Biblio

Scene text detection methods have developed greatly in the past few years. However, due to the limitation of the diversity of the text background of natural scene, the previous methods often failed when detecting more complicated text instances (e.g., super-long text and arbitrarily shaped text). In this paper, a text detection method based on multi -channel bounding box fusion is designed to address the problem. Firstly, the convolutional neural network is used as the basic network for feature extraction, including shallow text feature map and deep semantic text feature map. Secondly, the whole convolutional network is used for upsampling of feature map and fusion of feature map at each layer, so as to obtain pixel-level text and non-text classification results. Then, two independent text detection boxes channels are designed: the boundary box regression channel and get the bounding box directly on the score map channel. Finally, the result is obtained by combining multi-channel boundary box fusion mechanism with the detection box of the two channels. Experiments on ICDAR2013 and ICDAR2015 demonstrate that the proposed method achieves competitive results in scene text detection.

For the purpose of stealthiness, trigger-based Hardware Trojans(HTs) tend to have at least one trigger signal with an extremely low transition probability to evade the functional verification. In this paper, we discuss the correlation between poor testability and low transition probability, and then propose a kind of systematic Trojan trigger model with extremely low transition probability but reasonable testability, which can disable the Controllability and Observability for hardware Trojan Detection (COTD) technique, an efficient HT detection method based on circuits testability. Based on experiments and tests on circuits, we propose that the more imbalanced 0/1-controllability can indicate the lower transition probability. And a trigger signal identification method using the imbalanced 0/1-controllability is proposed. Experiments on ISCAS benchmarks show that the proposed method can obtain a 100% true positive rate and average 5.67% false positive rate for the trigger signal.

Nowadays malicious vendors can easily insert hardware Trojans into integrated circuit chips as the entire integrated chip supply chain involves numerous design houses and manufacturers on a global scale. It is thereby becoming a necessity to expose any possible hardware Trojans, if they ever exist in a chip. A typical Trojan circuit is made of a trigger and a payload that are interconnected with a trigger net. As trigger net can be viewed as the signature of a hardware Trojan, in this paper, we propose a gate-level hardware Trojan detection method and model that can be applied to screen the entire chip for trigger nets. In specific, we extract the trigger-net features for each net from known netlists and use the machine learning method to train multiple detection models according to the trigger modes. The detection models are used to identify suspicious trigger nets from the netlist of the integrated circuit under detection, and score each net in terms of suspiciousness value. By flagging the top 2% suspicious nets with the highest suspiciousness values, we shall be able to detect majority hardware Trojans, with an average accuracy rate of 96%.

There is an increasing concern about possible hostile modification done to ICs, which are used in various critical applications. Such malicious modifications are referred to as Hardware Trojan. A novel procedure to detect these malicious Trojans using Genetic algorithm along with the logical masking technique which masks the Trojan module when embedded is presented in this paper. The circuit features such as transition probability and SCOAP are used as suitable parameters to identify the rare nodes which are more susceptible for Trojan insertion. A set of test patterns called optimal test patterns are generated using Genetic algorithm to claim that these test vectors are more feasible to detect the presence of Trojan in the circuit under test. The proposed methodologies are validated in accordance with ISCAS '85 and ISCAS '89 benchmark circuits. The experimental results proven that it achieves maximum Trigger coverage, Trojan coverage and is also able to successfully mask the inserted Trojan when it is triggered by the optimal test patterns.

Commodity hypervisors are widely deployed to support virtual machines (VMs) on multiprocessor hardware. Their growing complexity poses a security risk. To enable formal verification over such a large codebase, we introduce microverification, a new approach that decomposes a commodity hypervisor into a small core and a set of untrusted services so that we can prove security properties of the entire hypervisor by verifying the core alone. To verify the multiprocessor hypervisor core, we introduce security-preserving layers to modularize the proof without hiding information leakage so we can prove each layer of the implementation refines its specification, and the top layer specification is refined by all layers of the core implementation. To verify commodity hypervisor features that require dynamically changing information flow, we introduce data oracles to mask intentional information flow. We can then prove noninterference at the top layer specification and guarantee the resulting security properties hold for the entire hypervisor implementation. Using microverification, we retrofitted the Linux KVM hypervisor with only modest modifications to its codebase. Using Coq, we proved that the hypervisor protects the confidentiality and integrity of VM data, while retaining KVM’s functionality and performance. Our work is the first machine-checked security proof for a commodity multiprocessor hypervisor.

