Biblio

The success of deep learning based steganography has shifted focus of researchers from traditional steganography approaches to deep learning based steganography. Various deep steganographic models have been developed for improved security, capacity and invisibility. In this work a multi-data deep learning steganography model has been developed using a well known deep learning model called Generative Adversarial Networks (GAN) more specifically using deep convolutional Generative Adversarial Networks (DCGAN). The model is capable of hiding two different messages, meant for two different receivers, inside a single cover image. The proposed model consists of four networks namely Generator, Steganalyzer Extractor1 and Extractor2 network. The Generator hides two secret messages inside one cover image which are extracted using two different extractors. The Steganalyzer network differentiates between the cover and stego images generated by the generator network. The experiment has been carried out on CelebA dataset. Two commonly used distortion metrics Peak signal-to-Noise ratio (PSNR) and Structural Similarity Index Metric (SSIM) are used for measuring the distortion in the stego image The results of experimentation show that the stego images generated have good imperceptibility and high extraction rates.

Existing anomaly detection models show success in detecting abnormal images with generative adversarial networks on the insufficient annotation of anomalous samples. However, existing models cannot accurately identify the anomaly samples which are close to the normal samples. We assume that the main reason is that these methods ignore the diversity of patterns in normal samples. To alleviate the above issue, this paper proposes a novel anomaly detection framework based on generative adversarial network, called ADe-GAN. More concretely, we construct a self-supervised learning task to fully explore the pattern information and latent representations of input images. In model inferring stage, we design a new abnormality score approach by jointly considering the pattern information and reconstruction errors to improve the performance of anomaly detection. Extensive experiments show that the ADe-GAN outperforms the state-of-the-art methods over several real-world datasets.

Generative adversarial networks (GANs) have been increasingly popular among deep learning methods. With many GANs-based models developed since its emergence, among which are conditional generative adversarial networks, progressive growing of generative adversarial networks, Wasserstein generative adversarial networks and so on. These frameworks are currently widely applied in areas such as remote sensing cybersecurity, medical, and architecture. Especially, they have solved problems of cloud removal, semantic segmentation, image-to-image translation and data argumentation in remote sensing. For example, WGANs and ProGANs can be applied in data argumentation, and cGANs can be applied in semantic argumentation and image-to-image translation. This article provides an overview of structures of multiple GANs-based models and what areas they can be applied in remote sensing.

Fuzzing is a popular technique for finding soft-ware vulnerabilities. Despite their success, the state-of-art fuzzers will inevitably produce a large number of low-quality inputs. In recent years, Machine Learning (ML) based selection strategies have reported promising results. However, the existing ML-based fuzzers are limited by the lack of training data. Because the mutation strategy of fuzzing can not effectively generate useful input, it is prohibitively expensive to collect enough inputs to train models. In this paper, propose a generative adversarial networks based solution to generate a large number of inputs to solve the problem of insufficient data. We implement the proposal in the American Fuzzy Lop (AFL), and the experimental results show that it can find more crashes at the same time compared with the original AFL.

Deep learning has been widely applied for jobs involving face inpainting, however, there are usually some problems, such as incoherent inpainting edges, lack of diversity of generated images and other problems. In order to get more feature information and improve the inpainting effect, we therefore propose a Generative Adversarial Network of Skip Connection (SCGAN), which connects the encoder layers and the decoder layers by skip connection in the generator. The coherence and consistency of the image inpainting edges are improved, and the finer features of the image inpainting are refined, simultaneously using the discriminator's local and global double discriminators model. We also employ WGAN-GP loss to enhance model stability during training, prevent model collapse, and increase the variety of inpainting face images. Finally, experiments on the CelebA dataset and the LFW dataset are performed, and the model's performance is assessed using the PSNR and SSIM indices. Our model's face image inpainting is more realistic and coherent than that of other models, and the model training is more reliable.

