Automated Synthetic Traffic Dataset Generation via Diffusion-Based Inpainting Pipeline

Synthetic Traffic generation plays a vital role in research related to computer networking. Most of the efforts taken so far by several researchers in the network community are relied on the synthetic traffic generation. Therefore to generate the synthetic traffic; traffic generators are used. Further this synthetic traffic behaves like real traffic and can be used to capture the behavior the network or device under test. Hence based on the design and features most of the researchers classify traffic generators into several categories. This contribution goes towards the classification of traffic generators so that it should help practitioners to use appropriate tools in their experimental works.

Network traffic data are critical for network research. With the help of synthetic traffic, researchers can readily generate data for network simulation and performance evaluation. However, the state-of-the-art traffic generators are either too simple to generate realistic traffic or require the implementation of original applications and user operations. We propose Synthetic PAcket Traffic Generative Adversarial Networks (SPATGAN) that are capable of generating synthetic traffic. The framework includes a server agent and a client agent, which transmit synthetic packets to each other and take the opponent's synthetic packets as conditional labels for the built-in Timing Synthesis Generative Adversarial Networks (TSynGAN) and a Packet Synthesis Generative Adversarial Networks (PSynGAN) to generate synthetic traffic. The evaluations demonstrate that the proposed framework can generate traffic whose distribution resembles real traffic distribution.

This study examines the role of linguistic regional variations in synthetic dataset generation and their impact on emotion detection performance. Emotion detection is essential for natural language processing (NLP) applications such as social media analysis, customer service, and mental health monitoring. To explore this, synthetic datasets were generated using a state-of-the-art language model, incorporating English variations from the United States, United Kingdom, and India, alongside a general baseline dataset. Two levels of prompt specificity were employed to assess the influence of regional linguistic nuances. Statistical analyses—including frequency distribution, term frequency-inverse document frequency (TF-IDF), type–token ratio (TTR), hapax legomena, pointwise mutual information (PMI) scores, and key-phrase extraction—revealed significant linguistic diversity and regional distinctions in the generated datasets. To evaluate their effectiveness, classification experiments were conducted with two models using bidirectional encoder representations from transformers (BERT) and its de-noising sequence to sequence variation (BART), beginning with zero-shot classification on the contextualized affect representations for emotion recognition (CARER) dataset, followed by fine-tuning with both baseline and region-specific datasets. Results demonstrated that region-specific datasets, particularly those generated with detailed prompts, significantly improved classification accuracy compared to the baseline. These findings underscore the importance of incorporating global linguistic variations in synthetic dataset generation, offering insights into how regional adaptations can enhance emotion detection models for diverse NLP applications.

The availability of fine grained time series data is a pre-requisite for research in smart-grids. While data for transmission systems is relatively easily obtainable, issues related to data collection, security and privacy hinder the widespread public availability/accessibility of such datasets at the distribution system level. This has prevented the larger research community from effectively applying sophisticated machine learning algorithms to significantly improve the distribution-level accuracy of predictions and increase the efficiency of grid operations. Synthetic dataset generation has proven to be a promising solution for addressing data availability issues in various domains such as computer vision, natural language processing and medicine. However, its exploration in the smart grid context remains unsatisfactory. Previous works have tried to generate synthetic datasets by modeling the underlying system dynamics: an approach which is difficult, time consuming, error prone and often times infeasible in many problems. In this work, we propose a novel data-driven approach to synthetic dataset generation by utilizing deep generative adversarial networks (GAN) to learn the conditional probability distribution of essential features in the real dataset and generate samples based on the learned distribution. To evaluate our synthetically generated dataset, we measure the maximum mean discrepancy (MMD) between real and synthetic datasets as probability distributions, and show that their sampling distance converges. To further validate our synthetic dataset, we perform common smart grid tasks such as k-means clustering and short-term prediction on both datasets. Experimental results show the efficacy of our synthetic dataset approach: the real and synthetic datasets are indistinguishable by solely examining the output of these tasks.

The availability of large-scale datasets has helped unleash the true potential of deep convolutional neural networks (CNNs). However, for the single-image denoising problem, capturing a real dataset is an unacceptably expensive and cumbersome procedure. Consequently, image denoising algorithms are mostly developed and evaluated on synthetic data that is usually generated with a widespread assumption of additive white Gaussian noise (AWGN). While the CNNs achieve impressive results on these synthetic datasets, they do not perform well when applied on real camera images, as reported in recent benchmark datasets. This is mainly because the AWGN is not adequate for modeling the real camera noise which is signal-dependent and heavily transformed by the camera imaging pipeline. In this paper, we present a framework that models camera imaging pipeline in forward and reverse directions. It allows us to produce any number of realistic image pairs for denoising both in RAW and sRGB spaces. By training a new image denoising network on realistic synthetic data, we achieve the state-of-the-art performance on real camera benchmark datasets. The parameters in our models are ~5 times lesser than the previous best method for RAW denoising. Furthermore, we demonstrate that the proposed framework generalizes beyond image denoising problem e.g., for color matching in stereoscopic cinema. The source code and pre-trained models are available at https://github.com/swz30/CycleISP.

