Biblio

With the ever-increasing energy demand and enormous development of generation capacity, modern bulk power systems are mostly pushed to operate with narrower security boundaries. Therefore, timely and reliable assessment of power system security is an inevitable necessity to prevent widespread blackouts and cascading outages. In this paper, a new white-box decision tree-based preventive strategy is presented to evaluate and enhance the power system dynamic security versus the credible N-K contingencies originating from transient instabilities. As well, a competent operating measure is expertly defined to detect and identify the islanding and non-islanding conditions with the aid of a wide-area phasor measurement system. The newly developed strategy is outlined by a three-level simulation with the aim of guaranteeing the power system dynamic security. In the first-level, six hundred islanding and non-islanding scenarios are generated using an enhanced version of the ID3 algorithm, referred to as the C4.5 algorithms. In the second-level, optimal C4.5 decision trees are offline trained based on operating parameters achieved by the reduction error pruning method. In the third level, however, all trained decision trees are rigorously investigated offline and online; and then, the most accurate and reliable decision tree is selected. The newly developed strategy is examined on the IEEE New England 39-bus test system, and its effectiveness is assured by simulation studies.

Generative Adversarial Network (GAN) models have been widely used in various fields. More recently, styleGAN and styleGAN2 have been developed to synthesize faces that are indistinguishable to the human eyes, which could pose a threat to public security. But latest work has shown that it is possible to identify fakes using powerful CNN networks as classifiers. However, the reliability of these techniques is unknown. Therefore, in this paper we focus on the generation of content-preserving images from fake faces to spoof classifiers. Two GAN-based frameworks are proposed to achieve the goal in the white-box and black-box. For the white-box, a network without up/down sampling is proposed to generate face images to confuse the classifier. In the black-box scenario (where the classifier is unknown), real data is introduced as a guidance for GAN structure to make it adversarial, and a Real Extractor as an auxiliary network to constrain the feature distance between the generated images and the real data to enhance the adversarial capability. Experimental results show that the proposed method effectively reduces the detection accuracy of forensic models with good transferability.

Weakly supervised segmentation methods using bounding box annotations focus on obtaining a pixel-level mask from each box containing an object. Existing methods typically depend on a class-agnostic mask generator, which operates on the low-level information intrinsic to an image. In this work, we utilize higher-level information from the behavior of a trained object detector, by seeking the smallest areas of the image from which the object detector produces almost the same result as it does from the whole image. These areas constitute a bounding-box attribution map (BBAM), which identifies the target object in its bounding box and thus serves as pseudo ground-truth for weakly supervised semantic and instance segmentation. This approach significantly outperforms recent comparable techniques on both the PASCAL VOC and MS COCO benchmarks in weakly supervised semantic and instance segmentation. In addition, we provide a detailed analysis of our method, offering deeper insight into the behavior of the BBAM.

The most representative method of deep learning is data-driven learning. These methods are often data-dependent, and lack of data leads to poor learning. There is a GAN method that creates a likely image as a way to solve a problem that lacks data. The GAN determines that the discriminator is fake/real with respect to the image created so that the generator learns. However, overfitting problems when the discriminator becomes overly dependent on the learning data. In this paper, we explain overfitting problem when the discriminator decides to fake/real using xAI. Depending on the area of the described image, it is possible to limit the learning of the discriminator to avoid overfitting. By doing so, the generator can produce similar but more diverse images.

Inter-substation Generic Object Oriented Substation Events (GOOSE) communications that are used for critical protection functions have several cyber-security vulnerabilities. GOOSE messages are directly mapped to the Layer 2 Ethernet without network and transport layer headers that provide data encapsulation. The high-status-number attack is a malicious attack on GOOSE messages that allows hackers to completely take over intelligent electronic devices (IEDs) subscribing to GOOSE communications. The status-number parameter of GOOSE messages, stNum is tampered with in these attacks. Given the strict delivery time requirement of 3 ms for GOOSE messaging, it is infeasible to encrypt the GOOSE payload. This work proposes to secure the sensitive stNum parameter of the GOOSE payload using systematically encoded polynomial codes. Exploiting linear codes allows for the security features to be encoded in linear time, in contrast to complex hashing algorithms. At the subscribing IED, the security feature is used to verify that the stNum parameter has not been tampered with during transmission in the insecure medium. The decoding and verification using syndrome computation at the subscriber IED is also accomplished in linear time.

