Abstract: Network traffic research relies on large-scale, high-quality traffic data. However, obtaining such data remains difficult because of privacy constraints, collection costs, class imbalance, and continuous updates. These challenges have increased researchers’ interest in traffic generation. Although many generation methods have been proposed, existing studies and surveys often overlook two key questions: what form of traffic is generated and what practical objectives it can support. Based on 113 candidate records published from 2019 to 2026, this survey provides a detailed analysis of 39 representative network traffic generation studies through the lenses of representation levels and objective consistency. We organize existing methods into four representation levels and analyze how generated data relate to usage scenarios. We find that many methods preserve information that is useful for downstream tasks such as classification and intrusion detection, but task usefulness does not guarantee replayability or usability in real network environments. High-level representations are easier to model, yet they often discard protocol semantics, packet dependencies, and communication logic. We therefore distinguish task consistency from protocol consistency and show that the latter remains underexplored. We further summarize evaluation practices, discuss level-specific metrics, and highlight future directions including controllable generation, protocol-aware state-consistent synthesis, and engineering-oriented evaluation.

网络流量生成的表示层级与目标一致性综述

[1]Adeleke OA, Bastin N, Gurkan D, 2023. Network traffic generation: a survey and methodology. ACM Comput Surv, 55(2):28.

[2]Alsaedi A, Moustafa N, Tari Z, et al., 2020. TON_IoT telemetry dataset: a new generation dataset of IoT and IIoT for data-driven intrusion detection systems. IEEE Access, 8:165130-165150.

[3]Cai SH, Zhao XY, Chen JF, et al., 2025. CT-SSSA: malicious traffic augmentation based on classifier TransGAN and spatial-channel synergistic self-attention. Knowl-Based Syst, 329:114285.

[4]Carillo R, Cerasuolo F, Bovenzi G, et al., 2025. Explainable federated class incremental learning for encrypted network traffic classification. Comput Netw, 269:111448.

[5]Chai HY, Jiang T, Yu L, 2024. Diffusion model-based mobile traffic generation with open data for network planning and optimization. Proc 30^th Conf on Knowledge Discovery and Data Mining, p.4828-4838.

[6]Chai HY, Qi XQ, Li Y, 2025a. Spatio-temporal knowledge driven diffusion model for mobile traffic generation. IEEE Trans Mob Comput, 24(6):4939-4956.

[7]Chai HY, Zhang SY, Qi XQ, et al., 2025b. UoMo: a universal model of mobile traffic forecasting for wireless network optimization. Proc 31^st Conf on Knowledge Discovery and Data Mining V.2, p.4308-4319.

[8]Charlier J, Singh A, Ormazabal G, et al., 2019. SynGAN: towards generating synthetic network attacks using GANs.

[9]Chawla NV, Bowyer KW, Hall LO, et al., 2002. SMOTE: synthetic minority over-sampling technique. J Artif Intell Res, 16(1):321-357.

[10]Chen YC, 2017. A tutorial on kernel density estimation and recent advances. Biostat Epidemiol, 1(1):161-187.

[11]Cheng A, 2019. PAC-GAN: packet generation of network traffic using generative adversarial networks. IEEE 10^th Annual Information Technology, Electronics and Mobile Communication Conf, p.728-734.

[12]Chu A, Jiang X, Liu SN, et al., 2024. Feasibility of state space models for network traffic generation. Proc SIGCOMM Workshop on Networks for AI Computing, p.9-17.

[13]Chu A, Jiang X, Liu SN, et al., 2026. NetSSM: multi-flow and state-aware network trace generation using state-space models. Proc ACM Netw, 4(CoNEXT1):6.

[14]Delgado-Soto JA, de Vergara JEL, González I, et al., 2025. GPT on the wire: towards realistic network traffic conversations generated with large language models. Comput Netw, 265:111308.

[15]Dowoo B, Jung Y, Choi C, 2019. PcapGAN: packet capture file generator by style-based generative adversarial networks. 18^th IEEE Int Conf on Machine Learning and Applications, p.1149-1154.

[16]Draper-Gil G, Lashkari AH, Mamun MSI, et al., 2016. Characterization of encrypted and VPN traffic using time-related features. Proc 2^nd Int Conf on Information Systems Security and Privacy, p.407-414.

[17]Du LF, He JJ, Li T, et al., 2023. DBWE-Corbat: background network traffic generation using dynamic word embedding and contrastive learning for cyber range. Comput Secur, 129:103202.

[18]Fernandes DAB, Neto M, Soares LFB, et al., 2015. On the self-similarity of traffic generated by network traffic simulators. In: Obaidat MS, Nicopolitidis P, Zarai F (Eds.), Modeling and Simulation of Computer Networks and Systems. Morgan Kaufmann, Waltham, p.285-311.

