{"title":"基于深度学习的调制和信道编码联合识别","authors":"Fang Deng, Xingrong Huang, Shujun Sun, Shengliang Peng","doi":"10.1109/ICCT56141.2022.10073366","DOIUrl":null,"url":null,"abstract":"The tasks of identifying which modulation format and channel coding scheme have been utilized by the wireless signal are crucial to intelligent communications, electronic warfare, and spectrum management. Currently, most research on these tasks digs into either modulation identification or channel coding identification, while the joint identification of modulation and channel coding has not been fully investigated. In this paper, we propose two joint identification algorithms based on deep learning to handle the tasks of modulation identification and channel coding identification simultaneously. The first algorithm adopts a successive architecture, in which two neural networks are used to handle the two identification tasks, respectively. The second algorithm exploits multi-task deep learning, with which two identification tasks can be completed using a single neural network. Experiments results show that, considering the candidate set of three modulation formats and two channel coding schemes, both algorithms achieve satisfactory identification accuracy, and the latter is superior to the former with the accuracy improvement of 10%.","PeriodicalId":294057,"journal":{"name":"2022 IEEE 22nd International Conference on Communication Technology (ICCT)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Joint Identification of Modulation and Channel Coding Based on Deep Learning\",\"authors\":\"Fang Deng, Xingrong Huang, Shujun Sun, Shengliang Peng\",\"doi\":\"10.1109/ICCT56141.2022.10073366\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The tasks of identifying which modulation format and channel coding scheme have been utilized by the wireless signal are crucial to intelligent communications, electronic warfare, and spectrum management. Currently, most research on these tasks digs into either modulation identification or channel coding identification, while the joint identification of modulation and channel coding has not been fully investigated. In this paper, we propose two joint identification algorithms based on deep learning to handle the tasks of modulation identification and channel coding identification simultaneously. The first algorithm adopts a successive architecture, in which two neural networks are used to handle the two identification tasks, respectively. The second algorithm exploits multi-task deep learning, with which two identification tasks can be completed using a single neural network. Experiments results show that, considering the candidate set of three modulation formats and two channel coding schemes, both algorithms achieve satisfactory identification accuracy, and the latter is superior to the former with the accuracy improvement of 10%.\",\"PeriodicalId\":294057,\"journal\":{\"name\":\"2022 IEEE 22nd International Conference on Communication Technology (ICCT)\",\"volume\":\"12 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-11-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE 22nd International Conference on Communication Technology (ICCT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCT56141.2022.10073366\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 22nd International Conference on Communication Technology (ICCT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCT56141.2022.10073366","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Joint Identification of Modulation and Channel Coding Based on Deep Learning
The tasks of identifying which modulation format and channel coding scheme have been utilized by the wireless signal are crucial to intelligent communications, electronic warfare, and spectrum management. Currently, most research on these tasks digs into either modulation identification or channel coding identification, while the joint identification of modulation and channel coding has not been fully investigated. In this paper, we propose two joint identification algorithms based on deep learning to handle the tasks of modulation identification and channel coding identification simultaneously. The first algorithm adopts a successive architecture, in which two neural networks are used to handle the two identification tasks, respectively. The second algorithm exploits multi-task deep learning, with which two identification tasks can be completed using a single neural network. Experiments results show that, considering the candidate set of three modulation formats and two channel coding schemes, both algorithms achieve satisfactory identification accuracy, and the latter is superior to the former with the accuracy improvement of 10%.