In an era of technological revolution in which everything becomes smarter and connected, the blockchain can introduce a new model for energy transactions able to grant more simplicity, security and transparency for end-users. The blockchain technology is characterized by a distributed architecture without a trusted and centralized authority, and, therefore, it appears as the perfect solutions for managing exchanges between peers. In this paper, a market algorithm that can be easily transferred to a smart contract for maximizing the match between produced and consumed energy in a micro-grid is presented. The algorithm supports energy transactions between peers (both producers and consumers) and could be one of the main executables implemented using a blockchain platform. The case study presented in this paper shows how the end-users through the blockchain could select among the possible energy transactions those more suitable to offer specific ancillary services to the grid operator without involving the grid operator itself or a third-party aggregator.

This thesis focuses on the detection of malware in industrial Internet. The basic flow of the detection of malware contains feature extraction and sample identification. API graph can effectively represent the behavior information of malware. However, due to the high algorithm complexity of solving the problem of subgraph isomorphism, the efficiency of analysis based on graph structure feature is low. Due to the different scales of API graph of different malicious codes, the API graph needs to be normalized. Considering the difficulties of sample collection and manual marking, it is necessary to expand the number of malware samples in industrial Internet. This paper proposes a method that combines PageRank with TF-IDF to process the API graph. Besides, this paper proposes a method to construct the adversarial samples of malwares based on GAN.

Trustworthy and secure operation of the cyber-power system calls for resilience against malicious and accidental failures. The objective of a resilient system is to withstand and recover operation of the system to supply critical loads despite multiple contingencies in the system. To take timely actions, we need to continuously measure the cyberphysical security of the system. We propose a cyber-physical security assessment metric (CP-SAM) based on quantitative factors affecting resiliency and utilizing concepts from graph theoretic analysis, probabilistic model of availability, attack graph metrics, and vulnerabilities across different layers of the microgrid system. These factors are integrated into a single metric using a multi-criteria decision making (MCDM) technique, Choquet Integral to compute CP-SAM. The developed metric will be valuable for i) monitoring the microgrid resiliency considering a holistic cyber-physical model; and ii) enable better decision-making to select best possible mitigation strategies towards resilient microgrid system. Developed CP-SAM can be extended for active distribution system and has been validated in a real-world power-grid test-bed to monitor the microgrid resiliency.

In this work, we examine the method of enabling computers to understand human interaction by constructing a generative conversational agent. An experimental approach in trying to apply the techniques of natural language processing using recurrent neural networks (RNNs) to emulate the concept of textual entailment or human reasoning is presented. To achieve this functionality, our experiment involves developing an integrated Long Short-Term Memory cell neural network (LSTM) system enhanced with an attention mechanism. The results achieved by the model are shown in terms of the number of epochs versus loss graphs as well as a brief illustration of the model's conversational capabilities.

This paper presents the design and the verification of an efficient image protection method based on the QR code, which is a type of two-dimensional barcode widely used in various fields. For this purpose, we design a new image protection system consisting of a secure image generator and a secure image recognizer. One adds a new pre-processing block to the typical QR code generator and the other combines the existing QR code reader with a new post-processing block. The new architecture provides image de-identification. It is also flexible, allowing the use of text-based compression and encryption. We have implemented prototype applications for verifying the functions of the secure image generator and those of the secure image recognizer. As a result, it is shown that the proposed architecture can be used as a good solution for image privacy protection, especially in offline environments.