The Internet of Vehicles (IoV) is becoming increasingly vulnerable to zero-day (ZD) cyberattacks, which often bypass conventional intrusion detection systems. To mitigate this challenge, this study proposes Zero-Day Bidirectional Encoder Representations from Transformers approach (ZDBERTa), a zero-shot learning (ZSL)-based framework for ZD attack detection, evaluated on the CICIoV2024 dataset. Unlike conventional AI models, ZSL enables the classification of attack types not previously encountered during the training phase. Two dataset variants are formed: Variant 1, created through synthetic traffic generation using a mixture of pattern-based, crossover, and mutation techniques, and Variant 2, augmented with a Generative Adversarial Network (GAN). To replicate realistic zero-day conditions, denial-of-service (DoS) attacks were omitted during training and introduced only at testing. The proposed ZDBERTa incorporates a Byte-Pair Encoding (BPE) tokenizer, a multi-layer transformer encoder, and a classification head for prediction, enabling the model to capture semantic patterns and identify previously unseen threats. The experimental results demonstrate that ZDBERTa achieves 86.677% accuracy on Variant 1, highlighting the complexity of zero-day detection, while performance significantly improves to 99.315% on Variant 2, underscoring the effectiveness of GAN-based augmentation. To the best of our knowledge, this is the first research to explore ZD detection within CICIoV2024, contributing a novel direction toward resilient IoV cybersecurity.

As is known, the term virtual studio is usually referred to the integration of synthetic and real environments: if an action occurs in the real world, the same must be reproduced in the artificial scene. The integration may be realized using one of the available systems which are generally characterized by high and expensive technology. Our work proposes an approach with two important advantages: low cost and possibility of building collaborative work. The system consists in a location named SM (generator of Synthetic worlds Meshing between real and virtual worlds) and in a differently located site indicated as V. The system SM is a powerful graphics oriented machine, which is able both to make highly realistic real time rendering of a complex virtual world, and to mesh the virtual scene with the video signal received from the V system. We suppose the V system characterized by a uniform background and a subject captured by a web cam whose video frames are sent to the SM system. In order to obtain the right information about the position of each video camera in the real world coordinate system and the zoom parameters, we propose an easy approach based on detecting the shape variations of a flag, with known aspect and dimension, placed in a defined position in the uniform background. This means that in a particular frame the scene modifications are codified in a few parameters related to the flag variations, so the integration between real and virtual becomes easy. The mesh results are sent to V, while just the selected meshed image is available for a generic user connected to the net service. The system may be applied in different contexts, for example video conferences and multiplayer virtual sets.© (2000) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Multi-camera systems (MCSs) are pivotal in aviation surveillance and autonomous navigation due to their wide coverage and high-resolution sensing. However, challenges such as complex setup, time-consuming data acquisition, and costly testing hinder research progress. To address these, we introduce MCS-Sim, a photo-realistic MCSsimulator for UAV visual perception research. MCS-Sim integrates vision sensor configurations, vehicle dynamics, and dynamic scenes, enabling rapid virtual prototyping and multi-task dataset generation. It supports dense flow estimation, 3D reconstruction, visual simultaneous localization and mapping, object detection, and tracking. With a hardware-in-loop interface, MCS-Sim facilitates closed-loop simulation for system validation. Experiments demonstrate its effectiveness in synthetic dataset generation, visual perception algorithm testing, and closed-loop simulation. Here we show that MCS-Sim significantly advances multi-camera UAV visual perception research, offering a versatile platform for future innovations.

Synthetic dataset generation in Computer Vision, particularly for industrial applications, is still underexplored. Industrial defect segmentation, for instance, requires highly accurate labels, yet acquiring such data is costly and time-consuming. To address this challenge, we propose a novel diffusion-based pipeline for generating high-fidelity industrial datasets with minimal supervision. Our approach conditions the diffusion model on enriched bounding-box representations to produce precise segmentation masks, ensuring realistic and accurately localized defect synthesis. Compared to existing layout-conditioned generative methods, our approach improves defect consistency and spatial accuracy. We introduce two quantitative metrics to evaluate the effectiveness of our method and assess its impact on a downstream segmentation task trained on real and synthetic data. Our results demonstrate that diffusion-based synthesis can bridge the gap between artificial and real-world industrial data, fostering more reliable and cost-efficient segmentation models.

High-quality, pixel-level annotated datasets are crucial for training deep learning models, while their creation is often labor-intensive, time-consuming, and costly. Generative diffusion models have then gained prominence for producing synthetic datasets, yet existing text-to-data methods struggle with generating complex scenes involving multiple objects and intricate spatial arrangements. To address these limitations, we introduce FlexDataset, a framework that pioneers the composition-to-data (C2D) paradigm. FlexDataset generates high-fidelity synthetic datasets with versatile annotations, tailored for tasks like salient object detection, depth estimation, and segmentation. Leveraging a meticulously designed composition-to-image (C2I) framework, it offers precise positional and categorical control. Our Versatile Annotation Generation (VAG) Plan A further enhances efficiency by exploiting rich latent representations through tuned perception decoders, reducing annotation time by nearly fivefold. FlexDataset allows unlimited generation of customized, multi-instance and multi-category (MIMC) annotated data. Extensive experiments show that FlexDataset sets a new standard in synthetic dataset generation across multiple datasets and tasks, including zero-shot and long-tail scenarios.