In order to improve the continuity of cryptographic services and ensure the quality of services in the cloud environment, a dynamic migration framework of cryptographic service virtual machines based on the network shared storage system is proposed. Based on the study of the security threats in the migration process, a dynamic migration attack model is established, and the security requirement of dynamic migration is analyzed. It designs and implements the dynamic security migration management software, which includes a dynamic migration security enhancement module based on the Libvirt API, role-based access control policy, and transmission channel protection module. A cryptographic service virtual machine migration environment is built, and the designed management software and security mechanism are verified and tested. The experimental results show that the method proposed in the paper can effectively improve the security of cryptographic service virtual machine migration.

Surveillance videos can capture anomalies in real scenarios and play an important role in security systems. Anomaly events are unpredictable, which reflect the unsupervised nature of the problem. In addition, it is difficult to construct a complete video dataset which contains all normal events. Based on the diversity of normal events, this paper proposes a memory-enhanced unsupervised method for anomaly detection. The proposed method reconstructs video events by combining prototype features and encoded features to detect anomaly events. Furthermore, a memory module is introduced to better store the prototype patterns of normal events. Experimental results in various benchmark datasets demonstrate the effectiveness and robustness of the proposed method.

The growth of inter-dependency intricacies of Supervisory Control and Data Acquisition (SCADA) systems in industrial operations generates a likelihood of increased vulnerability to malicious threats and machine learning approaches have been extensively utilized in the research for vulnerability detection. Nonetheless, to improve security, an enhanced vulnerability detection using hyper-parameter-tune machine learning is proposed for early detection, classification and mitigation of SCADA communication and transmission networks by classifying benign, or malicious DNS attacks. The proposed scheme, an ensemble optimizer (GentleBoost) upon hyper-parameter tuning, gave a comparative achievement. From the simulation results, the proposed scheme had an outstanding performance within the shortest possible time with an accuracy of 99.49%, 99.23% for precision, and a recall rate of 99.75%. Also, the model was compared to other contemporary algorithms and outperformed all the other algorithms proving to be an approach to keep abreast of the SCADA network vulnerabilities and attacks.

In recent years, the complexity of software vulnera-bilities has continued to increase. Manual vulnerability detection methods alone no longer meet the demand. With the rapid development of the deep learning, many neural network models have been widely applied to source code vulnerability detection. The variant of recurrent neural network (RNN), bidirectional Long Short-Term Memory (BiLSTM), has been a popular choice in vulnerability detection. However, is BiLSTM the most suitable choice? To answer this question, we conducted a series of experiments to investigate the effectiveness of different neural network models for source code vulnerability detection. The results shows that the variants of RNN, gated recurrent unit (GRU) and bidirectional GRU, are more capable of detecting source code fragments with mixed vulnerability types. And the concatenated convolutional neural network is more capable of detecting source code fragments of single vulnerability types.

Vulnerabilities in the source code of software are critical issues in the realm of software engineering. Coping with vulnerabilities in software source code is becoming more challenging due to several aspects of complexity and volume. Deep learning has gained popularity throughout the years as a means of addressing such issues. In this paper, we propose an evaluation of vulnerability detection performance on source code representations and evaluate how Machine Learning (ML) strategies can improve them. The structure of our experiment consists of 3 Deep Neural Networks (DNNs) in conjunction with five different source code representations; Abstract Syntax Trees (ASTs), Code Gadgets (CGs), Semantics-based Vulnerability Candidates (SeVCs), Lexed Code Representations (LCRs), and Composite Code Representations (CCRs). Experimental results show that employing different ML strategies in conjunction with the base model structure influences the performance results to a varying degree. However, ML-based techniques suffer from poor performance on class imbalance handling when used in conjunction with source code representations for software vulnerability detection.

