Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065710
Yiming Shi, Zhen Meng, Xianling Zeng, Anfu Zhou
People-aware is a way of perceiving a person's iden-tity through their biometrics. Currently, it plays a very important role in identity verification in application scenarios such as smart homes and security checks. Compared to traditional person identification technologies, mm Wave based person sensing has the unique advantages of being non-contact, not affected by the environment, and highly private and confidential. The current direct output of person point cloud results from TI packaged sensors suffers from small quantities and unclear target contours, limiting various point cloud data recognition applications such as gait recognition, status recognition, etc. In this paper, we collect mm Wave datasets of people walking and propose an Optimise-CFAR target detection optimisation algorithm based on the signal processing process, which can effectively remove the number of edge noise points and thus improve the quality of the point cloud output, and process the point cloud data of people into time series with the help of a person identification model. After experimental analysis, we found that the optimised point cloud data was able to improve the average accuracy of person classification by 93%.
{"title":"People-Aware mmWave Point Cloud Processing Algorithm","authors":"Yiming Shi, Zhen Meng, Xianling Zeng, Anfu Zhou","doi":"10.1109/ICCC56324.2022.10065710","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065710","url":null,"abstract":"People-aware is a way of perceiving a person's iden-tity through their biometrics. Currently, it plays a very important role in identity verification in application scenarios such as smart homes and security checks. Compared to traditional person identification technologies, mm Wave based person sensing has the unique advantages of being non-contact, not affected by the environment, and highly private and confidential. The current direct output of person point cloud results from TI packaged sensors suffers from small quantities and unclear target contours, limiting various point cloud data recognition applications such as gait recognition, status recognition, etc. In this paper, we collect mm Wave datasets of people walking and propose an Optimise-CFAR target detection optimisation algorithm based on the signal processing process, which can effectively remove the number of edge noise points and thus improve the quality of the point cloud output, and process the point cloud data of people into time series with the help of a person identification model. After experimental analysis, we found that the optimised point cloud data was able to improve the average accuracy of person classification by 93%.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121542921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065661
Shengwu Wu, Shuai Zhang, Qiyishu Li, Zhikun Xu
In this paper, the availability of the non-orthogonal multiple access (NOMA) to the satellite communication system is verified firstly by analyzing the distribution of the absolute value of any two satellite users' signal-to-noise ratio (SNR) gap. Then, the uplink and downlink NOMA schemes which are suitable for the satellite communications are studied. Finally, the link-level simulation is carried out under the satellite channel. The simulation results show that compared with the orthogonal multiple access (OMA), NOMA performs better in the satellite communication systems.
{"title":"Study of Non-Orthogonal Multiple Access Technology for Satellite Communications","authors":"Shengwu Wu, Shuai Zhang, Qiyishu Li, Zhikun Xu","doi":"10.1109/ICCC56324.2022.10065661","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065661","url":null,"abstract":"In this paper, the availability of the non-orthogonal multiple access (NOMA) to the satellite communication system is verified firstly by analyzing the distribution of the absolute value of any two satellite users' signal-to-noise ratio (SNR) gap. Then, the uplink and downlink NOMA schemes which are suitable for the satellite communications are studied. Finally, the link-level simulation is carried out under the satellite channel. The simulation results show that compared with the orthogonal multiple access (OMA), NOMA performs better in the satellite communication systems.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"9 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114001197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Direct sequence spread spectrum technology can improve the bit error rate performance in the application of filter bank based multicarrier system. Aiming at a communication system that combines the spread spectrum technology and the offset quadrature amplitude modulation based filter bank multi-carrier (FBMC/OQAM), this paper proposes a pilot frequency design scheme. The traditional pre-pilot design scheme does not consider the problems of high overhead and low power efficiency, therefore the channel estimation result is inaccurate. The proposed method in this paper combines the symmetry of the spread-spectrum signal and the imaginary interference factor, reduces other overheads except for the pilot, and improves the power of the pilot symbol. The experimental results show that the proposed method has a better bit error rate performance and lower mean square error than the auxiliary pilot method which results in no overhead of additional symbol positions around the pilot. In addition, this study analyzed the spectral efficiency, power efficiency and residual imaginary interference of the pilot design scheme.