[19]Goodfellow IJ, Pouget-Abadie J, Mirza M, et al., 2014. Generative adversarial networks. Proc 28^th Int Conf on Neural Information Processing Systems, p.2672-2680. https://dl.acm.org/doi/10.5555/2969033.2969125

[20]Ho J, Jain A, Abbeel P, 2020. Denoising diffusion probabilistic models. Proc 34^th Int Conf on Neural Information Processing Systems, Article 574. https://dl.acm.org/doi/abs/10.5555/3495724.3496298

[21]Hochreiter S, Schmidhuber J, 1997. Long short-term memory. Neur Comput, 9(8):1735-1780.

[22]Huang YZ, Li XH, Geng JX, et al., 2026. TrafficT5: multi-stage self-correcting framework for traffic generation. Comput Netw, 274:111858.

[23]Hui SD, Wang HD, Wang ZH, et al., 2022. Knowledge enhanced GAN for IoT traffic generation. Proc ACM Web Conf, p.3336-3346.

[24]Jablaoui R, Liouane N, 2026. GA-CNN-BiGRU-IDS: a robust framework for intrusion detection system based on GA for data augmentation and hybrid CNN-BiGRU model for spatiotemporal feature extraction. Comput Electr Eng, 130:110900.

[25]Jiang X, Liu SN, Gember-Jacobson A, et al., 2024. NetDiffusion: network data augmentation through protocol-constrained traffic generation. Proc ACM Meas Anal Comput Syst, 8(1):11.

[26]Jonath KK, Naz A, Zhang SG, 2026. GMM-cGAN: mitigating data scarcity and label noise for robust encrypted malicious traffic classification. Comput Netw, 274:111823.

[27]Kattadige C, Muramudalige SR, Choi KN, et al., 2021. VideoTrain: a generative adversarial framework for synthetic video traffic generation. IEEE 22^nd Int Symp on a World of Wireless, Mobile and Multimedia Networks, p.209-218.

[28]Kennedy J, Eberhart R, 1995. Particle swarm optimization. Proc Int Conf on Neural Networks, p.1942-1948.

[29]Kholgh DK, Kostakos P, 2023. PAC-GPT: a novel approach to generating synthetic network traffic with GPT-3. IEEE Access, 11:114936-114951.

[30]Kingma DP, Welling M, 2014. Auto-encoding variational Bayes.

[31]Koroniotis N, Moustafa N, Sitnikova E, et al., 2019. Towards the development of realistic botnet dataset in the Internet of Things for network forensic analytics: Bot-IoT dataset. Future Gener Comput Syst, 100:779-796.

[32]Li J, Li X, 2022. 5G network traffic prediction based on EEMD-GAN. Proc 7^th Int Conf on Cyber Security and Information Engineering, p.408-412.

[33]Li T, Hui SD, Zhang SY, et al., 2024. Mobile user traffic generation via multi-scale hierarchical GAN. ACM Trans Knowl Discov Data, 18(8):189.

[34]Lin ZN, Jain A, Wang C, et al., 2020. Using GANs for sharing networked time series data: challenges, initial promise, and open questions. Proc ACM Int Measurement Conf, p.464-483.

[35]Manocchio LD, Layeghy S, Portmann M, 2021. FlowGAN - synthetic network flow generation using generative adversarial networks. IEEE 24^th Int Conf on Computational Science and Engineering, p.168-176.

[36]Meddahi A, Drira H, Meddahi A, 2021. SIP-GAN: generative adversarial networks for SIP traffic generation. Int Symp on Networks, Computers and Communications, p.1-6.

[37]Meslet-Millet F, Mouysset S, Chaput E, 2022. NeCSTGen: an approach for realistic network traffic generation using deep learning. IEEE Global Communications Conf, p.3108-3113.

[38]Meta AI, 2024. Llama3/MODEL_CARD.md. https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md [Accessed on May 25, 2026].

[39]Moustafa N, Slay J, 2015. UNSW-NB15: a comprehensive data set for network intrusion detection systems (UNSW-NB15 network data set). Military Communications and Information Systems Conf, p.1-6.

[40]Mozo A, González-Prieto Á, Pastor A, et al., 2022. Synthetic flow-based cryptomining attack generation through generative adversarial networks. Sci Rep, 12(1):2091.

[41]Nukavarapu SK, Ayyat M, Nadeem T, 2022. MirageNet - towards a GAN-based framework for synthetic network traffic generation. IEEE Global Communications Conf, p.3089-3095.

[42]Oh S, Oh MJ, Im JK, et al., 2025. Spatio-temporal data augmentation method for network traffic prediction. IEEE Access, 13:138686-138698.

[43]Poisson M, Carnier RM, Fukuda K, 2024. GothX: a generator of customizable, legitimate and malicious IoT network traffic. Proc 17^th Cyber Security Experimentation and Test Workshop, p.65-73.

[44]Qi XQ, Chai HY, Yu L, et al., 2024. Regional features conditioned diffusion models for 5G network traffic generation. Proc 32^nd ACM Int Conf on Advances in Geographic Information Systems, p.396-409.

