communication of devices. The way of data analysis can not identify the malicious NFC devices from outside. Thus, we propose an intrusion detection system (IDS) based on radio frequency (RF) signals for NFC security, called NFC-IDS. We use the random forests algorithm to select the four most important feature extracted from RF signals and compare random forests, support vector machine (SVM), and k-Nearest Neighbor (k-NN) algorithms to detect intrusions. The experimental results on two real electric motorcycles show that every NFC device has its unique physical signal characteristics, which can be used to detect intrusions with high accuracy and robustness.

Unmanned Aerial Vehicles (UAVs), also known as drones, have recently been proposed as flying base stations for providing reliable service to IoT devices. However, due to the lack of effective authentication schemes, UAVs are often hijacked by adversaries, which raises a high potential for sensitive information leakage. Therefore, designing a real-time authentication scheme is essential to enhance UAV safety. Up to the present, several works exist about pilot authentication by classifying radio-control signals. As propagating through the open environment, radio-control signals can be sniffed, analyzed, and simulated, posing significant threats to UAV security. For this reason, we propose a novel deep learning-based drone pilot authentication scheme, DP-Authentication, to protect UAVs from malicious radio-manipulated attacks. Specifically, we collect UAV flight data from the onboard PX4 flight stack and feed them into the authentication scheme to validate pilot legal status dynamically. As verified by comprehensive experiments, the proposed authentication scheme can authenticate pilots with an accuracy of 95.24% and detect malicious hijacking with an accuracy of 96.82%. Thanks to the low system overhead, it holds great promise for deployment on the UAV side to monitor pilot legal status in real-time.

As emerging technologies such as mobile communication, vehicle to everything, and artificial intelligence are widely used in intelligent connected vehicles, drivers can gain a convenient and colorful driving experience. While these tech-nologies enrich the driving experience, they also bring a series of vulnerable interfaces to the vehicle. These interfaces can be used by hackers to attack other nodes of in-vehicle network that lack authentication and encryption. For this, researchers design scheme to encrypt and authenticate messages to protect in-vehicle networks, but this scheme would occupy the bandwidth resources of in-vehicle network. Therefore, researchers propose parameter monitoring-based intrusion detection system (IDS), information theory-based IDS, and fingerprint-based IDS, which do not occupy bandwidth. However, most IDSs either cannot locate the source of the attack, cannot detect aperiodic frames, or need to know the non-public mapping between electronic control units (ECUs) and identifiers (IDs) of in-vehicle network. To solve these weaknesses, we propose a novel IDS that establishes voltage fingerprints for each ID. This system can detect period and aperiodic malicious frames and locate the source of attack without knowing the mapping between ECUs and IDs. The experimental results on actual vehicles demonstrate that our scheme is robust against real scenarios.

5G and artificial intelligence greatly promote the development of intelligent and connected vehicle (ICV). However, ICV opens more ports to the outside world, making it easy for hackers to intrude controller area network (CAN) and control ICV. Therefore, many researchers design intrusion detection systems (IDSs) to detect vehicle intrusion in real-time. In this paper, we propose a highly camouflaged attack method called the same origin method execution (SOME) attack. The intrusion messages of this attack have the same characteristics as normal messages and can bypass most existing IDSs. To detect this attack, we design a reliable IDS for ICV based on a generative adversarial network (GAN) called GVIDS. It takes CAN messages as the input sample and trains the IDS model to distinguish the legality of messages. Experiments on two real vehicles show that GVIDS can detect most existing attacks, including spoofing, bus-off, masquerade, and SOME attacks. The average detection accuracy of GVIDS is 96.64%, and the average running time of each detection is only 0.18 ms. In addition, the experiment also shows that the detection performance of GVIDS is not affected by the value of identifiers in CAN messages.

With the popularity of intelligent and connected vehicles, driver identification based on driver behaviors, which has great significance in driving safety, vehicle alarm system and other situations, attracts more and more attention. However, the implementation of driver identification also faces some problems, such as insufficient computing power, slow transmission rate and so on. In order to solve these problems, people use mobile edge computing technology to transfer the computing process, which makes drive identification more feasible in practical scenarios. Many researchers have done a lot of works for the accuracy of driver identification. Nonetheless, few of them pay attention to the identification efficiency in the scenarios where drivers are constantly added to the model. Different from the works before, we propose a new driver identification scheme based on incremental learning for the first time, which can greatly improve the efficiency of model retraining when new driver is added. We extract the driver behavior characteristics data from two physical vehicles and train an incremental learning model in edge server. The experimental results show that the our scheme can accurately identify drivers. With the number of drivers in model increasing, although the identification accuracy slowly decreases, it remains above 93%, which means our scheme against the catastrophic forgetting well. Comparing with the existing identification schemes, our scheme requires less time and memory resources for model retraining, which is suitable for practical scenarios.

