{"title":"TL-ConvLSTM:一种基于迁移学习的卷积LSTM,用于识别和预测未来环境中的流量","authors":"Bikash Chandra Singh;Peter Foytik;Rafael Diaz;Sachin Shetty","doi":"10.1109/JSYST.2025.3569445","DOIUrl":null,"url":null,"abstract":"Forecasting and categorizing cellular traffic flows and their types are essential functions in intelligent network systems to ensure efficient network optimization. The ever-evolving nature of 5G networks results in fluctuations in traffic patterns over time, leading to a phenomenon known as model drift. Consequently, accurately predicting and identifying cellular traffic patterns becomes a complex task. To tackle this challenge, this article introduces an innovative approach called <italic>TL-ConvLSTM</i>, which combines transfer learning with convolutional long short-term memory (ConvLSTM) to effectively combat model drift and provide precise forecasting and recognition of cellular traffic within the network. To accomplish this, we initiate the training of <italic>TL-ConvLSTM</i> by estimating its parameters from the source domain. We then employ the Kolmogorov–Smirnov method to adapt the model within the target domain, fine tuning its weights. To improve the precision of this model adaptation, we systematically explore optimal learning windows. This exploration includes adjusting window size for time-series data and feature dimensions to capture dynamic traffic patterns in a 5G environment. Furthermore, we make use of the Amarisoft 5G testbed in our lab to create a 12-day time-series dataset. This dataset includes various features related to traffic flows and their patterns. We showcase the effectiveness of our approach through a set of experiments.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"19 2","pages":"358-369"},"PeriodicalIF":4.4000,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"TL-ConvLSTM: A Transfer-Learning-Based Convolutional LSTM to Identify and Forecast Traffic in the NextG Environments\",\"authors\":\"Bikash Chandra Singh;Peter Foytik;Rafael Diaz;Sachin Shetty\",\"doi\":\"10.1109/JSYST.2025.3569445\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Forecasting and categorizing cellular traffic flows and their types are essential functions in intelligent network systems to ensure efficient network optimization. The ever-evolving nature of 5G networks results in fluctuations in traffic patterns over time, leading to a phenomenon known as model drift. Consequently, accurately predicting and identifying cellular traffic patterns becomes a complex task. To tackle this challenge, this article introduces an innovative approach called <italic>TL-ConvLSTM</i>, which combines transfer learning with convolutional long short-term memory (ConvLSTM) to effectively combat model drift and provide precise forecasting and recognition of cellular traffic within the network. To accomplish this, we initiate the training of <italic>TL-ConvLSTM</i> by estimating its parameters from the source domain. We then employ the Kolmogorov–Smirnov method to adapt the model within the target domain, fine tuning its weights. To improve the precision of this model adaptation, we systematically explore optimal learning windows. This exploration includes adjusting window size for time-series data and feature dimensions to capture dynamic traffic patterns in a 5G environment. Furthermore, we make use of the Amarisoft 5G testbed in our lab to create a 12-day time-series dataset. This dataset includes various features related to traffic flows and their patterns. We showcase the effectiveness of our approach through a set of experiments.\",\"PeriodicalId\":55017,\"journal\":{\"name\":\"IEEE Systems Journal\",\"volume\":\"19 2\",\"pages\":\"358-369\"},\"PeriodicalIF\":4.4000,\"publicationDate\":\"2025-03-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Systems Journal\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/11016909/\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, INFORMATION SYSTEMS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Systems Journal","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/11016909/","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
TL-ConvLSTM: A Transfer-Learning-Based Convolutional LSTM to Identify and Forecast Traffic in the NextG Environments
Forecasting and categorizing cellular traffic flows and their types are essential functions in intelligent network systems to ensure efficient network optimization. The ever-evolving nature of 5G networks results in fluctuations in traffic patterns over time, leading to a phenomenon known as model drift. Consequently, accurately predicting and identifying cellular traffic patterns becomes a complex task. To tackle this challenge, this article introduces an innovative approach called TL-ConvLSTM, which combines transfer learning with convolutional long short-term memory (ConvLSTM) to effectively combat model drift and provide precise forecasting and recognition of cellular traffic within the network. To accomplish this, we initiate the training of TL-ConvLSTM by estimating its parameters from the source domain. We then employ the Kolmogorov–Smirnov method to adapt the model within the target domain, fine tuning its weights. To improve the precision of this model adaptation, we systematically explore optimal learning windows. This exploration includes adjusting window size for time-series data and feature dimensions to capture dynamic traffic patterns in a 5G environment. Furthermore, we make use of the Amarisoft 5G testbed in our lab to create a 12-day time-series dataset. This dataset includes various features related to traffic flows and their patterns. We showcase the effectiveness of our approach through a set of experiments.
期刊介绍:
This publication provides a systems-level, focused forum for application-oriented manuscripts that address complex systems and system-of-systems of national and global significance. It intends to encourage and facilitate cooperation and interaction among IEEE Societies with systems-level and systems engineering interest, and to attract non-IEEE contributors and readers from around the globe. Our IEEE Systems Council job is to address issues in new ways that are not solvable in the domains of the existing IEEE or other societies or global organizations. These problems do not fit within traditional hierarchical boundaries. For example, disaster response such as that triggered by Hurricane Katrina, tsunamis, or current volcanic eruptions is not solvable by pure engineering solutions. We need to think about changing and enlarging the paradigm to include systems issues.