{"title":"Pilot Design with Low Overhead in Spread-Spectrum-Based FBMC/OQAM Systems","authors":"Kaiwen Huang, Zhongnian Li, Hongbo Xu, Guoping Zhang","doi":"10.1109/ICCC56324.2022.10066052","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10066052","url":null,"abstract":"Direct sequence spread spectrum technology can improve the bit error rate performance in the application of filter bank based multicarrier system. Aiming at a communication system that combines the spread spectrum technology and the offset quadrature amplitude modulation based filter bank multi-carrier (FBMC/OQAM), this paper proposes a pilot frequency design scheme. The traditional pre-pilot design scheme does not consider the problems of high overhead and low power efficiency, therefore the channel estimation result is inaccurate. The proposed method in this paper combines the symmetry of the spread-spectrum signal and the imaginary interference factor, reduces other overheads except for the pilot, and improves the power of the pilot symbol. The experimental results show that the proposed method has a better bit error rate performance and lower mean square error than the auxiliary pilot method which results in no overhead of additional symbol positions around the pilot. In addition, this study analyzed the spectral efficiency, power efficiency and residual imaginary interference of the pilot design scheme.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124494978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065890
Xinjian Zhao, Fei Xia, Guoquan Yuan, Shi Chen, Song Zhang
Network intrusion detection (NID) has attracted much attention as it is essential in preventing security threats and protecting networks from attacks. However, existing methods face the following challenges: (1) poor feature extraction capability; (2) not well-designed to address the class imbalance problem; (3) failure to take full use of label information and learn classification-oriented features, degrading the NID performance. To this end, we proposed SC-Net, a two-stage training model with deep supervised learning and a normalized classifier, to overcome the abovementioned challenges. During the pretraining stage, the learned embedding will be optimized by both a supervised contrastive loss and a classification loss, so that the embedding with the same label will be more compact in the feature space. After that, in the finetuning stage, the weight of the classifier will be normalized for catering to classification tasks in scenarios of a class imbalance dataset. The experiment shows that SC-Net outperforms all comparative models in four metrics.
{"title":"SC-Net: Network Intrusion Detection with Deep Supervised Contrastive Learning and Normalized Classifier","authors":"Xinjian Zhao, Fei Xia, Guoquan Yuan, Shi Chen, Song Zhang","doi":"10.1109/ICCC56324.2022.10065890","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065890","url":null,"abstract":"Network intrusion detection (NID) has attracted much attention as it is essential in preventing security threats and protecting networks from attacks. However, existing methods face the following challenges: (1) poor feature extraction capability; (2) not well-designed to address the class imbalance problem; (3) failure to take full use of label information and learn classification-oriented features, degrading the NID performance. To this end, we proposed SC-Net, a two-stage training model with deep supervised learning and a normalized classifier, to overcome the abovementioned challenges. During the pretraining stage, the learned embedding will be optimized by both a supervised contrastive loss and a classification loss, so that the embedding with the same label will be more compact in the feature space. After that, in the finetuning stage, the weight of the classifier will be normalized for catering to classification tasks in scenarios of a class imbalance dataset. The experiment shows that SC-Net outperforms all comparative models in four metrics.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132481734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065828
Yuan Zhang
At present, one of the effective ways to deal with the widespread of COVID-19 is to control the source of infection. As the gate of population flow in various places, the security inspection department plays a vital role in screening positive patients in the population; To solve the problems of credibility and lack of human resources, this paper establishes a security check framework based on the blockchain and combines machine learning with the blockchain: the blockchain records the abnormal results of COVID-19 nucleic acid detection and the abnormal conditions detected by the security inspection system (such as no mask, high temperature); Use machine learning to realize mask recognition and other functions. The architecture, data flow, and key elements are presented and discussed. The study findings could solve the security problem under the epidemic and provide relevant enlightenment for the effective combination and application of machine learning and blockchain.
