Search

Motivated by modern network communication applications which require low latency, we study codes that correct erasures with low decoding delay. We provide a simple explicit construction that yields convolutional codes that can correct both burst and arbitrary erasures under a maximum decoding delay constraint $T$. Our proposed code has efficient encoding/decoding algorithms and requires a field size that is linear in $T$. We study the performance of our code over the Gilbert-Elliot channel; our simulation results show significant performance gains over low-delay codes existing in the literature., Comment: Accepted for publication in IEEE ICASSP 2023

Proactive failure detection of instances is vitally essential to microservice systems because an instance failure can propagate to the whole system and degrade the system's performance. Over the years, many single-modal (i.e., metrics, logs, or traces) data-based nomaly detection methods have been proposed. However, they tend to miss a large number of failures and generate numerous false alarms because they ignore the correlation of multimodal data. In this work, we propose AnoFusion, an unsupervised failure detection approach, to proactively detect instance failures through multimodal data for microservice systems. It applies a Graph Transformer Network (GTN) to learn the correlation of the heterogeneous multimodal data and integrates a Graph Attention Network (GAT) with Gated Recurrent Unit (GRU) to address the challenges introduced by dynamically changing multimodal data. We evaluate the performance of AnoFusion through two datasets, demonstrating that it achieves the F1-score of 0.857 and 0.922, respectively, outperforming the state-of-the-art failure detection approaches.

Future networks are expected to support various ultra-reliable low-latency communications via wireless links. To avoid the loss of packets and keep the low latency, sliding network coding (SNC) is an emerging technology by generating redundant packets that are the linear combination of the original data packets from the current block and some previous blocks. However, how to take the advantage of re-transmission for SNC is still an open problem since higher reliability could be achieved at the expense of large latency caused by round-trip time (RTT). To deal with this issue, in this paper, we consider the idea of adjusting the transmission phase and the number of the redundant packets for SNC with re-transmission. Specifically, If RTT is large, most of the redundant packets are sent at the first transmission, otherwise, re-transmission will be used. We first derive a concise and tight lower bound of the block error probability of SNC without re-transmission. Then, based on the bound, the theoretical expressions of the proposed re-transmission schemes are derived regarding the block error probability, the average code length, and the average packet latency. Results show that the proposed SNC with re-transmission improves block error probability and keeps the low latency.

The increased use of social media among digitally anonymous users, sharing their thoughts and opinions, can facilitate participation and collaboration. However, this anonymity feature which gives users freedom of speech and allows them to conduct activities without being judged by others can also encourage cyberbullying and hate speech. Predators can hide their identity and reach a wide range of audience anytime and anywhere. According to the detrimental effect of cyberbullying, there is a growing need for cyberbullying detection approaches. In this survey paper, a comparative analysis of the automated cyberbullying techniques from different perspectives is discussed including data annotation, data preprocessing, and feature engineering. In addition, the importance of emojis in expressing emotions as well as their influence on sentiment classification and text comprehension leads us to discuss the role of incorporating emojis in the process of cyberbullying detection and their influence on the detection performance. Furthermore, the different domains for using self-supervised learning (SSL) as an annotation technique for cyberbullying detection are explored.

Due to the limited local computation resource, industrial vehicular computation requires offloading the computation tasks with time-delay sensitive and complex demands to other intelligent devices (IDs) once the data is sensed and collected collaboratively. This paper considers offloading partial computation tasks of the industrial vehicles (IVs) to multiple available IDs of the industrial mobile edge computing (MEC), including unmanned aerial vehicles (UAVs), and the fixed-position MEC servers, to optimize the system cost including execution time, energy consumption, and the ID rental price. Moreover, to increase the access probability of IV by the UAVs, the geographical area is divided into small partitions and schedule the UAVs regarding the regional IV density dynamically. A minimum incremental task allocation (MITA) algorithm is proposed to divide the whole task and assign the divided units for the minimum cost increment each time. Experimental results show the proposed solution can significantly reduce the system cost.

Zero-shot voice conversion (VC) where both source and target speakers are unseen in the training dataset has become a new research direction. Using speaker embeddings instead of one-hot vectors to represent speaker identity is a key point, which makes VC models work on unseen speakers. In our work, a newly designed neural network was used to adjust the speaker embeddings of unseen speakers. This enables speaker embeddings to perform better on zero-shot VC. In addition, disentangled representation of features is the mainstream method to achieve zero-shot VC. In terms of input features of VC model, we use Mel-cepstral and F0 as simple acoustic features (SAF) rather than Mel-spectrograms. This avoids F0 conflicts in decoder that existed in the previous methods. The evaluations demonstrate that our proposed methods improve the quality of converted speech in terms of naturalness and similarity.