True Random number Generator (TRNG) is critical for secure communications. In this work, we explain in details regarding our recent solution on TRNG using random telegraph noise (RTN) including the benefits and the disadvantages. Security check is performed using the NIST randomness tests for both the RTN-based TRNG and various conventional pseudo random umber generator. The newly-proposed design shows excellent randomness, power consumption, low design complexity, small area and high speed, making it a suitable candidate for future cryptographically secured applications within the internet of things.

In this paper is devoted methods for generating keys for cryptographic algorithms. Hash algorithms were analysed and learned linear and nonlinear. It was made up improved key generation algorithm and software.

VSS (Verifiable Secret Sharing) protocols are used in a number of block-chain systems, such as Dfinity and Ouroboros to generate unpredicted random number flow, they can be used to determine the proposer list and the voting powers of the voters at each height. To prevent random numbers from being predicted and attackers from corrupting a sufficient number of participants to violate the underlying trust assumptions, updatable VSS protocol in distributed protocols is important. The updatable VSS universal setup is also a hot topic in zkSNARKS protocols such as Sonic [19]. The way that we make it updatable is to execute the share exchange process repeatedly on chain, this process is challenging to be implemented in asynchronous network model, because it involves the wrong shares and the complaints, it requires the participant has the same view towards the qualified key generators, we take this process on chain and rely on BFT consensus mechanism to solve this. The group secret is thus updatable on chain. This is an enhancement to Dfinity. Therefore, even if all the coefficients of the random polynomials of epoch n are leaked, the attacker can use them only in epoch n+2. And the threshold group members of the DKG protocol can be updated along with the updates of the staked accounts and nodes.

Security preservation is considered as one of the major concerns in this digital world, mainly for performing any online transactions. As the time progress, it witnesses an enormous amount of security threats and stealing different kind of digital information over the online network. In this regard, lots of cryptographic algorithms based on secret key generation techniques have been implemented to boost up the security aspect of network systems that preserve the confidentiality of digital information. Despite this, intelligent intruders are still able to crack the key generation technique, thus stealing the data. In this research article, we propose an innovative approach for generating a pseudo-pseudo-random key sequence that serves as a base for the encryption/decryption process. The key generation process is carried out by extracting the essential features from a facial image and based on the extracted features; a pseudo-random key sequence that acts as a primary entity for the efficient encryption/decryption process is generated. Experimental findings related to the pseudo-random key is validated through chi-square, runs up-down and performs a period of subsequence test. Outcomes of these have subsequently passed in achieving an ideal key.

Present day's data security plays a vital role in digital human life. Data is a valuable asset to any organization and hence its security from external attacks is very important. Information security is not only an important aspect but essential, to secure data from unapproved access. Data encryption, decryption and key management are the key factors in data protection. It is very important to have the right data security solution to meet the challenging threats. Cryptosystem implementation and random number generators are crucial for Cryptosystem applications such as security applications, space applications, military applications and smart cards et al. In this paper, we present the implementation of hybrid cryptosystem based on the True Random number Generator, pseudo Random number Generator and whitening the data by using the ASIC design flow.

Directional Overcurrent Relays (DOCRs) play an essential role in the power system protection to guarantee the reliability, speed of relay operation and avoiding mal-trip in the primary and backup relays when unintentional fault conditions occur in the system. Moreover, the dual setting protection scheme is more efficient protection schemes for offering fast response protection and providing flexibility in the coordination of relay. In this paper, the Adaptive Modified Firefly Algorithm (AMFA) is used to determine the optimal coordination of dual setting DOCRs in the ring distribution system. The AMFA is completed by choosing the minimum value of pickup current (\textbackslashtextbackslashpmbI\textbackslashtextbackslashpmbP) and time dial setting (TDS). On the other hand, dual setting DOCRs protection scheme also proposed for operating in both forward and reverse directions that consisted of individual time current characteristics (TCC) curve for each direction. The previous method is applied to the ring distribution system network of PT. Pupuk Sriwidjaja by considering the fault on each bus. The result illustration that the AMFA within dual setting protection scheme is significantly reaching the optimized coordination and the relay coordination is certain for all simulation scenarios with the minimum operation. The AMFA has been successfully implemented in MATLAB software programming.