{"title":"Blockchain-Based Security Check Framework in the State of the COVID-19 Epidemic","authors":"Yuan Zhang","doi":"10.1109/ICCC56324.2022.10065828","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065828","url":null,"abstract":"At present, one of the effective ways to deal with the widespread of COVID-19 is to control the source of infection. As the gate of population flow in various places, the security inspection department plays a vital role in screening positive patients in the population; To solve the problems of credibility and lack of human resources, this paper establishes a security check framework based on the blockchain and combines machine learning with the blockchain: the blockchain records the abnormal results of COVID-19 nucleic acid detection and the abnormal conditions detected by the security inspection system (such as no mask, high temperature); Use machine learning to realize mask recognition and other functions. The architecture, data flow, and key elements are presented and discussed. The study findings could solve the security problem under the epidemic and provide relevant enlightenment for the effective combination and application of machine learning and blockchain.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130037572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065764
Bai Yuxuan, Huang Junfei, Lin Zhaowen
In the traditional collaborative filtering recommendation algorithm, the similarity calculation of users is only based on cosine similarity; in the rating prediction link, only the direct neighbors of users are used for prediction. Therefore, under the circumstance that the rating matrix of enterprises on policies is highly sparse, traditional collaborative filtering has the problem that it cannot accurately predict the attitudes of enterprises towards policies and implement policies to corresponding enterprises in a timely manner. This paper proposes an enterprise-oriented policy push algorithm, which incorporates the extreme attitudes and characteristics of enterprises into the similarity calculation process. When the rating matrix is highly sparse and cannot be predicted accurately by relying on direct neighbors, iterative prediction is performed by referring to indirect neighbors and using z-score to eliminate rating bias. The experiments are carried out on the enterprise-policy dataset collected in the article and the film-trust dataset commonly used in recommender systems. The experimental results show that the algorithm reduces the mean absolute error by 5.67% and 1.54% respectively compared with the iterative rating prediction algorithm, which shows that the algorithm has achieved good optimization in the recommendation accuracy.
{"title":"Enterprise-Oriented Policy Push Algorithm","authors":"Bai Yuxuan, Huang Junfei, Lin Zhaowen","doi":"10.1109/ICCC56324.2022.10065764","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065764","url":null,"abstract":"In the traditional collaborative filtering recommendation algorithm, the similarity calculation of users is only based on cosine similarity; in the rating prediction link, only the direct neighbors of users are used for prediction. Therefore, under the circumstance that the rating matrix of enterprises on policies is highly sparse, traditional collaborative filtering has the problem that it cannot accurately predict the attitudes of enterprises towards policies and implement policies to corresponding enterprises in a timely manner. This paper proposes an enterprise-oriented policy push algorithm, which incorporates the extreme attitudes and characteristics of enterprises into the similarity calculation process. When the rating matrix is highly sparse and cannot be predicted accurately by relying on direct neighbors, iterative prediction is performed by referring to indirect neighbors and using z-score to eliminate rating bias. The experiments are carried out on the enterprise-policy dataset collected in the article and the film-trust dataset commonly used in recommender systems. The experimental results show that the algorithm reduces the mean absolute error by 5.67% and 1.54% respectively compared with the iterative rating prediction algorithm, which shows that the algorithm has achieved good optimization in the recommendation accuracy.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130131108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065735
Hua Zhang, Sen Xu, Jincan Xin, S. Xiong
With continuous development of the three driving forces of Artificial Intelligence (AI), e.g., ecomputing power, algorithms and data related technologies, AI is setting off a new round of technological revolution in human society. With the development of AI/ML technology, the wireless communication system is also developing rapidly. This paper summarizes the AI/ML based systems in mobile communication networks, and through researches for network AI of various international standards organizations, the existing AI/ML research purpose and progress of core networks, network management and wireless networks are introduced respectively. Besides, the relevant signaling procedures for one of the typical AI/ML radio network use cases, e.g., AI/ML mobility optimization is introduced and more potential ideas for AI/ML technology to enable wireless networks are provided.
