IEEE TNSE Special Issue 2022 : IEEE TNSE Special Issue on Next-generation Traffic Measurement with Network-wide Perspective and Artificial Intelligence
Call For Papers
Traffic measurement is deemed as the bedrock of the next-generation network systems. Its function is to monitor network traffic at all protocol layers, from the physical layer to the applications, and to capture traffic patterns, relationships, and anomalies in the time dimension and volume dimension to support fundamental network functions and upper-layer services, such as load balancing, routing, intrusion detection, traffic engineering, and performance diagnosis. Recently, the explosive Internet traffic growth, the emerging networking paradigms, and the surging network service demands have opened new challenges for traffic measurement, which have gained significant attention from both academia and industry. However, the state-of-the-art solutions, which mainly focus on single-point measurement scenarios and derive probabilistic formulas to measure elementary metrics like frequency, cardinality, and persistence, cannot meet the arising heterogeneous and fine-grained measurement requirements on performance, throughput, scalability, response time, and diversity. For example, modern switches can forward packets at extremely high throughput (up to several Gpps), practically two orders of magnitude higher than the throughput of existing sketch solutions. For another example, application-oriented sketches that provide timely and accurate features beyond elementary metrics to applications like traffic engineering and intrusion detection systems can undoubtedly benefit such systems, while the design of such sketches and the derivation of measurement formulas are non-trivial problems. Thus, there is an urgent need for systematic and in-depth research on network-wide and AI-powered traffic measurement methods to meet new network traffic characteristics and support emerging applications.
It is expected that the next-generation network management systems will feature network-wide measurement algorithms. The big network data is distributed in nature as the sources and destinations of connections may span the entire network. It is thus essential to aggregate the views of multiple measurement points to build a network-wide perception and capture comprehensive and accurate traffic information. Besides, with a proper task breakdown schema, multiple measurement points can federatively run measurements in a completely parallel and distributed manner, reducing the computation overhead and hardware requirement. Another latest trend involves artificial intelligence technologies that allow seamless aggregation of multi-resolution and heterogeneous network traffic data while advancing traffic measurement systems' design, deployment, and application. Additionally, the interplay between traffic measurement and artificial intelligence is bidirectional. Besides using AI intelligence to power traffic measurement, traffic measurement methods can also aid the AI systems since AI systems are often deployed in a decentralized environment where communication plays an important role. For instance, sketches, a family of traditional measurement methods, can naturally compress the input data and approximate its distribution. It can be used as the medium to transfer gradients among distributed AI systems, striking a tradeoff among information compression, convergence time, and system accuracy.
This special issue is intended to encourage scholars and experts to systematically discuss the latest research progress and development trends for next-generation traffic measurement, promote in-depth research, and share academic and technical achievements.
The topics of this SI include, but are not limited to:
- Network-wide traffic measurement algorithms and systems
- Use of artificial intelligence, machine learning, and data analytics in network traffic measurement and its applications
- Network-wide and/or AI-powered traffic measurement for the Internet, edge networks, data center networks, cloud-based systems, software-defined networks, online social networks, online services, and next-generation networks
- Traffic measurement with programmable hardware and software platforms
- Traffic measurement with privacy preservation and anonymization
- AI-powered design, simulation, modeling, analysis, and visualization for next-generation traffic measurement
- Validation and repeatability of network-wide and/or AI-powered traffic measurements, shared datasets, or collaborative platforms
- Novel applications of network-wide and AI-powered traffic measurement for load balancing, flow scheduling, network management, and network evolution
- Novel applications of network-wide and AI-powered traffic measurement for security, anomaly/vulnerability/attack detection, and user profiling/privacy
- Novel applications of traffic measurement for artificial intelligence and machine learning
Papers submitted to this special issue for possible publication must be original and must not be under consideration for publication in any other journal or conference.
All papers are to be submitted through the journal editorial submission system. At the beginning of the submission process in the submission system, authors need to select “Special Issue on Next-generation Traffic Measurement with Network-wide Perspective and Artificial Intelligence” as the article type.
All manuscripts must be prepared according to the journal publication guidelines, which can also be found on the website provided above. Papers will be evaluated following the journal's standard review process.
All papers will be reviewed by at least three reviewers for their technical merit, scope, and relevance to the CFP.
Manuscripts Due: 15 December 2022
First-Round Peer Reviews to Authors: 15 March 2023
Revised Manuscripts Due: 1 May 2023
Final Decision: 1 June 2023
Estimated Publication Date: Fourth Quarter 2023
He Huang (Lead), Soochow University, China
Shigang Chen, University of Florida, USA
Ran Ben Basat, University College London, UK
Haipeng Dai, Nanjing University, China
Amirhosein Taherkordi, University of Oslo, Norway
Jun (Jim) Xu, Georgia Institute of Technology, USA