Vulnerability detection identifies defects in various commercial software. Because most vulnerability detection methods are based on the source code, they are not useful if the source code is unavailable. In this paper, we propose a binary vulnerability detection method and use our tool named BVD that extracts binary features with the help of an intermediate language and then detects the vulnerabilities using an embedding model. Sufficiently robust features allow the binaries compiled in cross-architecture to be compared. Consequently, a similarity evaluation provides more accurate results.

The security of Internet of Things (IoT) devices has become a matter of great concern in recent years. The existence of security holes in the executable programs in the IoT devices has resulted in difficult to estimate security risks. For a long time, vulnerability detection is mainly completed by manual debugging and analysis, and the detection efficiency is low and the accuracy is difficult to guarantee. In this paper, the mainstream automated vulnerability analysis methods in recent years are studied, and a vulnerability detection algorithm based on symbol execution is presented. The detection algorithm is suitable for lightweight applications in small and medium-sized IoT devices. It realizes three functions: buffer overflow vulnerability detection, encryption reliability detection and protection state detection. The robustness of the detection algorithm was tested in the experiment, and the detection of overflow vulnerability program was completed within 2.75 seconds, and the detection of encryption reliability was completed within 1.79 seconds. Repeating the test with multiple sets of data showed a small difference of less than 6.4 milliseconds. The results show that the symbol execution detection algorithm presented in this paper has high detection efficiency and more robust accuracy and robustness.

The recent events regarding worldwide human health sped up research efforts and resulted in the tremendous development of new technologies and applications. The last decade proved that new technologies find a proper place in worldwide human health and wellbeing, therefore the security of data during wireless transmission in medical facilities and for medical devices has become a research area of considerable importance. To provide enhanced security using conventional visible light wireless communication, we propose in this paper a novel communication protocol based on asymmetric encryption with a private key. We base the wireless communication protocol described in this work on a data encryption method using block chipers, and we propose it for medical facilities and devices with visible light transmission technology embedded. The asymmetric encryption with a private key algorithm, as part of a transmission protocol, aim to assure the security of sensitive medical data during wireless communication.

Recently, the physical layer security (PLS) is becoming an important research area for visible light communication (VLC) systems. In this paper, the secrecy rate performance is investigated for an indoor multi-user visible light communication (VLC) system using artificial noise (AN). In the considered model, all users simultaneously communicate with the legitimate receiver under wiretap channels. The legitimate receiver uses the minimum mean squared error (MMSE) equalizer to detect the received signals. Both lower bound and upper bound of the secrecy rate are obtained for the case that users' signals are uniformly distributed. Simulation results verify the theoretical findings and show the system secrecy rate performance for various positions of illegal eavesdropper.

A grouping-proof protocol aims to generate an evidence that two or more RFID (Radio Frequency Identification) tags in a group are coexistent, which has been widely deployed in practical scenarios, such as healthcare, supply-chain management, and so on. However, existing grouping-proof protocols have many issues in security and efficiency, either incompatible with EPCglobal Class-1 Generation-2 (C1G2) standard, or vulnerable to different attacks. In this paper, we propose a lightweight grouping-proof protocol which only utilizes bitwise operations (AND, XOR) and 128-bit pseudorandom number generator (PRNG). 2-round interactions between the reader and the tags allow them to cooperate on fast authentication in parallel mode where the reader broadcasts its round messages rather than hang on for the prior tag and then fabricate apposite output for the next tag consecutively. Our design enables the reader to aggregate the first round proofs (to bind the membership of tags in the same group) generated by the tags to an authenticator of constant size (independent of the number of tags) that can then be used by the tags to generate the second round proofs (and that will be validated by the verifier). Formal security (i.e., PPT adversary cannot counterfeit valid grouping-proof that can be accepted by any verifier) of the proposed protocol relies on the hardness of the learning parity with noise (LPN) problem, which can resist against quantum computing attacks. Other appealing features (e.g., robustness, anonymity, etc.) are also inspected. Performance evaluation shows its applicability to C1G2 RFID.

