{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,22]],"date-time":"2026-03-22T09:59:55Z","timestamp":1774173595405,"version":"3.50.1"},"reference-count":57,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T00:00:00Z","timestamp":1771372800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000780","name":"European Commission","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100021856","name":"Ministero dell'Universit\u00e0 e della Ricerca","doi-asserted-by":"publisher","award":["D.D.104-02\/02\/2022"],"award-info":[{"award-number":["D.D.104-02\/02\/2022"]}],"id":[{"id":"10.13039\/501100021856","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Computer Networks"],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1016\/j.comnet.2026.112090","type":"journal-article","created":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T16:57:04Z","timestamp":1772038624000},"page":"112090","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["A multimodal and perturbation-aware learning approach for robust traffic classification"],"prefix":"10.1016","volume":"280","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-6141-0188","authenticated-orcid":false,"given":"Idio","family":"Guarino","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0997-9724","authenticated-orcid":false,"given":"Giampaolo","family":"Bovenzi","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7395-4222","authenticated-orcid":false,"given":"Alfredo","family":"Nascita","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6230-2958","authenticated-orcid":false,"given":"Domenico","family":"Ciuonzo","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3467-1166","authenticated-orcid":false,"given":"Damiano","family":"Carra","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0221-7444","authenticated-orcid":false,"given":"Antonio","family":"Pescap\u00e8","sequence":"additional","affiliation":[]}],"member":"78","reference":[{"issue":"1\u20134","key":"10.1016\/j.comnet.2026.112090_bib0001","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1109\/SURV.2008.080406","article-title":"A survey of techniques for internet traffic classification using machine learning","volume":"10","author":"Nguyen","year":"2008","journal-title":"IEEE Commun. Surv. Tutorials"},{"key":"10.1016\/j.comnet.2026.112090_bib0002","first-page":"1","article-title":"Towards the deployment of machine learning solutions in network traffic classification: a systematic survey","author":"Pacheco","year":"2018","journal-title":"IEEE Commun. Surv. Tutorials"},{"issue":"9","key":"10.1016\/j.comnet.2026.112090_bib0003","doi-asserted-by":"crossref","first-page":"168","DOI":"10.1109\/MCOM.001.2300361","article-title":"AI-powered internet traffic classification: past, present, and future","volume":"62","author":"Aceto","year":"2024","journal-title":"IEEE Commun. Mag."},{"issue":"2","key":"10.1016\/j.comnet.2026.112090_bib0004","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1145\/1129582.1129589","article-title":"Traffic classification on the fly","volume":"36","author":"Bernaille","year":"2006","journal-title":"ACM SIGCOMM Comput. Commun. Rev."},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0005","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1145\/1198255.1198257","article-title":"Traffic classification through simple statistical fingerprinting","volume":"37","author":"Crotti","year":"2007","journal-title":"ACM SIGCOMM Comput. Commun. Rev."},{"key":"10.1016\/j.comnet.2026.112090_bib0006","series-title":"IEEE Int. Conf. Commun. (ICC)","first-page":"1","article-title":"Identification of traffic flows hiding behind TCP port 80","author":"Dainotti","year":"2010"},{"key":"10.1016\/j.comnet.2026.112090_bib0007","series-title":"Proceedings of the ACM Internet Measurement Conference (IMC)","first-page":"65","article-title":"Wire-Speed statistical classification of network traffic on commodity hardware","author":"P. M. Santiago del Rio","year":"2012"},{"key":"10.1016\/j.comnet.2026.112090_bib0008","series-title":"IEEE 4th International Conference on Computing, Communications and Security (ICCCS)","first-page":"1","article-title":"MIRAGE: mobile-app traffic capture and ground-truth creation","author":"Aceto","year":"2019"},{"key":"10.1016\/j.comnet.2026.112090_bib0009","doi-asserted-by":"crossref","DOI":"10.1016\/j.comnet.2022.109452","article-title":"Contextual counters and multimodal deep learning for activity-level traffic classification of mobile communication apps during COVID-19 pandemic","volume":"219","author":"Guarino","year":"2022","journal-title":"Elsevier Comput. Netw."