Speech enhancement is often applied for speech-based systems due to the proneness of speech signals to additive background noise. While speech processing-based methods are traditionally used for speech enhancement, with advancements in deep learning technologies, many efforts have been made to implement them for speech enhancement. Using deep learning, the networks learn mapping functions from noisy data to clean ones and then learn to reconstruct the clean speech signals. As a consequence, deep learning methods can reduce what is so-called musical noise that is often found in traditional speech enhancement methods. Currently, one popular deep learning architecture for speech enhancement is generative adversarial networks (GAN). However, the cross-entropy loss that is employed in GAN often causes the training to be unstable. So, in many implementations of GAN, the cross-entropy loss is replaced with the least-square loss. In this paper, to improve the training stability of GAN using cross-entropy loss, we propose to use deep regret analytic generative adversarial networks (Dragan) for speech enhancements. It is based on applying a gradient penalty on cross-entropy loss. We also employ relativistic rules to stabilize the training of GAN. Then, we applied it to the least square and Dragan losses. Our experiments suggest that the proposed method improve the quality of speech better than the least-square loss on several objective quality metrics.

In recent years, generative adversarial networks (GAN) have become a research hotspot in the field of deep learning. Researchers apply them to the field of anomaly detection and are committed to effectively and accurately identifying abnormal images in practical applications. In anomaly detection, traditional supervised learning algorithms have limitations in training with a large number of known labeled samples. Therefore, the anomaly detection model of unsupervised learning GAN is the research object for discussion and research. Firstly, the basic principles of GAN are introduced. Secondly, several typical GAN-based anomaly detection models are sorted out in detail. Then by comparing the similarities and differences of each derivative model, discuss and summarize their respective advantages, limitations and application scenarios. Finally, the problems and challenges faced by GAN in anomaly detection are discussed, and future research directions are prospected.

As one of the defensive solutions against cyberattacks, an Intrusion Detection System (IDS) plays an important role in observing the network state and alerting suspicious actions that can break down the system. There are many attempts of adopting Machine Learning (ML) in IDS to achieve high performance in intrusion detection. However, all of them necessitate a large amount of labeled data. In addition, labeling attack data is a time-consuming and expensive human-labor operation, it makes existing ML methods difficult to deploy in a new system or yields lower results due to a lack of labels on pre-trained data. To address these issues, we propose a semi-supervised IDS model that leverages Generative Adversarial Networks (GANs) and Adversarial AutoEncoder (AAE), called a semi-supervised adversarial autoencoder (SAAE). Our SAAE experimental results on two public datasets for benchmarking ML-based IDS, including NF-CSE-CIC-IDS2018 and NF-UNSW-NB15, demonstrate the effectiveness of AAE and GAN in case of using only a small number of labeled data. In particular, our approach outperforms other ML methods with the highest detection rates in spite of the scarcity of labeled data for model training, even with only 1% labeled data.

Imbalanced class distribution can cause information loss and missed/false alarms for deep learning and machine-learning algorithms. The detection performance of traditional intrusion detection systems tend to degenerate due to skewed class distribution caused by the uneven allocation of observations in different kinds of attacks. To combat class imbalance and improve network intrusion detection performance, we adopt the conditional generative adversarial network (CTGAN) that enables the generation of samples of specific classes of interest. CTGAN builds on the generative adversarial networks (GAN) architecture to model tabular data and generate high quality synthetic data by conditionally sampling rows from the generated model. Oversampling using CTGAN adds instances to the minority class such that both data in the majority and the minority class are of equal distribution. The generated security alerts are used for training classifiers that realize critical alert detection. The proposed scheme is evaluated on a real-world dataset collected from security operation center of a large enterprise. The experiment results show that detection accuracy can be substantially improved when CTGAN is adopted to produce a balanced security-alert dataset. We believe the proposed CTGAN-based approach can cast new light on building effective systems for critical alert detection with reduced missed/false alarms.

With the progress of cryptography computer science, designing cryptographic algorithms using deep learning is a very innovative research direction. Google Brain designed a communication model using generation adversarial network and explored the encrypted communication algorithm based on machine learning. However, the encrypted communication model it designed lacks quantitative evaluation. When some plaintexts and keys are leaked at the same time, the security of communication cannot be guaranteed. This model is optimized to enhance the security by adjusting the optimizer, modifying the activation function, and increasing batch normalization to improve communication speed of optimization. Experiments were performed on 16 bits and 64 bits plaintexts communication. With plaintext and key leak rate of 0.75, the decryption error rate of the decryptor is 0.01 and the attacker can't guess any valid information about the communication.