The goal of the edge computing framework is to solve the problem of management and control in the access of massive 5G terminals in the power Internet of things. Firstly, this paper analyzes the needs of IOT agent in 5G ubiquitous connection, equipment management and control, intelligent computing and other aspects. In order to meet with these needs, paper develops the functions and processes of the edge computing framework, including unified access of heterogeneous devices, protocol adaptation, edge computing, cloud edge collaboration, security control and so on. Finally, the performance of edge computing framework is verified by the pressure test of 5G wireless ubiquitous connection.

In the era of machine learning, mobile users are able to submit their symptoms to doctors at any time, anywhere for personal diagnosis. It is prevalent to exploit edge computing for real-time diagnosis services in order to reduce transmission latency. Although data-driven machine learning is powerful, it inevitably compromises privacy by relying on vast amounts of medical data to build a diagnostic model. Therefore, it is necessary to protect data privacy without accessing local data. However, the blossom has also been accompanied by various problems, i.e., the limitation of training data, vulnerabilities, and privacy concern. As a solution to these above challenges, in this paper, we design a lightweight privacy-preserving medical diagnosis mechanism on edge. Our method redesigns the extreme gradient boosting (XGBoost) model based on the edge-cloud model, which adopts encrypted model parameters instead of local data to reduce amounts of ciphertext computation to plaintext computation, thus realizing lightweight privacy preservation on resource-limited edges. Additionally, the proposed scheme is able to provide a secure diagnosis on edge while maintaining privacy to ensure an accurate and timely diagnosis. The proposed system with secure computation could securely construct the XGBoost model with lightweight overhead, and efficiently provide a medical diagnosis without privacy leakage. Our security analysis and experimental evaluation indicate the security, effectiveness, and efficiency of the proposed system.

Due to broadcast nature of acoustic signal, underwater acoustic sensor networks face security challenge. In the paper, we propose a physical layer security transmission scheme with cooperative jamming. The proposed scheme takes advantage of the long propagation delay of the underwater acoustic channel to interfere with eavesdropper without affecting the reception of intended users. The results of both simulation and field experiment show that the proposed mechanism can improve the secrecy capacity of the network and effectively jam eavesdropper.

The ongoing expansion of underwater Internet of Things techniques promote diverse categories of maritime intelligent systems, e.g., Underwater Acoustic Sensor Networks (UASNs), Underwater Wireless Networks (UWNs), especially multiple Autonomous Underwater Vehicle (AUV) based UWNs have produced many civil and military applications. To enhance the network management and scalability, in this paper, the technique of Software-Defined Networking (SDN) technique is introduced, leading to the paradigm of Software-Defined multi-AUV-based UWNs (SD-UWNs). With SD-UWNs, the network architecture is divided into three functional layers: data layer, control layer, and application layer, and the network administration is re-defined by a framework of software-defined beacon. To manage the network, a control model based on artificial potential field and network topology theory is constructed. On account of the efficient data sharing ability of SD-UWNs, a proactive alarming-enabled path planning scheme is proposed, wherein all potential categories of obstacle avoidance scenes are taken into account. Evaluation results indicate that the proposed SD-UWN is more efficient in scheduling the cooperative network function than the traditional approaches and can secure exact path planning.

Underwater wireless optical communications are studied through single photon detection, photon states modulation and quantum key encryption. These studies will promote the development of optical communication applications in underwater vehicles and underwater sensor networks.

Incipient faults are faults at their initial stages and occur before permanent faults occur. It is very important to detect incipient faults timely and accurately for the safe and stable operation of the power system. At present, most of the detection methods for incipient faults are designed for the detection of a single device’s incipient fault, but a unified detection for multiple devices cannot be achieved. In order to increase the fault detection capability and enable detection expandability, this paper proposes a waveform vector embedding (WVE) method to embed incipient fault waveforms of different devices into waveform vectors. Then, we utilize the waveform vectors and formulate them into a waveform dictionary. To improve the efficiency of embedding the waveform signature into the learning process, we build a loss function that prevents overflow and overfitting of softmax function during when learning power system waveforms. We use the real data collected from an IEEE Power & Energy Society technical report to verify the feasibility of this method. For the result verification, we compare the superiority of this method with Logistic Regression and Support Vector Machine in different scenarios.