},{"key":"10.1016\/j.comnet.2026.112090_bib0010","article-title":"The applications of deep learning on traffic identification","author":"Wang","year":"2015","journal-title":"BlackHat USA"},{"key":"10.1016\/j.comnet.2026.112090_bib0011","series-title":"IEEE International Conference on Intelligence and Security Informatics (ISI)","first-page":"43","article-title":"End-to-end encrypted traffic classification with one-dimensional convolution neural networks","author":"Wang","year":"2017"},{"issue":"2","key":"10.1016\/j.comnet.2026.112090_bib0012","doi-asserted-by":"crossref","first-page":"445","DOI":"10.1109\/TNSM.2019.2899085","article-title":"Mobile encrypted traffic classification using deep learning: experimental evaluation, lessons learned, and challenges","volume":"16","author":"Aceto","year":"2019","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"key":"10.1016\/j.comnet.2026.112090_bib0013","doi-asserted-by":"crossref","first-page":"18042","DOI":"10.1109\/ACCESS.2017.2747560","article-title":"Network traffic classifier with convolutional and recurrent neural networks for internet of things","volume":"5","author":"Lopez-Martin","year":"2017","journal-title":"IEEE Access"},{"issue":"3","key":"10.1016\/j.comnet.2026.112090_bib0014","article-title":"Deep packet: a novel approach for encrypted traffic classification using deep learning","volume":"24","author":"Lotfollahi","year":"2020","journal-title":"Springer Soft Comput."},{"key":"10.1016\/j.comnet.2026.112090_bib0015","doi-asserted-by":"crossref","DOI":"10.1016\/j.comnet.2019.106944","article-title":"MIMETIC: Mobile encrypted traffic classification using multimodal deep learning","volume":"165","author":"Aceto","year":"2019","journal-title":"Elsevier Comput. Netw."},{"issue":"4","key":"10.1016\/j.comnet.2026.112090_bib0016","doi-asserted-by":"crossref","first-page":"4225","DOI":"10.1109\/TNSM.2021.3098157","article-title":"XAI meets mobile traffic classification: understanding and improving multimodal deep learning architectures","volume":"18","author":"Nascita","year":"2021","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0017","doi-asserted-by":"crossref","DOI":"10.1007\/s11704-023-2386-4","article-title":"MMCo: Using multimodal deep learning to detect malicious traffic with noisy labels","volume":"18","author":"Yuan","year":"2023","journal-title":"Front. Comput. Sci."},{"key":"10.1016\/j.comnet.2026.112090_bib0018","doi-asserted-by":"crossref","first-page":"182065","DOI":"10.1109\/ACCESS.2020.3029190","article-title":"Automatic mobile app identification from encrypted traffic with hybrid neural networks","volume":"8","author":"Wang","year":"2020","journal-title":"IEEE Access"},{"key":"10.1016\/j.comnet.2026.112090_bib0019","series-title":"Proceedings of the Springer International Conference on Information Systems Security and Privacy (ICISSP)","first-page":"407","article-title":"Characterization of encrypted and VPN traffic using time-related features","author":"Draper-Gil","year":"2016"},{"key":"10.1016\/j.comnet.2026.112090_bib0020","series-title":"IEEE 20th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob)","first-page":"663","article-title":"Mirage-app\u00d7act-2024: a novel dataset for mobile app and activity traffic analysis","author":"Guarino","year":"2024"},{"key":"10.1016\/j.comnet.2026.112090_bib0021","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1016\/j.jnca.2018.10.018","article-title":"Mobile app traffic flow feature extraction and selection for improving classification robustness","volume":"125","author":"Liu","year":"2019","journal-title":"Elsevier J. Netw. Comput. Appl."},{"issue":"2","key":"10.1016\/j.comnet.2026.112090_bib0022","doi-asserted-by":"crossref","first-page":"1962","DOI":"10.1109\/TNSM.2021.3052888","article-title":"Adversarial network traffic: towards evaluating the robustness of deep-learning-based network traffic classification","volume":"18","author":"Sadeghzadeh","year":"2021","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"key":"10.1016\/j.comnet.2026.112090_bib0023","doi-asserted-by":"crossref","first-page":"3225","DOI":"10.1109\/TIFS.2022.3201377","article-title":"TANTRA: timing-based adversarial network traffic reshaping attack","volume":"17","author":"Sharon","year":"2022","journal-title":"IEEE Trans. Inf. Forensics Secur."},{"issue":"3","key":"10.1016\/j.comnet.2026.112090_bib0024","doi-asserted-by":"crossref","first-page":"1294","DOI":"10.1109\/TNET.2021.3137084","article-title":"Adversarial attacks against deep learning-based network intrusion detection systems and defense mechanisms","volume":"30","author":"Zhang","year":"2022","journal-title":"IEEE\/ACM Trans. Networking"},{"key":"10.1016\/j.comnet.2026.112090_bib0025","series-title":"IEEE 7th Network Traffic Measurement and Analysis Conference (TMA)","first-page":"1","article-title":"Many or few samples?: comparing transfer, contrastive and meta-learning in encrypted traffic classification","author":"Guarino","year":"2023"},{"key":"10.1016\/j.comnet.2026.112090_bib0026","series-title":"Proceedings of the 32nd USENIX Security Symposium","first-page":"625","article-title":"Rosetta: enabling robust TLS encrypted traffic classification in diverse network environments with TCP-aware traffic augmentation","author":"Xie","year":"2023"},{"key":"10.1016\/j.comnet.2026.112090_bib0027","doi-asserted-by":"crossref","first-page":"2524","DOI":"10.1109\/TIFS.2023.3267885","article-title":"Prism: real-time privacy protection against temporal network traffic analyzers","volume":"18","author":"Li","year":"2023","journal-title":"IEEE Trans. Inf. Forensics Secur."