Traditional text emotion analysis methods are primarily devoted to studying extended texts, such as news reports and full-length documents. Microblogs are considered short texts that are often characterized by large noises, new words, and abbreviations. Previous emotion classification methods usually fail to extract significant features and achieve poor classification effect when applied to processing of short texts or micro-texts. This study proposes a microblog emotion classification model, namely, CNN_Text_Word2vec, on the basis of convolutional neural network (CNN) to solve the above-mentioned problems. CNN_Text_Word2vec introduces a word2vec neural network model to train distributed word embeddings on every single word. The trained word vectors are used as input features for the model to learn microblog text features through parallel convolution layers with multiple convo-lution kernels of different sizes. Experiment results show that the overall accuracy rate of CNN_Text_Word2vec is 7.0% higher than that achieved by current mainstream methods, such as SVM, LSTM and RNN. Moreover, this study explores the impact of different semantic units on the accuracy of CNN_Text_Word2vec, specifically in processing of Chinese texts. The experimental results show that comparing to using feature vectors obtained from training words, feature vector obtained from training Chinese characters yields a better performance.

Recently, the research of Internet of Things (IoT) and Multimedia Big Data (MBD) has been growing tremendously. Both IoT and MBD have a lot of multimedia data, which can be tampered easily. Therefore, the research of multimedia forensics is necessary. Copy-move is an important branch of multimedia forensics. In this paper, a novel copy-move forgery detection scheme using combined features and transitive matching is proposed. First, SIFT and LIOP are extracted as combined features from the input image. Second, transitive matching is used to improve the matching relationship. Third, a filtering approach using image segmentation is proposed to filter out false matches. Fourth, affine transformations are estimated between these image patches. Finally, duplicated regions are located based on those affine transformations. The experimental results demonstrate that the proposed scheme can achieve much better detection results on the public database under various attacks.

The increasing demand for portable electronic devices and hybrid electric vehicles stimulates the development of supercapacitors as an advanced energy storage system. Here, we demonstrate a binder-free nickel hydroxide@nanoporous gold/Ni foam (Ni(OH)2@NPG/Ni foam) electrode for high-performance supercapacitors, which is prepared by a facile three-step fabrication route including electrodeposition of Au-Sn alloy on Ni foam, chemical dealloying of Sn and electrodepostion of Ni(OH)2 on NPG/Ni foam. Such Ni(OH)2@ NPG/Ni foam electrode is composed of a thin layer of conformable Ni(OH)2 nanoflakes supported on three-dimensional (3D) hierarchically porous NPG/Ni foam substrate. The resulting Ni(OH)2@NPG/Ni foam electrode can offer highways for both electron transfer and ion transport and lead to an excellent electrochemical performance with an ultrahigh specific capacitance of 3,380 F g–1 at a current density of 2 A g–1. Even when the current density was increased to 50 A g–1, it still retained a high capacitance of 1,927 F g–1. The promising performance of the Ni(OH)2@NPG/Ni foam electrode is mainly ascribed to the 3D hierarchical porosity and the highly conductive network on the NPG/Ni foam composite current collector, as well as the conformal electrodeposition of Ni(OH)2 active material on the NPG/Ni foam, which induces the formation of interconnected porosity both on the top surface and on the inner surface of the electrode. This inspiring electrochemical performance would make the as-designed electrode material become one of the most promising candidates for future electrochemical energy storage systems.

The unbroken amplfication of a versatile urban setup is challenged by huge Big Data processing. Understanding the voluminous data generated in a smart urban environment for decision making is a challenging task. Big Data analytics is performed to obtain useful insights about the massive data. The existing conventional techniques are not suitable to get a useful insight due to the huge volume of data. Big Data analytics has attracted significant attention in the context of large-scale data computation and processing. This paper presents a Hadoop-based architecture to deal with Big Data loading and processing. The proposed architecture is composed of two different modules, i.e., Big Data loading and Big Data processing. The performance and efficiency of data loading is tested to propose a customized methodology for loading Big Data to a distributed and processing platform, i.e., Hadoop. To examine data ingestion into Hadoop, data loading is performed and compared repeatedly against different decisions. The experimental results are recorded for various attributes along with manual and traditional data loading to highlight the efficiency of our proposed solution. On the other hand, the processing is achieved using YARN cluster management framework with specific customization of dynamic scheduling. In addition, the effectiveness of our proposed solution regarding processing and computation is also highlighted and decorated in the context of throughput.