Road safety is crucial in transportation infrastructure management, with potholes posing significant hazards such as accidents, vehicle damage, and increased maintenance costs. Traditional pothole detection methods rely on reactive and labor-intensive processes, including manual inspections or infrequent public reports. This project introduces a proactive approach to pothole identification using Convolutional Neural Networks (CNNs) with the YOLOv8 algorithm. By leveraging deep learning and computer vision, the system enables real-time, autonomous pothole detection and localization from vehicle-mounted cameras or drones. The YOLOv8 algorithm's accuracy, efficiency, and real-time performance make it well-suited for this task. Training the model on diverse datasets enhances its robustness across various road conditions, lighting variations, and camera perspectives. Evaluation across multiple road segments yielded promising results.

In this paper we present some simulation results about web traffic differentiation. Traditionally, the differentiated services architecture (DiffServ) has been under consideration to providing different services and QoS in Internet. The flow size of the web traffic is important when we intend to achieve certain QoS to these kind of flows. We propose a DiffServ model where short flows are managed with high priority and the maximum QoS. We intend to promote some long flows, but this promotion must not affects the short flows performance. For this purpose, firstly we need to detect the gaps in the bandwidth committed to QoS. Second, some long flows can be promoted over high priority queue, instead of low priority queue. Our proposal has special attention for the very long flows which reduce the web traffic performance. We must detect and penalize these extreme long flows in order to avoid the promotion of these flows and permit others aspirants to be promoted. With these actions we get to improve the global performance of the web traffic. We have used ns2 network simulator tool for the simulation with the PackMime-HTTP object for the realistic synthetic Web traffic generation. We have analyzed some simulation results about web transmission latency and packet loss.

Recently, there has been increasing interest in developing and deploying deep graph learning algorithms for various tasks, such as fraud detection and recommender systems. However, there is a limited number of publicly available graph-structured datasets, most of which are small compared with production-sized applications or limited in their application domain. In this work, we tackle this shortcoming by proposing a synthetic graph generation tool that enables scaling datasets to production-size graphs with trillions of edges and billions of nodes. The proposed method comprises a series of parametric models that can either be randomly initialized or fit to proprietary datasets. These models can then be released to researchers to study graph methods on the synthetic data, facilitating prototype development and novel applications. We demonstrate the generalizability of the framework across various datasets, mimicking their structural and feature distributions, as well as the ability to scale them to varying sizes, demonstrating their usefulness for benchmarking and model development. Code can be found on GitHub.

Having synthetic image generation and automatic labelling as two separate processes remains one of the main limitations of automatic generation of large real-world synthetic datasets. To overcome this drawback, a methodology to perform both tasks in a simultaneous and automatic manner is proposed. To resemble real-world scenarios, a diverse set of rendering configurations of illumination, locations, and sizes are presented. For testing, three synthetic datasets (S, M and SM), oriented to the construction field, were generated. Faster R-CNN, RetinaNet, and YoloV4 detection algorithms were used to independently evaluate the datasets using the COCO evaluation metrics and the PascalVOC Mean Average Accuracy metric. Results show that, in general, the S dataset performed 1.2% better in the evaluation metrics and that the SM dataset obtained better plots of training and validation loss curves in each detector; highlighting the combinational usage of images with single and multiple objects as a better generalisation approach.

The discomfort, invasiveness, and inconvenience associated with traditional glucose monitoring methods continue to hinder patient compliance and the effectiveness of diabetes management. This simulation-driven feasibility study presents the design of the Photoacoustic Glucose Monitoring Device (PAGMD), a non-invasive glucose sensing system that integrates photoacoustic spectroscopy, machine learning, and IoT connectivity to enable real-time glucose estimation. PAGMD employs pulsed near-infrared light to induce glucose-dependent acoustic signals, which are detected by ultrasonic transducers and analyzed by machine learning algorithms for concentration prediction. The device features a hybrid energy harvesting system, combining thermoelectric and piezoelectric elements, to support continuous, battery-assisted operation and long-term usability. The system was theoretically validated using a multi-stage simulation pipeline that included Monte Carlo photon modeling, finite element acoustic simulation, and synthetic glucose dataset generation across a range of physiological conditions. The machine learning models achieved a coefficient of determination of 0.98 and a mean absolute relative difference of 6.97%, with 98.2% of predictions falling within clinically acceptable error zones. While current findings are based on simulated environments, a future validation roadmap encompassing in vitro, ex vivo, and in vivo studies is proposed to support clinical translation. This work should be regarded as a simulation-driven feasibility study; the reported benchmarks represent design targets for future prototyping rather than experimentally validated hardware results. By prioritizing non-invasiveness, ease of use, and adaptive intelligence, PAGMD represents a promising step toward inclusive, personalized metabolic health monitoring, especially for individuals with disabilities or limited dexterity.