In the Internet of Things (IoT) era, WiFi is now commonly implemented worldwide as a convenient wireless data transmission technology and brings much convenience to people’s life. Personnel counting, which plays an indispensable role in many areas, such as indoor crowd control, public safety, and marketing analytics, is also important in some special events. In emergencies, such as bank robberies, where police need to capture the number of robbers and hostages in a robbed bank, counting the number of people within the region using WiFi perception technology is more reliable and safer than traditional counting methods based on video streaming. In the early work of others, researchers used received signal strength indicator (RSSI) from the MAC layer for personnel counting studies. In recent years, it has been found that the channel state information (CSI) from the physical layer is more stable and the personnel counting method based on CSI has a higher accuracy rate. In their excellent works, however, researchers do not take into consideration the unpredictability of crowd behavior in real life. Meanwhile, the experimental accuracy needs to be greatly improved. This paper, therefore, focuses on proposing a novel CSI-based WiFi perception personnel counting method using the Long Short Term Memory (LSTM) algorithm that is useful for hidden counting the number of people in different situations. We conducted people counting experiments under three real experimental scenarios. In addition, we have compared three other algorithms for machine learning and optimized some parameters to achieve an overall accuracy of over 97% for our method.

With emerging technologies such as 5G, artificial intelligence, and other emerging technologies widely used in intelligent connected vehicles (ICVs), users can obtain more personalized service and more comfortable experiences. Although these technologies significantly facilitate our quality of life, they also bring a series of vulnerable interfaces, which threaten the security of in-vehicle networks, such as the controller area network (CAN) bus. Therefore, many researchers design intrusion detection systems (IDSs) to detect malicious frames. However, most IDSs cannot locate the sender electronic control unit (ECU) of the malicious frames, the compromised ECU. This means vehicles cannot take timely defensive measures against the ECU, which seriously endangers the safety of users. In order to identify the sender more accurately, we design a lightweight sender identification scheme based on the physical layer characteristics of vehicles. It does not increase the load and calculation burden of the CAN bus, and it can accurately map multiple identifiers (IDs) to each ECU without developer documentation. When compromised ECUs send malicious frames to attack vehicles by spoofing or masquerading, the scheme is able to accurately identify the sender, with an average accuracy rate of over 95%.

Intelligent and connected vehicles (ICVs) have become the mainstream in the development of automobile industry. Many emerging technologies have been proposed to provide users with comfortable and convenient driving experience. However, even though these technologies significantly improve the quality of service, some of the communication interfaces they used are vulnerable and easily attacked. Note that although many malicious attacks can be carried out in various ways, their final step must be in the in-vehicle network, i.e., the controller area network (CAN) bus. In order to protect the security of the CAN bus, it is of great importance to design an intrusion detection system (IDS), which can monitor the message transmission in real time. In this article, we design a novel external IDS based on vehicle voltage signals, named VehicleEIDS. It does not occupy the bandwidth or computing resources of the CAN bus and maintains the original CAN bus protocol as well. The system can be directly installed in the automobile gateway to monitor the external intrusion, and can be connected to the CAN bus as an independent external device to protect the automobile security. In addition, VehicleEIDS is robust against the factors of vehicle status, the number of attacking electronic control units (ECUs), and the sending frequency of attack data. It is only related to the voltage signals of external intrusion device. Once external intrusion devices send attack data to the CAN bus, VehicleEIDS can quickly identify its abnormal voltage signals, with the accuracy of more than 97%.

Although intelligent connected vehicles (ICVs) can better assist drivers and improve their driving experience, they have huge network security problems and are frequently attacked. This is because the vehicle network is connected to the Internet, which expands the attack surface of ICV, and attackers have more ways to launch attacks. In recent years, many security experts fight against attackers and propose various types of vehicle intrusion detection systems (IDSs) to protect the controller area network (CAN). However, with the continuous enhancement of attack means, especially the appearance of the masquerade attack, most IDSs are no longer applicable. In this article, we design a new fingerprint-based vehicle IDS to protect the CAN, called ClockIDS. It establishes a unique fingerprint for each electronic control unit (ECU) based on clock skew. On this basis, ClockIDS realizes the functions of intrusion detection and attack source identification by utilizing the empirical rule and dynamic time warping. It neither occupies the bandwidth of CAN bus nor needs to modify the CAN protocol. Our experiments on two real vehicles show that ClockIDS can establish a unique fingerprint for ECU without being affected by the size of message period, and can detect three types of attack with a detection accuracy of 98.63%. In addition, this system can identify the attack source, and the average recognition accuracy is 96.77%. Furthermore, ClockIDS has high real-time performance, and the average time cost of each detection is only 1.99 ms.

Nowadays, more and more external interfaces are added into intelligent and connected vehicles. The in-vehicle network, especially the controller area network (CAN), is no longer a closed environment, which provides more approaches for attackers to invade. To resist attacks, numerous researchers have proposed intrusion detection systems (IDSs). However, attackers can intrude CAN bus in a more advanced way, such as masquerade attack, which leads to failures of most IDS. To counter masquerade attacks, we propose an efficient vehicle IDS based on clock behavior, called VehicleCIDS. First, the system uses recursive least squares (RLS) algorithm to estimate the clock behavior of each electronic control unit (ECU). Then, a statistical method called empirical rule is used to detect attack messages. Finally, it utilizes dynamic time warping (DTW) to identify attackers. The experimental results on real vehicles show that the recognition rate of VehicleCIDS can achieve 98.52% in intrusion detection and 87.71% in attacker identification.