},{"key":"10.1016\/j.comnet.2026.112090_bib0028","doi-asserted-by":"crossref","first-page":"7659","DOI":"10.1109\/TIFS.2024.3442530","article-title":"Cactus: obfuscating bidirectional encrypted TCP traffic at client side","volume":"19","author":"Xie","year":"2024","journal-title":"IEEE Trans. Inf. Forensics Secur."},{"key":"10.1016\/j.comnet.2026.112090_bib0029","series-title":"Proceedings of the ACM Web Conference (WWW)","first-page":"4417-4428","article-title":"AN-Net: an anti-noise network for anonymous traffic classification","author":"Deng","year":"2024"},{"key":"10.1016\/j.comnet.2026.112090_bib0030","series-title":"Springer Nature Passive and Active Measurement (PAM)","first-page":"159","article-title":"Data augmentation for traffic classification","author":"Wang","year":"2024"},{"key":"10.1016\/j.comnet.2026.112090_bib0031","doi-asserted-by":"crossref","DOI":"10.1016\/j.comnet.2024.110790","article-title":"Towards universal and transferable adversarial attacks against network traffic classification","author":"Ding","year":"2024","journal-title":"Elsevier Comput. Netw."},{"key":"10.1016\/j.comnet.2026.112090_bib0032","series-title":"Proceedings of the 33rd ACM International Conference on Information & Knowledge Management (CIKM)","first-page":"2910","article-title":"TrafCL: robust encrypted malicious traffic detection via contrastive learning","author":"Yang","year":"2024"},{"key":"10.1016\/j.comnet.2026.112090_bib0033","article-title":"The art of time-bending: data augmentation and early prediction for efficient traffic classification","volume":"252","author":"Hajaj","year":"2024","journal-title":"Elsevier Expert Syst. Appl."},{"issue":"3","key":"10.1016\/j.comnet.2026.112090_bib0034","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1109\/LNET.2024.3435723","article-title":"A packet sequence permutation-aware approach to robust network traffic classification","volume":"6","author":"Jiang","year":"2024","journal-title":"IEEE Netw.Lett."},{"issue":"3","key":"10.1016\/j.comnet.2026.112090_bib0035","doi-asserted-by":"crossref","first-page":"3054","DOI":"10.1109\/TNSM.2024.3366848","article-title":"Self-supervised traffic classification: flow embedding and few-shot solutions","volume":"21","author":"Horowicz","year":"2024","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"key":"10.1016\/j.comnet.2026.112090_bib0036","series-title":"Proceedings of the 34th USENIX Security Symposium","first-page":"625","article-title":"Robustifying ML-powered network classifiers with PANTS","author":"Jin","year":"2025"},{"key":"10.1016\/j.comnet.2026.112090_bib0037","article-title":"Adversarial attacks based on time-series features for traffic detection","volume":"148","author":"Lu","year":"2025","journal-title":"Elsevier Compu. Secur."},{"key":"10.1016\/j.comnet.2026.112090_bib0038","article-title":"Entropy-regulated cross-modal generative fusion for multimodal network intrusion detection","volume":"126","author":"Wang","year":"2026","journal-title":"Elsevier Inf. Fusion"},{"key":"10.1016\/j.comnet.2026.112090_bib0039","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1016\/j.jpdc.2019.03.003","article-title":"The security of machine learning in an adversarial setting: a survey","volume":"130","author":"Wang","year":"2019","journal-title":"J. Parallel Distrib. Comput."},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0040","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1109\/MNET.2012.6135854","article-title":"Issues and future directions in traffic classification","volume":"26","author":"Dainotti","year":"2012","journal-title":"IEEE Netw."},{"key":"10.1016\/j.comnet.2026.112090_bib0041","series-title":"IEEE Network Traffic Measurement and Analysis Conference (TMA)","first-page":"1","article-title":"Middleboxes in the internet: a HTTP perspective","author":"Huang","year":"2017"},{"key":"10.1016\/j.comnet.2026.112090_bib0042","series-title":"Proceedings of the 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25)","first-page":"1523","article-title":"CATO: end-to-end optimization of ML-based traffic analysis pipelines","author":"Wan","year":"2025"},{"issue":"3","key":"10.1016\/j.comnet.2026.112090_bib0043","doi-asserted-by":"crossref","first-page":"800","DOI":"10.1109\/TNSM.2019.2933358","article-title":"A survey on big data for network traffic monitoring and analysis","volume":"16","author":"D\u2019Alconzo","year":"2019","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"key":"10.1016\/j.comnet.2026.112090_bib0044","doi-asserted-by":"crossref","DOI":"10.1016\/j.comnet.2024.110374","article-title":"MEMENTO: a novel approach for class incremental learning of encrypted traffic","volume":"245","author":"Cerasuolo","year":"2024","journal-title":"Elsevier Comput. Netw."