In this work, an IF(intermediate frequency) module of a hyperspectral microwave radiometer based on a polyphase filter bank (PFB) and Discrete Fourier Transformation (DFT)is introduced. The IF module is designed with an 800MSPS sampling-rate ADC and a Xilinx Virtex-7 FPGA. The module can achieve 512 channels and a bandwidth of 400M and process all the sampled data in real-time. The test results of this module are given and analyzed, such as linearity, accuracy, etc. It can be used in various applications of microwave remote sensing. The system has strong expandability.

Accidents in Nuclear Power Plants (NPPs) can occur for a variety of causes. However, among these, the scale of accidents due to human error can be greater than expected. Accordingly, researches are being actively conducted using artificial intelligence to reduce human error. Most of the research shows high performance based on the numerical data on NPPs, but the expandability of researches using only numerical data is limited. Therefore, in this study, abnormal diagnosis was performed using artificial intelligence based on image data. The methods applied to abnormal diagnosis are the deep neural network, convolution neural network, and convolution recurrent neural network. Consequently, in nuclear power plants, it is expected that the application of more methodologies can be expanded not only in numerical data but also in image-based data.

Internet of things (IoT) healthcare devices, like other IoT devices, typically use proprietary protocol communications. Usually, these proprietary protocols are not audited and may present security flaws. Further, new proprietary protocols are desgined in the field of IoT devices, like data-over-sound communications. Data-over-sound is a new method of communication based on audio with increasing popularity due to its low hardware requirements. Only a speaker and a microphone are needed instead of the specific antennas required by Bluetooth or Wi-Fi protocols. In this paper, we analyze, audit and reverse engineer a modern IoT healthcare device used for performing electrocardiograms (ECG). The audited device is currently used in multiple hospitals and allows remote health monitoring of a patient with heart disease. For this auditing, we follow a black-box reverse-engineering approach and used STRIDE threat analysis methodology to assess all possible attacks. Following this methodology, we successfully reverse the proprietary data-over-sound protocol used by the IoT healthcare device and subsequently identified several vulnerabilities associated with the device. These vulnerabilities were analyzed through several experiments to classify and test them. We were able to successfully manipulate ECG results and fake heart illnesses. Furthermore, all attacks identified do not need any patient interaction, being this a transparent process which is difficult to detect. Finally, we suggest several short-term solutions, centred in the device isolation, as well as long-term solutions, centred in involved encryption capabilities.

Blockchain is well known for its storage consistency, decentralization and tamper-proof, but the privacy disclosure and difficulty in auditing discourage the innovative application of blockchain technology. As compared to public blockchain and private blockchain, consortium blockchain is widely used across different industries and use cases due to its privacy-preserving ability, auditability and high transaction rate. However, the present co-audit of privacy-preserving data on consortium blockchain is inefficient. Private data is usually encrypted by a session key before being published on a consortium blockchain for privacy preservation. The session key is shared with transaction parties and auditors for their access. For decentralizing auditorial power, multiple auditors on the consortium blockchain jointly undertake the responsibility of auditing. The distribution of the session key to an auditor requires individually encrypting the session key with the public key of the auditor. The transaction initiator needs to be online when each auditor asks for the session key, and one encryption of the session key for each auditor consumes resources. This work proposes GAChain and applies group key agreement technology to efficiently co-audit privacy-preserving data on consortium blockchain. Multiple auditors on the consortium blockchain form a group and utilize the blockchain to generate a shared group encryption key and their respective group decryption keys. The session key is encrypted only once by the group encryption key and stored on the consortium blockchain together with the encrypted private data. Auditors then obtain the encrypted session key from the chain and decrypt it with their respective group decryption key for co-auditing. The group key generation is involved only when the group forms or group membership changes, which happens very infrequently on the consortium blockchain. We implement the prototype of GAChain based on Hyperledger Fabric framework. Our experimental studies demonstrate that GAChain improves the co-audit efficiency of transactions containing private data on Fabric, and its incurred overhead is moderate.