The industrial Internet of Things has become the new driving force for the automobile industry, making people's travel increasingly convenient. However, there are still a multitude of challenges that need to be tackled, including but not limited to illegal driver detection, legal driver identification, and driving behavior evaluation. At present, many researchers have attempted to solve issues of illegal driver detection and legal driver identification by using deep learning network, but there are still quite a few limitations in the collection and analysis of driving behavior data. Moreover, the problem of driving behavior evaluation has been paid little attention. Therefore, in this article we conduct a comprehensive study on driving behavior habits and establish a multitask learning (MTL) network to solve the abovementioned problems. First, we collect original data from a real vehicle and extract the driving behavior characteristics. Then, a novel MTL network composed of long short-term memory network, support vector domain description model and feedforward neural network is established, which achieves illegal driver detection, legal driver identification, and driving behavior evaluation. Extensive experiments illustrate that the proposed MTL network not only supports parallel learning to reduce time and space costs, but also has excellent performances and robustness for the three tasks.

Advanced technologies are constantly emerging in automobile industry, which not only provides drivers with a comfortable driving experience, but also enhances the safety of passengers. However, there are still some security issues need to be solved in automobiles, such as automobile driver fingerprinting. At present, identification technologies, such as fingerprint recognition and iris recognition, cannot monitor the driver's identity in real-time manner. Therefore, it is of great significance to design a real-time automobile driver fingerprinting scheme to ensure the safety of people's properties and even lives. Different from previous work concerning automobile driver fingerprinting, in this article, we conduct a comprehensive study on behavioral characteristics of drivers in two vehicles, namely Luxgen U5 SUV and Buick Regal. We exploit the actual data of the controller area network to construct a driver identity comparison library by extracting and processing the feature data. Then, we construct a combined model based on convolutional neural network and support vector domain description to achieve efficient automobile driver fingerprinting. Extensive experimental results show that the proposed driver fingerprinting scheme can dynamically match the driver's identity in real time without affecting the normal driving.

With the continuous expansion and deepening of Intelligent and Connected Vehicles (ICVs), advanced technology continues to emerge, making ICVs more intelligent to provide services for drivers and protect them. It is noticed that the emergence of almost all advanced technologies is based on the use of automotive data. Using automobile data can not only restore the current driving state, but also realize the identification of the driver. Different from previous works about driver identification, we don't use any manufacturer's Controller Area Network (CAN) protocol to parse vehicle's data and don't use any external sensor data. We first rely on the broadcast feature of the CAN bus, and use the automotive diagnostic tool to get all real-time data from the On-Board Diagnostic (OBD-II) port. Then, we use Feature scaling and Principal Component Analysis (PCA) algorithm to preprocess the data. Finally, we use k-Nearest Neighbor (k-NN) algorithm and Naive Bayes algorithm combined with voting mechanism to successfully identify the driver's identity. The experimental results show that the recognition rate of ten drivers is 100%.

As the mainstream of future automotive industry, Intelligent and Connected Vehicles (ICVs) have versatile connections between themselves and external devices, although able to provide more conveniences and better driving experiences for the users, also bring forward lots of intrusion portals for the malicious attackers. It is noticed that the final step of almost all attacks in available works, must be at the in-vehicle network, i.e., the CAN bus. Actually, the characteristics of CAN data, specifically, the broadcast transmission on the CAN bus, as well as the unencrypted authentication strategy make the CAN bus vulnerable to various attacks. Different from previous works about CAN bus, we present in this paper a comprehensive study on the in-vehicle network of a modern ICV (a Luxgen SUV), from the perspective of the vehicle auxiliary system. We first clarify the complicated communication process among the smart key, Body Control Module (BCM), and Key Control Unit (KCU), identify the loophole among the Luxgen auxiliary system, and then introduce a practical method to utilize this vulnerability. Finally, extensive experiments have been conducted on the Luxgen SUV where a wireless diagnostic equipment was utilized to achieve successful remote invasion in road tests.

The widespread use of Internet of Things (IoT), especially smart wearables, will play an important role in improving the quality of medical care, bringing convenience for patients and improving the management level of hospitals. However, due to the limitation of communication protocols, there exists non unified architecture that can connect all intelligent things in smart hospitals, which is made possible by the emergence of the Narrowband IoT (NB-IoT). In light of this, we propose an architecture to connect intelligent things in smart hospitals based on NB-IoT, and introduce edge computing to deal with the requirement of latency in medical process. As a case study, we develop an infusion monitoring system to monitor the real-time drop rate and the volume of remaining drug during the intravenous infusion. Finally, we discuss the challenges and future directions for building a smart hospital by connecting intelligent things.