},{"key":"10.1016\/j.comnet.2026.112090_bib0045","series-title":"IEEE\/IFIP Networking Conference (IFIP Networking)","first-page":"757","article-title":"XAI For interpretable multimodal architectures with contextual input in mobile network traffic classification","author":"Cerasuolo","year":"2024"},{"key":"10.1016\/j.comnet.2026.112090_bib0046","series-title":"Proceedings of the ACM Workshop on Secure Programmable Network Infrastructure (SPIN \u201920)","first-page":"35","article-title":"Measuring TCP round-trip time in the data plane","author":"Chen","year":"2020"},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0047","first-page":"1","article-title":"A survey on image data augmentation for deep learning","volume":"6","author":"Shorten","year":"2019","journal-title":"Springer J. Big Data"},{"key":"10.1016\/j.comnet.2026.112090_bib0048","series-title":"Proceedings of the 37th ACM International Conference on Machine Learning (CML\u201920)","first-page":"1597","article-title":"A simple framework for contrastive learning of visual representations","author":"Chen","year":"2020"},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0049","first-page":"857","article-title":"Self-supervised learning: generative or contrastive","volume":"35","author":"Liu","year":"2023","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"10.1016\/j.comnet.2026.112090_bib0050","series-title":"Proceedings of the 34th International Conference on Neural Information Processing Systems (NIPS \u201920)","first-page":"18661","article-title":"Supervised contrastive learning","volume":"33","author":"Khosla","year":"2020"},{"issue":"2","key":"10.1016\/j.comnet.2026.112090_bib0051","doi-asserted-by":"crossref","first-page":"224","DOI":"10.1109\/TPAMI.1979.4766909","article-title":"A cluster separation measure","volume":"PAMI-1","author":"Davies","year":"1979","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.comnet.2026.112090_bib0052","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: a graphical aid to the interpretation and validation of cluster analysis","volume":"20","author":"Rousseeuw","year":"1987","journal-title":"Elsevier J. Comput. Appl. Math."},{"key":"10.1016\/j.comnet.2026.112090_bib0053","series-title":"Proc. of the 33rd International Conference on Neural Information Processing Systems (NeurIPS)","first-page":"12316","article-title":"Beyond temperature scaling: obtaining well-calibrated multiclass probabilities with dirichlet calibration","author":"Kull","year":"2019"},{"key":"10.1016\/j.comnet.2026.112090_bib0054","series-title":"Proceedings of the ACM 29th AAAI Conference on Artificial Intelligence (AAAI\u201915)","first-page":"2901-2907","article-title":"Obtaining well calibrated probabilities using bayesian binning","author":"Naeini","year":"2015"},{"issue":"4","key":"10.1016\/j.comnet.2026.112090_bib0055","doi-asserted-by":"crossref","first-page":"4103","DOI":"10.1109\/TNSM.2021.3122940","article-title":"Deep learning and zero-day traffic classification: lessons learned from a commercial-grade dataset","volume":"18","author":"Yang","year":"2021","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"issue":"1","key":"10.1016\/j.comnet.2026.112090_bib0056","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1109\/TNSM.2022.3203389","article-title":"Lossdetection: real-time packet loss monitoring system for sampled traffic data","volume":"20","author":"Wu","year":"2023","journal-title":"IEEE Trans. Netw. Serv. Manage."},{"issue":"6","key":"10.1016\/j.comnet.2026.112090_bib0057","doi-asserted-by":"crossref","first-page":"2614","DOI":"10.1109\/TNET.2023.3262564","article-title":"Enhancing TCP via hysteresis switching: theoretical analysis and empirical evaluation","volume":"31","author":"Ahmed","year":"2023","journal-title":"IEEE\/ACM Trans. Networking"}],"container-title":["Computer Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1389128626001027?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1389128626001027?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,3,22]],"date-time":"2026-03-22T09:42:49Z","timestamp":1774172569000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S1389128626001027"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":57,"alternative-id":["S1389128626001027"],"URL":"https:\/\/doi.org\/10.1016\/j.comnet.2026.112090","relation":{},"ISSN":["1389-1286"],"issn-type":[{"value":"1389-1286","type":"print"}],"subject":[],"published":{"date-parts":[[2026,5]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"A multimodal and perturbation-aware learning approach for robust traffic classification","name":"articletitle","label":"Article Title"},{"value":"Computer Networks","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.comnet.2026.112090","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 The Author(s). Published by Elsevier B.V.","name":"copyright","label":"Copyright"}],"article-number":"112090"}}