[45]Ring M, Schlör D, Landes D, et al., 2019. Flow-based network traffic generation using generative adversarial networks. Comput Secur, 82:156-172.

[46]Shafi M, Lashkari AH, Roudsari AH, 2025. NTLFlowLyzer: towards generating an intrusion detection dataset and intruders behavior profiling through network and transport layers traffic analysis and pattern extraction. Comput Secur, 148:104160.

[47]Shahid MR, Blanc G, Jmila H, et al., 2020. Generative deep learning for Internet of Things network traffic generation. IEEE 25^th Pacific Rim Int Symp on Dependable Computing, p.70-79.

[48]Shapira T, Shavitt Y, 2021. FlowPic: a generic representation for encrypted traffic classification and applications identification. IEEE Trans Netw Serv Manage, 18(2):1218-1232.

[49]Sharafaldin I, Habibi Lashkari A, Ghorbani AA, 2018. Toward generating a new intrusion detection dataset and intrusion traffic characterization. Proc 4^th Int Conf on Information Systems Security and Privacy, p.108-116.

[50]Sharafaldin I, Lashkari AH, Hakak S, et al., 2019. Developing realistic distributed denial of service (DDoS) attack dataset and taxonomy. Int Carnahan Conf on Security Technology, p.1-8.

[51]Shawkat M, Badawi M, El-ghamrawy S, et al., 2022. An optimized FP-growth algorithm for discovery of association rules. J Supercomput, 78:5479-5506.

[52]Shin CY, Choi YS, Kim MS, 2025. Data augmentation-based enhancement for efficient network traffic classification. IEEE Access, 13:6006-6028.

[53]Sivaroopan N, Bandara D, Madarasingha C, et al., 2024. NetDiffus: network traffic generation by diffusion models through time-series imaging. Comput Netw, 251:110616.

[54]Sivaroopan N, Silva K, Madarasingha C, et al., 2026. A comprehensive survey on synthetic network traffic generation. IEEE Commun Surv Tutor, 28:5949-5983.

[55]Soper J, Xu Y, Foo E, et al., 2024. Improved packet-level synthetic network traffic generation. IEEE 23^rd Int Conf on Trust, Security and Privacy in Computing and Communications, p.1928-1934.

[56]Sun DY, Chen JQ, Gong C, et al., 2024. NetDPSyn: synthesizing network traces under differential privacy. Proc ACM on Internet Measurement Conf, p.545-554.

[57]Sun PS, Yun XC, Li SH, et al., 2025. AdvTG: an adversarial traffic generation framework to deceive DL-based malicious traffic detection models. Proc ACM on Web Conf, p.3147-3159.

[58]Tavallaee M, Bagheri E, Lu W, et al., 2009. A detailed analysis of the KDD CUP 99 data set. IEEE Symp on Computational Intelligence for Security and Defense Applications, p.1-6.

[59]Wang BY, Wang BS, Wei ZL, et al., 2025. MFSI: multi-flow based service identification for encrypted network traffic. Comput Netw, 265:111283.

[60]Wang MX, Yang N, Forcade-Perkins NJ, et al., 2024. ProGen: projection-based adversarial attack generation against network intrusion detection. IEEE Trans Inform Forensics Secur, 19:5476-5491.

[61]Yang LM, Wang YJ, Liu L, et al., 2025. unFlowS: an unsupervised construction scheme of flow spectrum for network traffic detection. IEEE Trans Inform Forensics Secur, 20:3330-3345.

[62]Yin YC, Lin Z, Jin M, et al., 2022. Practical GAN-based synthetic IP header trace generation using NetShare. Proc ACM SIGCOMM Conf, p.458-472.

[63]Zhang HZ, 2024. TransFlowGAN: generation and balancing of network traffic data. Proc Asia Pacific Conf on Computing Technologies, Communications and Networking, p.51-59.

[64]Zhang JH, Tang JQ, Zhang X, et al., 2015. A survey of network traffic generation. 3^rd Int Conf on Cyberspace Technology, p.1-6.

[65]Zhang S, Azizi S, Joshi A, et al., 2025. Towards behavior grammar-driven IoT network traffic generation using MUD specifications. Proc 7^th Joint Workshop on CPS & IoT Security and Privacy, p.98-104.

[66]Zhang SY, Li T, Jin DP, et al., 2024. NetDiff: a service-guided hierarchical diffusion model for network flow trace generation. Proc ACM Netw, 2(CoNEXT3):16.

[67]Zhu XF, Shu NN, Zheng BW, et al., 2022. A method to generate a ground truth distributed network traffic dataset. Proc 3^rd Int Conf on Control, Robotics and Intelligent System, p.219-224.

[68]Zion Y, Aharon P, Dubin R, et al., 2025. Enhancing encrypted Internet traffic classification through advanced data augmentation techniques. IEEE Int Conf on Communications, p.1-6.

Similar articles