The NIST standardization process for post-quantum cryptography has been drawing the attention of researchers to the submitted candidates. One direction of research consists in implementing those candidates on embedded systems and that exposes them to physical attacks in return. The Classic McEliece cryptosystem, which is among the four finalists of round 3 in the Key Encapsulation Mechanism category, builds its security on the hardness of the syndrome decoding problem, which is a classic hard problem in code-based cryptography. This cryptosystem was recently targeted by a laser fault injection attack leading to message recovery. Regrettably, the attack setting is very restrictive and it does not tolerate any error in the faulty syndrome. Moreover, it depends on the very strong attacker model of laser fault injection, and does not apply to optimised implementations of the algorithm that make optimal usage of the machine words capacity. In this article, we propose a to change the angle and perform a message-recovery attack that relies on side-channel information only. We improve on the previously published work in several key aspects. First, we show that side-channel information, obtained with power consumption analysis, is sufficient to obtain an integer syndrome, as required by the attack framework. This is done by leveraging classic machine learning techniques that recover the Hamming weight information very accurately. Second, we put forward a computationally-efficient method, based on a simple dot product and information-set decoding algorithms, to recover the message from the, possibly inaccurate, recovered integer syndrome. Finally, we present a masking countermeasure against the proposed attack.

We show that the most common flavors of noisy leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to a small statistical simulation error and a slight loss in the leakage parameter. The latter holds true in particular for one of the most used noisy-leakage models, where the noisiness is measured using the conditional average min-entropy (Naor and Segev, CRYPTO’09 and SICOMP’12). Our reductions between noisy and bounded leakage are achieved in two steps. First, we put forward a new leakage model (dubbed the dense leakage model) and prove that dense leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to small statistical distance. Second, we show that the most common noisy-leakage models fall within the class of dense leakage, with good parameters. Third, we prove lower bounds on the amount of bounded leakage required for simulation with sub-constant error, showing that our reductions are nearly optimal. In particular, our results imply that useful general simulation of noisy leakage based on statistical distance and mutual information is impossible. We also provide a complete picture of the relationships between different noisy-leakage models. Our result finds applications to leakage-resilient cryptography, where we are often able to lift security in the presence of bounded leakage to security in the presence of noisy leakage, both in the information-theoretic and in the computational setting. Remarkably, this lifting procedure makes only black-box use of the underlying schemes. Additionally, we show how to use lower bounds in communication complexity to prove that bounded-collusion protocols (Kumar, Meka, and Sahai, FOCS’19) for certain functions do not only require long transcripts, but also necessarily need to reveal enough information about the inputs.

Unmanned Aerial Vehicle (UAV)-based streaming media transmission may become unstable when the bit rate generated by the source load exceeds the channel capacity owing to the UAV location and speed change. The change of the location can affect the network connection, leading to reduced transmission rate; the change of the flying speed can increase the video payload due to more I-frames. To improve the transmission reliability, in this paper we design a Client-Server-Ground&User (C-S-G&U) framework, and propose an algorithm of splitting-merging stream (SMS) for multi-link concurrent transmission. We also establish multiple transport links and configure the routing rules for the cross-layer design. The multi-link transmission can achieve higher throughput and significantly smaller end-to-end delay than a single-link especially in a heavy load situation. The audio and video data are packaged into the payload by the Real-time Transport Protocol (RTP) before being transmitted over the User Datagram Protocol (UDP). The forward error correction (FEC) algorithm is implemented to promote the reliability of the UDP transmission, and an encryption algorithm to enhance security. In addition, we propose a Quality of Service (QoS) strategy so that the server and the user can control the UAV to adapt its transmission mode dynamically, according to the load, delay, and packet loss. Our design has been implemented on an engineering platform, whose efficacy has been verified through comprehensive experiments.