With the rapid pervasion of location-based services (LBSs), protection of location privacy has become a significant concern. In most continuous LBSs' privacy-preserving solutions, users need to transmit the location query data to an untrusted location service provider (LSP) to obtain query results, and the users discard these results immediately after using them. This results in an ineffective use of these results by future queries and in turn leads to a higher risk to user privacy from the LSP. To address these issues, we generally use caching to cache the query results for users' future queries. However, the minimization of the interaction between users and LSPs is a challenge. In this paper, we propose an enhanced user privacy scheme through caching and spatial K-anonymity (CSKA) in continuous LBSs; it adopts multi-level caching to reduce the risk of exposure of users' information to untrusted LSPs. In continuous LBS queries, our scheme first utilizes the Markov model to predict the next query location according to the user mobility. Then, according to the predicted location, cell's cache contribution rate, and data freshness, an algorithm for forming spatial K-anonymity is designed to improve the user's cache hit rate and enhance the user location privacy. The security analysis and simulation results demonstrate that our proposed CSKA scheme can provide higher privacy protection than a few previous methods, and it can minimize the overhead of the LBS server.

Malware writers frequently try to hide the activities of their agents within tunnelled traffic. Within the Kill Chain model the infection time is often measured in seconds, and if the infection is not detected and blocked, the malware agent, such as a bot, will often then set up a secret channel to communicate with its controller. In the case of ransomware the communicated payload may include the encryption key used for the infected host to register its infection. As a malware infection can spread across a network in seconds, it is often important to detect its activities on the air, in memory and at-rest. Malware increasingly uses encrypted channels for communicating with their controllers. This paper presents a new approach to discovering the cryptographic artefacts of real malware clients that use cryptographic libraries of the Microsoft Windows operating system. This enables malware secret communications to be discovered without any prior malware knowledge.

In the area of full duplex (FD)-enabled small cell networks, limited works have been done on consideration of cache and mobile edge communication (MEC). In this paper, a virtual FD-enabled small cell network with cache and MEC is investigated for two heterogeneous services, high-data-rate service and computation-sensitive service. In our proposed scheme, content caching and FD communication are closely combined to offer high-data-rate services without the cost of backhaul resource. Computing offloading is conducted to guarantee the delay requirement of users. Then we formulate a virtual resource allocation problem, in which user association, power control, caching and computing offloading policies and resource allocation are jointly considered. Since the original problem is a mixed combinatorial problem, necessary variables relaxation and reformulation are conducted to transfer the original problem to a convex problem. Furthermore, alternating direction method of multipliers (ADMM) algorithm is adopted to obtain the optimal solution. Finally, extensive simulations are conducted with different system configurations to verify the effectiveness of the proposed scheme.

Nodulation is an essential process for biological nitrogen (N2) fixation in legumes, but its regulation remains poorly understood. Here, a β-expansin gene, GmEXPB2, was found to be critical for soybean (Glycine max) nodulation. GmEXPB2 was preferentially expressed at the early stage of nodule development. β-Glucuronidase staining further showed that GmEXPB2 was mainly localized to the nodule vascular trace and nodule vascular bundles, as well as nodule cortical and parenchyma cells, suggesting that GmEXPB2 might be involved in cell wall modification and extension during nodule formation and development. Overexpression of GmEXPB2 dramatically modified soybean root architecture, increasing the size and number of cortical cells in the root meristematic and elongation zones and expanding root hair density and size of the root hair zone. Confocal microscopy with green fluorescent protein-labeled rhizobium USDA110 cells showed that the infection events were significantly enhanced in the GmEXPB2-overexpressing lines. Moreover, nodule primordium development was earlier in overexpressing lines compared with wild-type plants. Thereby, overexpression of GmEXPB2 in either transgenic soybean hairy roots or whole plants resulted in increased nodule number, nodule mass, and nitrogenase activity and thus elevated plant N and phosphorus content as well as biomass. In contrast, suppression of GmEXPB2 in soybean transgenic composite plants led to smaller infected cells and thus reduced number of big nodules, nodule mass, and nitrogenase activity, thereby inhibiting soybean growth. Taken together, we conclude that GmEXPB2 critically affects soybean nodulation through modifying root architecture and promoting nodule formation and development and subsequently impacts biological N2 fixation and growth of soybean.

The quality of feature has significant impact on the performance of detection techniques used for Denial-of-Service (DoS) attack. The features that fail to provide accurate characterization for network traffic records make the techniques suffer from low accuracy in detection. Although researches have been conducted and attempted to overcome this problem, there are some constraints in these works. In this paper, we propose a technique based on Euclidean Distance Map (EDM) for optimal feature extraction. The proposed technique runs analysis on original feature space (first-order statistics) and extracts the multivariate correlations between the first-order statistics. The extracted multivariate correlations, namely second-order statistics, preserve significant discriminative information for accurate characterizations of network traffic records, and these multivariate correlations can be the high-quality potential features for DoS attack detection. The effectiveness of the proposed technique is evaluated using KDD CUP 99 dataset and experimental analysis shows encouraging results. © 2011 Springer-Verlag.

