Efficiently and precisely identifying small items on traffic highways using unmanned aerial vehicle (UAV) platforms with limited resources is a crucial yet challenging job. This research suggests a speedy, accurate, and component-optimized road surveillance model (SAC-RSM) for UAVs. This model addresses slow detection speed, limited detection of small objects, and deployment difficulties. First, we designed a super-resolution-assisted learning branch in the network to balance the model’s detection speed and accuracy. This branch learns the feature representation from low to high resolution. This branch uses multiscale feature fusion in the encoding stage to enhance the feature representation of small objects, thereby enhancing their detection accuracy. Second, to avoid the problem of cross-layer convolution, which results in the loss of fine-grained information and low-learning efficiency, we propose using the convolution-to-space-convolution (CSPC) module in the backbone network to improve model detection’s robustness. Third, to achieve real-time detection, we realized the model using the Huawei Ascend compute architecture for neural networks (CANNs) framework to enable automatic quantization and parallel inference acceleration. Finally, we deployed the accelerated model to the embedded platform Atlas 200I developer kit (DK) A2. Compared to the baseline model, the proposed method shows significant increases in mean average precision (mAP) values for the VisDrone and DroneVehicle data sets, with increases of 17.4% and 9.4%, respectively. The proposed method achieves frames/s (FPS) of 38.3, which is 2.1 times faster than the baseline model, meeting the requirement for high-performance real-time detection in a UAV environment.

With the rising security issues in the Internet of Things (IoT), IoT threat intelligence (short for ITI) raises more and more concern. However, the lack of ITI knowledge graphs hinders the sharing and utilization of ITI that is usually in the form of unstructured text data scattered around the Internet. In this article, we propose a knowledge extraction method vital to the construction of ITI knowledge graphs. We first build an ITI ontology based on existing security ontologies and knowledge bases, providing an organized schema to incorporate ITI text data. Secondly, we design a joint model for ITI entity and relation extraction, namely, ITIRel, based on TPLinkerplus. To assist ITIRel in ITI entity recognition, we introduce domain knowledge to help the model learn the semantics of IoT security terms, which also improves the accuracy of relation extraction. We further optimize the tagging scheme of TPLinkerplus to enhance joint extraction performance. Finally, due to the lack of annotated ITI text data, we combine manual annotation and data augmentation techniques to create a new ITI dataset. Experiment results show that ITIRel establishes the new state-of-the-art on the dataset, which implies that our knowledge extraction method is suitable for ITI.

Traditional solutions for human action recognition usually rely on sensor or video methods. However, these methods have some limitations, such as inconvenient portability, light intensity influence, privacy protection, etc. In this article, an RFID-based nonwearable human action recognition scheme is proposed. In order to reduce the occlusion effect of the human body on the signal and increase the diversity of the reflected signal, a tags array is constructed. The data of phase and RSSI are fused as feature data to enhance the diversity of data. Furthermore, a combined processing method is proposed to eliminate thermal noise generated by the equipment and reduce the interference caused by the environment. Then, an action segmentation algorithm is designed to align the RF signals of human action. Finally, an efficient human action signal classification model is constructed using the spatiotemporal graph convolutional neural network (STGCN). Extensive experiments demonstrate that the overall accuracy rate of the system for human action recognition is 92.8%. Compared with the comparative mainstream recognition algorithms, STGCN shows better classification performance in terms of identification precision. In addition, multimodal RFID data fusion also improves the accuracy of identification.

Aiming at the shortcomings of the current YOLOv3 model, such as large size, slow response speed, and difficulty in deploying to real devices, this paper reconstructs the target detection model YOLOv3, and proposes a new lightweight target detection network YOLOv3-promote: Firstly, the G-Module combined with the Depth-Wise convolution is used to construct the backbone network of the entire model, and the attention mechanism is introduced and added to perform weighting operations on each channel to get more key features and remove redundant features, thereby strengthening the identification ability of feature network model’s to distinguish target objects among background; Secondly, in order to delete some less important channels to achieve the effect of compressing the model size and improving the calculation speed, the size of the scaling factor gamma in the batch normalization layer is used; Finally, based on NVIDIA’s TensorRT framework model conversion and half-precision acceleration were carried out, and the accelerated model was successfully deployed on the embedded platform Jetson Nano. The performed KITTI experimental results show that the inference speed of our proposed method is about 5 times that of the original model, the parameter volume is reduced to one tenth, the mAP is increased from 86.1% of the original model to 93.1%, and the FPS reaches 25.5fps, realizing the requirements of real-time detection with high precision.