The correctness of user traffic forwarding paths is an important goal of trusted transmission. Many network security issues are related to it, i.e., denial-of-service attacks, route hijacking, etc. The current path-aware network architecture can effectively overcome this issue through path verification. At present, the main problems of path verification are high communication and high computation overhead. To this aim, this article proposes a lightweight real-time verification mechanism of intradomain forwarding paths in autonomous systems to achieve a path verification architecture with no communication overhead and low computing overhead. The problem situation is that a packet finally reaches the destination, but its forwarding path is inconsistent with the expected path. The expected path refers to the packet forwarding path determined by the interior gateway protocols. If the actual forwarding path is different from the expected one, it is regarded as an incorrect forwarding path. This article focuses on the most typical intradomain routing environment. A few routers are set as the verification routers to block the traffic with incorrect forwarding paths and raise alerts. Experiments prove that this article effectively solves the problem of path verification and the problem of high communication and computing overhead.

This paper proposes a new two-stage encryption method to increase the cryptographic strength of the AES algorithm, which is based on stochastic error of a neural network. The composite encryption key in AES neural network cryptosystem are the weight matrices of synaptic connections between neurons and the metadata about the architecture of the neural network. The stochastic nature of the prediction error of the neural network provides an ever-changing pair key-ciphertext. Different topologies of the neural networks and the use of various activation functions increase the number of variations of the AES neural network decryption algorithm. The ciphertext is created by the forward propagation process. The encryption result is reversed back to plaintext by the reverse neural network functional operator.

We demonstrated hardware emulations to evaluate the error-correction performance for a FEC scheme with multilevel coding. It has enabled the measurement of BER to reach the order of 10−14 for the decoded signal.

With the rapid development of Ethernet, RS (544, 514) (KP4-forward error correction), which was widely used in high-speed Ethernet standards for its good performance-complexity trade-off, may not meet the demands of next-generation Ethernet for higher data transmission speed and better decoding performance. A concatenated code based on KP4-FEC has become a good solution because of its low complexity and excellent compatibility. For concatenated codes, aside from the selection of outer and inner codes, an efficient interleaving scheme is also very critical to deal with different channel conditions. Aiming at burst errors in wired communication, we propose a novel matrix interleaving scheme for concatenated codes which set the outer code as KP4-FEC and the inner code as Bose-Chaudhuri-Hocquenghem (BCH) code. In the proposed scheme, burst errors are evenly distributed to each BCH code as much as possible to improve their overall decoding efficiency. Meanwhile, the bit continuity in each symbol of the RS codeword is guaranteed during transmission, so the number of symbols affected by burst errors is minimized. Simulation results demonstrate that the proposed interleaving scheme can achieve a better decoding performance on burst-error channels than the original scheme. In some cases, the extra coding gain at the bit-error-rate (BER) of 1 × 10−15 can even reach 1 dB.

At present, our English error correction algorithm is slightly general, the error correction ability is also very limited, and its accuracy rate is also low, so it is very necessary to improve. This article will further explore the problem of syntax error correction, and the corresponding algorithm model will also be proposed. Based on deep learning technology to improve the error correction rate of English grammar, put forward the corresponding solution, put forward the Sep2seq-based English grammar error correction system model, and carry out a series of rectifications to improve its efficiency and accuracy. The basic architecture of TensorFLOW is used to implement the model, and the success of the error correction algorithm model is proved, which brings great improvement and progress to the success of error correction.

In real-time video streaming, data packets are transported over the network from a transmitter to a receiver. The quality of the received video fluctuates as the network conditions change, and it can degrade substantially when there is considerable packet loss. Forward error correction (FEC) techniques can be used to recover lost packets by incorporating redundant data. Conventional FEC schemes do not work well when scalable video coding (SVC) is adopted. In this paper, we propose a novel FEC scheme that overcomes the drawbacks of these schemes by considering the reference picture structure of SVC and weighting the reference pictures more when FEC redundancy is applied. The experimental results show that the proposed FEC scheme outperforms conventional FEC schemes.

Currently, the pulse position modulation (PPM) schemes are suffering from bandwidth application where the line rate is double that of the initial data rate. Thus, the offset pulse position modulation (OPPM) has been suggested to rectify this concern. Several attempts to improve the OPPM can be found in the open literature. This study focuses on the utilization of Reed Solomon (RS) codes to enhance the forward error correction (FEC) bit error rate, which is not yet explored. The performance of errors of the uncoded OPPM was compared to the one used by RS coded OPPM using the number of photons per pulse, the transmission's efficacy, and bandwidth growth. The results demonstrate that employing FEC coding would increase the system's error performance especially when the RS is operating at its finest settings. Specifically, when operating with a capacity that is equivalent to or even more 0.7, the OPPM with RS code outperforms the uncoded OPPM where the OPPM with MLSD needs only 1.2×103 photons per pulse with an ideal coding rate of about 3/4.