Various hardware security solutions have been developed recently to help counter hardware level attacks such as hardware Trojan, integrated circuit (IC) counterfeiting and intellectual property (IP) clone/piracy. However, existing solutions often provide specific types of protections. While these solutions achieve great success in preventing even advanced hardware attacks, the compatibility of among these hardware security methods are rarely discussed. The inconsistency hampers with the development of a comprehensive solution for hardware IC and IP from various attacks. In this paper, we develop a security primitive generator to help solve the compatibility issue among different protection techniques. Specifically, we focus on two modern IC/IP protection methods, logic locking and watermarking. A combined locking and watermarking technique is developed based on enhanced finite state machines (FSMs). The security primitive generator will take user-specified constraints and automatically generate an FSM module to perform both logic locking and watermarking. The generated FSM can be integrated into any designs for protection. Our experimental results show that the generator can facilitate circuit protection and provide the flexibility for users to achieve a better tradeoff between security levels and design overheads.

The completion and the correction of measurement data are the foundation of the ubiquitous power internet of things construction. However, data missing may occur during the data transporting process. Therefore, a model of missing load situation reconstruction based on the generative adversarial networks is proposed in this paper to overcome the disadvantage of depending on data of other relevant factors in conventional methods. Through the unsupervised training, the proposed model can automatically learn the complex features of loads that are difficult to model explicitly to fill the incomplete load data without using other relevant data. Meanwhile, a method of online correction is put forward to improve the robustness of the reconstruction model in different scenarios. The proposed method is fully data-driven and contains no explicit modeling process. The test results indicate that the proposed algorithm is well-matched for the various scenarios, including the discontinuous missing load reconstruction and the continuous missing load reconstruction even massive data missing. Specifically, the reconstruction error rate of the proposed algorithm is within 4% under the absence of 50% load data.

Language creates a compact representation of the world and allows the description of unlimited situations and objectives through compositionality. While these characterizations may foster instructing, conditioning or structuring interactive agent behavior, it remains an open-problem to correctly relate language understanding and reinforcement learning in even simple instruction following scenarios. This joint learning problem is alleviated through expert demonstrations, auxiliary losses, or neural inductive biases. In this paper, we propose an orthogonal approach called Hindsight Generation for Experience Replay (HIGhER) that extends the Hindsight Experience Replay approach to the language-conditioned policy setting. Whenever the agent does not fulfill its instruction, HIGhER learns to output a new directive that matches the agent trajectory, and it relabels the episode with a positive reward. To do so, HIGhER learns to map a state into an instruction by using past successful trajectories, which removes the need to have external expert interventions to relabel episodes as in vanilla HER. We show the efficiency of our approach in the BabyAI environment, and demonstrate how it complements other instruction following methods.

The Internet of Things technology has been used in a wide range of fields, ranging from industrial applications to individual lives. As a result, a massive amount of sensitive data is generated and transmitted by IoT devices. Those data may be accessed by a large number of complex users. Therefore, it is necessary to adopt an encryption scheme with access control to achieve more flexible and secure access to sensitive data. The Ciphertext Policy Attribute-Based Encryption (CP-ABE) can achieve access control while encrypting data can match the requirements mentioned above. However, the long ciphertext and the slow decryption operation makes it difficult to be used in most IoT devices which have limited memory size and computing capability. This paper proposes a modified CP-ABE scheme, which can implement the full security (adaptive security) under the access structure of AND gate. Moreover, the decryption overhead and the length of ciphertext are constant. Finally, the analysis and experiments prove the feasibility of our scheme.