The reliability and availability of network services are being threatened by the growing number of Denial-of-Service (DoS) attacks. Effective mechanisms for DoS attack detection are demanded. Therefore, we propose a multivariate correlation analysis approach to investigate and extract second-order statistics from the observed network traffic records. These second-order statistics extracted by the proposed analysis approach can provide important correlative information hiding among the features. By making use of this hidden information, the detection accuracy can be significantly enhanced. The effectiveness of the proposed multivariate correlation analysis approach is evaluated on the KDD CUP 99 dataset. The evaluation shows encouraging results with average 99.96% detection rate and 2.08% false positive rate. Comparisons also show that our multivariate correlation analysis based detection approach outperforms some other current researches in detecting DoS attacks. © 2011 Springer-Verlag.

Endophytic bacteria isolated from Guangdong bamboo (Bambusa blumeana) plants were obtained under aerobic and anaerobic cultural conditions. Using the acetylene reduction assay (ARA), 40 strains were identified as diazotrophs. Diazotrophic isolates were grouped into five clusters (I, II, III, IV and V) by SDS-PAGE whole-cell protein patterns. The analysis of full-length 16S rDNA sequences of the representative strains of each group indicated that the endophytic diazotrophs showed great diversity and belonged to the generaAzospirillum(α-subclass of proteobacteria),EscherichiaandPseudomonas(γ-subclass of proteobacteria), andAquaspirillum(β-subclass of proteobacteria).

Since the early 1990s, under the programmes of ‘corporatisation’ and ‘marketisation’, stock exchange listing has become a central platform for a more recent round of state-owned enterprises (SOEs) reforms in China.The Chinese government has permitted profitable business units of an SOE or the enterprise as a whole to list shares on domestic and overseas stock exchanges. As a result of initial public offerings (IPOs), certain Chinese SOEs have been transformed from quasi-government agencies into profit-oriented corporations, and partially privatised .This research aims to investigate two questions:To what extent and in what way do management accounting and control practices change in listed Chinese SOEs following IPOs?How does management accounting change link to other organisational change, as triggered by IPOs in listed Chinese SOEs?

This thesis examines the phenomenon of sell-side financial analysts (analysts hereafter) "doing" corporate governance. The term "doing" is used in the current study to designate the various ways in which some analysts in the US and the UK, across the past decade or so, have made corporate governance visible. The thesis examines how this has occurred, and the mechanisms and devices that have made it possible. Analysts, it is suggested, can be viewed as "agents of transparency", in so far as they have taken the evaluation of companies beyond the financials, to include corporate governance issues. The thesis focuses primarily on the corporate governance reports produced by analysts, the official documents issued by various organisations and institutions, selected financial and business newspapers and magazines, together with other documents such as textbooks of corporate governance, as well as academic and practioner publications on corporate governance. Through an examination of these materials, the thesis investigates the pre-conditions that made possible the appearance and development of the corporate governance work pursued by analysts in the early twenty-first century. It examines the evaluations performed by analysts of the corporate governance procedures adopted by companies. In particular, it focuses on the ways in which analysts benchmarked the corporate governance procedures of companies against formal regulations, and how comparisons of the governance procedures adopted by different companies were undertaken and facilitated by analysts. Benchmarking, and the making of comparisons of corporate governance practices through a range of devices, are examined. The thesis also examines the linking of corporate governance to the financials (such as profitability, stock price performance, and equity valuation) in the investment analyses performed by analysts. It concentrates on the way in which analysts integrated corporate governance issues in the investment decision making process. Attention is paid to the ideas that shaped and articulated the integration, as well as to the tools and devices deployed by analysts. This thesis argues that greater attention is needed to the "doing" of corporate governance by analysts, and its implications for these "agents of transparency" that have broadened the parameters through which transparency is assessed.

Proceedings of the 8th Australian Information Security Mangement Conference, Edith Cowan University, Perth Western Australia, 30th November 2010, Web servers and web-based applications are commonly used as attack targets. The main issues are how to prevent unauthorised access and to protect web servers from the attack. Intrusion Detection Systems (IDSs) are widely used security tools to detect cyber-attacks and malicious activities in computer systems and networks. In this paper, we focus on the detection of various web-based attacks using Geometrical Structure Anomaly Detection (GSAD) model and we also propose a novel algorithm for the selection of most discriminating features to improve the computational complexity of payload-based GSAD model. Linear Discriminant method (LDA) is used for the feature reduction and classification of the incoming network traffic. GSAD model is based on a pattern recognition technique used in image processing. It analyses the correlations between various payload features and uses Mahalanobis Distance Map (MDM) to calculate the difference between normal and abnormal network traffic. We focus on the detection of generic attacks, shell code attacks, polymorphic attacks and polymorphic blending attacks. We evaluate accuracy of GSAD model experimentally on the real-world attacks dataset created at Georgia Institute of Technology. We conducted preliminary experiments on the DARPA 99 dataset to evaluate the accuracy of feature reduction.