We propose the use of dual coding concatenation for mitigation of post-shaping burst errors in probabilistic amplitude shaping (PAS) architectures. The proposed dual coding concatenation for PAS is a hybrid integration of conventional reverse concatenation and forward concatenation, i.e., post-shaping forward error correction (FEC) layer and pre-shaping FEC layer, respectively. A low-complexity architecture based on parallel Bose–Chaudhuri–Hocquenghem (BCH) codes is introduced for the pre-shaping FEC layer. Proposed dual coding concatenation can relax bit error rate (BER) requirement after post-shaping soft-decision (SD) FEC codes by an order of magnitude, resulting in a gain of up to 0.25 dB depending on the complexity of post-shaping FEC. Also, combined shaping and coding performance was analyzed based on sphere shaping and the impact of shaping length on coding performance was demonstrated.

In this work, we implement a complete probabilistic amplitude shaping (PAS) architecture on a field-programmable gate array (FPGA) platform to study the interplay between probabilistic shaping (PS) and forward error correction (FEC). Due to the fully parallelized input–output interfaces based on look up table (LUT) and low computational complexity without high-precision multiplication, hierarchical distribution matching (HiDM) is chosen as the solution for real time probabilistic shaping. In terms of FEC, we select two kinds of the mainstream soft decision-forward error correction (SD-FEC) algorithms currently used in optical communication system, namely Open FEC (OFEC) and soft-decision quasi-cyclic low-density parity-check (SD-QC-LDPC) codes. Through FPGA experimental investigation, we studied the impact of probabilistic shaping on OFEC and LDPC, respectively, based on PS-16QAM under moderate shaping, and also the impact of probabilistic shaping on LDPC code based on PS-64QAM under weak/strong shaping. The FPGA experimental results show that if pre-FEC bit error rate (BER) is used as the predictor, moderate shaping induces no degradation on the OFEC performance, while strong shaping slightly degrades the error correction performance of LDPC. Nevertheless, there is no error floor when the output BER is around 10-15. However, if normalized generalized mutual information (NGMI) is selected as the predictor, the performance degradation of LDPC will become insignificant, which means pre-FEC BER may not a good predictor for LDPC in probabilistic shaping scenario. We also studied the impact of residual errors after FEC decoding on HiDM. The FPGA experimental results show that the increased BER after HiDM decoding is within 10 times compared to post-FEC BER.

With a record 400Gbps 100-piece-FPGA implementation, we investigate performance of the potential FEC schemes for OIF-800GZR. By comparing the power dissipation and correction threshold at 10−15 BER, we proposed the simplified OFEC for the 800G-ZR FEC.

JPEG-XS offers low complexity image compression for applications with constrained but reasonable bit-rate, and low latency. Our paper explores the deployment of JPEG-XS on lossy packet networks. To preserve low latency, Forward Error Correction (FEC) is envisioned as the protection mechanism of interest. Although the JPEG-XS codestream is not scalable in essence, we observe that the loss of a codestream fraction impacts the decoded image quality differently, depending on whether this codestream fraction corresponds to codestream headers, to coefficient significance information, or to low/high frequency data. Hence, we propose a rate-distortion optimal unequal error protection scheme that adapts the redundancy level of Reed-Solomon codes according to the rate of channel losses and the type of information protected by the code. Our experiments demonstrate that, at 5% loss rates, it reduces the Mean Squared Error by up to 92% and 65%, compared to a transmission without and with optimal but equal protection, respectively.

The discussion about forward error correction (FEC) used for 800G unamplified link (800LR) is ongoing. Aiming at two potential options for FEC bit error ratio (BER) threshold, we propose two FEC schemes, respectively based on channel-polarized (CP) multilevel coding (MLC) and bit interleaved coded modulation (BICM), with the same inner FEC code. The field-programmable gate array (FPGA) verification results indicate that with the same FEC overhead (OH), proposed CP-MLC outperforms BICM scheme with less resource and power consumption.