Random numbers have a wide usage in the area of cryptography. In practice, pseudo random number generators are used in place of true random number generators, as regeneration of them may be required. Therefore because of generation methods of pseudo random number sequences, statistical randomness tests have a vital importance. In this paper, a randomness test suite is specified for long binary sequences. In literature, there are many randomness tests and test suites. However, in most of them, to apply randomness test, long sequences are partitioned into a certain fixed length and the collection of short sequences obtained is evaluated instead. In this paper, instead of partitioning a long sequence into fixed length subsequences, a concept of dynamic partitioning is introduced in accordance with the random variable in consideration. Then statistical methods are applied. The suggested suite, containing four statistical tests: Collision Tests, Weight Test, Linear Complexity Test and Index Coincidence Test, all of them work with the idea of dynamic partitioning. Besides the adaptation of this approach to randomness tests, the index coincidence test is another contribution of this work. The distribution function and the application of all tests are given in the paper.

The code length and rate of length-compatible polar codes can be adaptively adjusted and changed because of the special coding structure. In this paper, we propose a method to construct length-compatible polar codes by employing physical layer encryption technology. The deletion way of frozen bits and generator matrix are random, which makes polar codes more flexible and safe. Simulation analysis shows that the proposed algorithm can not only effectively improve the performance of length-compatible polar codes but also realize the physical layer security encryption of the system.

While machine learning technologies have been remarkably advanced over the past several years, one of the fundamental requirements for the success of learning-based approaches would be the availability of high-quality data that thoroughly represent individual classes in a problem space. Unfortunately, it is not uncommon to observe a significant degree of class imbalance with only a few instances for minority classes in many datasets, including network traffic traces highly skewed toward a large number of normal connections while very small in quantity for attack instances. A well-known approach to addressing the class imbalance problem is data augmentation that generates synthetic instances belonging to minority classes. However, traditional statistical techniques may be limited since the extended data through statistical sampling should have the same density as original data instances with a minor degree of variation. This paper takes a learning-based approach to data augmentation to enable effective network anomaly detection. One of the critical challenges for the learning-based approach is the mode collapse problem resulting in a limited diversity of samples, which was also observed from our preliminary experimental result. To this end, we present a novel "Divide-Augment-Combine" (DAC) strategy, which groups the instances based on their characteristics and augments data on a group basis to represent a subset independently using a generative adversarial model. Our experimental results conducted with two recently collected public network datasets (UNSW-NB15 and IDS-2017) show that the proposed technique enhances performances up to 21.5% for identifying network anomalies.

Machine learning techniques help to understand underlying patterns in datasets to develop defense mechanisms against cyber attacks. Multilayer Perceptron (MLP) technique is a machine learning technique used in detecting attack vs. benign data. However, it is difficult to construct any effective model when there are imbalances in the dataset that prevent proper classification of attack samples in data. In this research, we use UGR'16 dataset to conduct data wrangling initially. This technique helps to prepare a test set from the original dataset to train the neural network model effectively. We experimented with a series of inputs of varying sizes (i.e. 10000, 50000, 1 million) to observe the performance of the MLP neural network model with distribution of features over accuracy. Later, we use Generative Adversarial Network (GAN) model that produces samples of different attack labels (e.g. blacklist, anomaly spam, ssh scan) for balancing the dataset. These samples are generated based on data from the UGR'16 dataset. Further experiments with MLP neural network model shows that a balanced attack sample dataset, made possible with GAN, produces more accurate results than an imbalanced one.

With the increasing scale of network, attacks against network emerge one after another, and security problems become increasingly prominent. Network intrusion detection system is a widely used and effective security means at present. In addition, with the development of machine learning technology, various intelligent intrusion detection algorithms also start to sprout. By flexibly combining these intelligent methods with intrusion detection technology, the comprehensive performance of intrusion detection can be improved, but the vulnerability of machine learning model in the adversarial environment can not be ignored. In this paper, we study the defense problem of network intrusion detection system against adversarial samples. More specifically, we design a defense algorithm for NIDS against adversarial samples by using bidirectional generative adversarial network. The generator learns the data distribution of normal samples during training, which is an implicit model reflecting the normal data distribution. After training, the adversarial sample detection module calculates the reconstruction error and the discriminator matching error of sample. Then, the adversarial samples are removed, which improves the robustness and accuracy of NIDS in the adversarial environment.