Summary Qualitative spatio‐temporal reasoning is an important problem in artificial intelligence and has been widely and successfully applied in geographic information system and spatio‐temporal database. Currently, action features can be found in spatio‐temporal domain and the existing spatio‐temporal formalisms are not suitable for dealing with dynamic spatio‐temporal knowledge. Thus, how to represent and reason dynamic spatio‐temporal knowledge has become an important research issue. In this article, we present a dynamic spatio‐temporal logic for representing and reasoning dynamic spatio‐temporal knowledge. is a natural combination of spatio‐temporal logic ‐8 based on ‐8 and propositional dynamic logic. Timed actions of can be considered as temporal terms, moving the regions of topological space from one time point to another. can capture actions that change spatial relations between regions over time. For a formula from , we present a construction of a Büchi tree automaton. At the same time, we prove that deciding the satisfiability problem of is an EXPTIME‐complete problem.

In order to solve the problem that the privacy preserving algorithm based on slicing technology is incapable of dealing with packet loss, this paper presents the redundancy algorithm for privacy preserving. The algorithm guarantees privacy by combining disturbance data and ensures redundancy via carrying hidden data. It also selects the routing tree that is generated by the CTP protocol as the routing path for data transmission. Through division at the source node, the method adds hidden information and disturbance data. This algorithm uses hidden data and adds perturbation data to improve the privacy preserving. Nonetheless, it can restore the original data when data are partly lost. According to the simulation via TOSSIM (TinyOS simulator), in the case of partial packet loss, the algorithm can completely restore the original data. Furthermore, the authors compared accuracy of proposed algorithm, probability of data reduction, data fitting degree, communication overhead, and PLR. As a result, it improves the reliability and privacy of data transmission while ensuring data redundancy.

The Radio Frequency Identification (RFID) technology has been integrated into healthcare systems for the purpose of improving healthcare management. However, people have concerns about the security and privacy of this kind of RFID systems. In order to solve the security problems faced by RFID‐based healthcare systems, a novel security authentication protocol based on Physical Unclonable Function (PUF) and Advanced Encryption Standard (AES) encryption algorithm is designed. The protocol uses PUF technology to output unique and random responses to different excitation inputs, encrypts the authentication information sent by the tag, and uses the AES encryption algorithm to encrypt the authentication information between the cloud database and the reader. At the same time, in the authentication process, once the communicating entity completes the identity authentication of the other two entities, it immediately starts to update the key. The security analysis and formal analysis of BAN (proposed by Burrows et al.) logic prove the security and correctness of the protocol. Analysis results show that the computation cost and security performance of the proposed protocol are better than the compared protocols. Our findings will contribute to further enhancing the security for RFID healthcare systems.

Description logic, as a logical foundation of knowledge representation and reasoning, plays an important role in the Semantic Web. In practical applications, many fields contain a large number of fuzzy spatio-temporal knowledge. With a large amount of fuzzy spatio-temporal knowledge and many corres ponding applications being incorporated into the Semantic Web, description logic becomes an effective method to solve the problem of fuzzy spatio-temporal knowledge representation and reasoning. Currently, many efforts have been done on fuzzy spatio-temporal extensions of description logics, and the literature on fuzzy spatio-temporal description logic has been booming. To address these issues and more importantly, in this paper, we provide a comprehensive survey of the research literature that applies description logics techniques in fuzzy spatio-temporal representation and reasoning. The paper serves as helping readers grasp the main results and highlighting the direction of fuzzy spatio-temporal representation and reasoning based on description logics.

Radio frequency identification technology (RFID) is widely used in identity authentication and payment, and it also becomes an indispensable part of daily life. Cloud-based RFID systems have broad application prospects, and can be provided as a service to individuals or organisations. For example, RFID cards can be used for cash-less payment, physical access control, temporary rights and identification in cloud environment. When an RFID card is used, there is a wireless transaction between the card and its reader, which could be attacked by several methods, including a relay attack. Relay attacks are difficult to completely prevent and a serious threat to RFID systems security. An attacker could use limited resources to build up this kind of attack and may need little knowledge of the underlying protocol. In recent years, researchers have proposed solutions using second channels to resist relay attack, such as using environmental measurements including noise, light and temperature. This paper describes research on the defence techniques for relay attacks in cloudbased RFID systems. The cloud-based architecture for RFID systems typically consists of RFID tags, card readers (fixed or mobile) and cloud-based server functionality.

Secure localization has become very important in wireless sensor networks. However, the conventional secure localization algorithms used in wireless sensor networks cannot deal with internal attacks and cannot identify malicious nodes. In this paper, a localization based on trust valuation, which can overcome a various attack types, such as spoofing attacks and Sybil attacks, is presented. The trust valuation is obtained via selection of the property set, which includes estimated distance, localization performance, position information of beacon nodes, and transmission time, and discussion of the threshold in the property set. In addition, the robustness of the proposed model is verified by analysis of attack intensity, localization error, and trust relationship for three typical scenes. The experimental results have shown that the proposed model is superior to the traditional secure localization models in terms of malicious nodes identification and performance improvement.