{"title":"Artificial Intelligence in Mobile Communication Network","authors":"Hua Zhang, Sen Xu, Jincan Xin, S. Xiong","doi":"10.1109/ICCC56324.2022.10065735","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065735","url":null,"abstract":"With continuous development of the three driving forces of Artificial Intelligence (AI), e.g., ecomputing power, algorithms and data related technologies, AI is setting off a new round of technological revolution in human society. With the development of AI/ML technology, the wireless communication system is also developing rapidly. This paper summarizes the AI/ML based systems in mobile communication networks, and through researches for network AI of various international standards organizations, the existing AI/ML research purpose and progress of core networks, network management and wireless networks are introduced respectively. Besides, the relevant signaling procedures for one of the typical AI/ML radio network use cases, e.g., AI/ML mobility optimization is introduced and more potential ideas for AI/ML technology to enable wireless networks are provided.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134295254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065602
Zebin Zhang, Bolin Qi, Shimin Ou, Chenjian Shi
Malformed sperm is an important cause of male infertility, and sperm morphology analysis (SMA) is an effective means to diagnose sperm morphology. Deep learning assists to enhance performance on precise SMA; however, existing deep learning based SMA methods mostly focus on single cell scale, which presents a challenge for obtaining single-sperm-image datasets (one sperm cell per image). It is also challenging to integrate current object detection models on low-performance devices. This paper presents a lightweight model for sperm detection. By removing 50% of convolutional kernels cutting the large-object-detecting head from YOLOv5s, our model got a similar precision to the original YOLOv3 (mAP.5 of 0.957 and 0.947, respectively), but with a model size of only 2.8 MB (123.6 MB of YOLOv3). There is a slight loss in precision compared to YOLOv5s (mAP.5:.95 of 0.604 with 14.4MB model size); however, our model still shows a significant advantage in reducing the number of parameters. Experimental results also indicated that MS COCO pre-training is helpful in sperm detection tasks, and the mosaic augmentation strongly enhances the precision for all YOLO models.
{"title":"Real-Time Sperm Detection Using Lightweight YOLOv5","authors":"Zebin Zhang, Bolin Qi, Shimin Ou, Chenjian Shi","doi":"10.1109/ICCC56324.2022.10065602","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065602","url":null,"abstract":"Malformed sperm is an important cause of male infertility, and sperm morphology analysis (SMA) is an effective means to diagnose sperm morphology. Deep learning assists to enhance performance on precise SMA; however, existing deep learning based SMA methods mostly focus on single cell scale, which presents a challenge for obtaining single-sperm-image datasets (one sperm cell per image). It is also challenging to integrate current object detection models on low-performance devices. This paper presents a lightweight model for sperm detection. By removing 50% of convolutional kernels cutting the large-object-detecting head from YOLOv5s, our model got a similar precision to the original YOLOv3 (mAP.5 of 0.957 and 0.947, respectively), but with a model size of only 2.8 MB (123.6 MB of YOLOv3). There is a slight loss in precision compared to YOLOv5s (mAP.5:.95 of 0.604 with 14.4MB model size); however, our model still shows a significant advantage in reducing the number of parameters. Experimental results also indicated that MS COCO pre-training is helpful in sperm detection tasks, and the mosaic augmentation strongly enhances the precision for all YOLO models.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134388001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065641
Peihan Shang, Xuan Zhou, Jinxing Hu
With the development of UAV technology, pedestrian trajectory extraction based on UAV video plays an increasingly prominent role in public safety. Aiming at the problems of small pedestrian targets in UAV video and the effect of pedestrian detection and tracking is easily blocked by obstacles in the scene, this paper analyzes the multi-target tracking algorithm framework based on detection, takes YOLOv5 as the target detection model, and gets a better anchor frame through KMeans++ clustering method. At the same time, CBAM attention module is integrated into YOLOv5 network to realize the effective extraction of small target features; With Deep_SORT is a target tracking model. By calculating the confidence of the trajectory, an improved correlation matching tracking algorithm framework is proposed to solve the problems of trajectory fracture caused by pedestrians being blocked, and provide a strong guarantee for extracting complete and reliable pedestrian trajectory information in UAV video. The mAP of this paper is 49.1 % on the VisDrone2019-DET dataset, and the MOTA is 48.0% on the VisDrone2019-MOT dataset. Experiments show that the pedestrian trajectory extraction method in this paper can extract more stable and continuous pedestrian trajectory.
{"title":"Pedestrian Trajectory Extraction Method Based on UAV Video","authors":"Peihan Shang, Xuan Zhou, Jinxing Hu","doi":"10.1109/ICCC56324.2022.10065641","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065641","url":null,"abstract":"With the development of UAV technology, pedestrian trajectory extraction based on UAV video plays an increasingly prominent role in public safety. Aiming at the problems of small pedestrian targets in UAV video and the effect of pedestrian detection and tracking is easily blocked by obstacles in the scene, this paper analyzes the multi-target tracking algorithm framework based on detection, takes YOLOv5 as the target detection model, and gets a better anchor frame through KMeans++ clustering method. At the same time, CBAM attention module is integrated into YOLOv5 network to realize the effective extraction of small target features; With Deep_SORT is a target tracking model. By calculating the confidence of the trajectory, an improved correlation matching tracking algorithm framework is proposed to solve the problems of trajectory fracture caused by pedestrians being blocked, and provide a strong guarantee for extracting complete and reliable pedestrian trajectory information in UAV video. The mAP of this paper is 49.1 % on the VisDrone2019-DET dataset, and the MOTA is 48.0% on the VisDrone2019-MOT dataset. Experiments show that the pedestrian trajectory extraction method in this paper can extract more stable and continuous pedestrian trajectory.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133020711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-09DOI: 10.1109/ICCC56324.2022.10065844
Chun-An Yang, Hongli Xu, Shixiao Fan, Xuan Cheng, Minghui Liu, Xiaomin Wang
Recently, Mobile Edge Cloud Computing (MECC) emerges as a promising partial offloading paradigm to provide computing services. However, the design of computation resource allocation policies for the MECC network inevitably encounters a challenging delay-sensitive two-queue optimization problem. Specifically, the coupled computation resource allocation of edge processing queue and cloud processing queue makes it difficult to guarantee the end-to-end delay requirements. This study investigates this problem with the stochasticity of computation request arrival, service time, and dynamic computation resources. We first model the MECC network as a two-stage tandem queue that consists of two sequential computation processing queues with multiple servers. A Deep Reinforcement Learning (DRL) algorithm, is then applied to learn a computation speed adjusting policy for the tandem queue, which can provide end-to-end delay insurance for multiple mobile applications while preventing the total computation resources of edge servers and cloud servers from overuse. Finally, extensive simulation results demonstrate that our approach can achieve better performance than others in dynamic network environment.
{"title":"Efficient Resource Allocation Policy for Cloud Edge End Framework by Reinforcement Learning","authors":"Chun-An Yang, Hongli Xu, Shixiao Fan, Xuan Cheng, Minghui Liu, Xiaomin Wang","doi":"10.1109/ICCC56324.2022.10065844","DOIUrl":"https://doi.org/10.1109/ICCC56324.2022.10065844","url":null,"abstract":"Recently, Mobile Edge Cloud Computing (MECC) emerges as a promising partial offloading paradigm to provide computing services. However, the design of computation resource allocation policies for the MECC network inevitably encounters a challenging delay-sensitive two-queue optimization problem. Specifically, the coupled computation resource allocation of edge processing queue and cloud processing queue makes it difficult to guarantee the end-to-end delay requirements. This study investigates this problem with the stochasticity of computation request arrival, service time, and dynamic computation resources. We first model the MECC network as a two-stage tandem queue that consists of two sequential computation processing queues with multiple servers. A Deep Reinforcement Learning (DRL) algorithm, is then applied to learn a computation speed adjusting policy for the tandem queue, which can provide end-to-end delay insurance for multiple mobile applications while preventing the total computation resources of edge servers and cloud servers from overuse. Finally, extensive simulation results demonstrate that our approach can achieve better performance than others in dynamic network environment.","PeriodicalId":263098,"journal":{"name":"2022 IEEE 8th International Conference on Computer and Communications (ICCC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133504661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：