Pub Date : 2019-05-09DOI: 10.1109/WOCC.2019.8770602
Saad Chaudhry, Xiaoya Zuo, Rugui Yao, Hailong Liu, T. Mekkawy
The recent emergence of spatial modulation (SM) as a new modulation concept maneuvers the random and unique properties of the multiple-input multiple-output (MIMO) wireless channels. In this paper, an enhancement in SM scheme is proposed by innovative usage of quadrature amplitude modulation (QAM) to add layers which fully exploit both the signal and spatial domains. Layered spatial modulation (LSM) controls the preset rotated signals with normalized amplitudes to classify the layers and to identify the transmitted symbols at the transceiver by completely avoiding inter-channel interference (ICI). High spectral efficiency can be achieved with the addition of only one transmit antenna for each layer by the LSM scheme. Simulation results show that the proposed method has significant improvement in spectral efficiency and bit-error-rate (BER) performance as compared to other MIMO schemes.
{"title":"Enhancing Spectral Efficiency and BER Performance of Spatial Modulation using Layered Architecture","authors":"Saad Chaudhry, Xiaoya Zuo, Rugui Yao, Hailong Liu, T. Mekkawy","doi":"10.1109/WOCC.2019.8770602","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770602","url":null,"abstract":"The recent emergence of spatial modulation (SM) as a new modulation concept maneuvers the random and unique properties of the multiple-input multiple-output (MIMO) wireless channels. In this paper, an enhancement in SM scheme is proposed by innovative usage of quadrature amplitude modulation (QAM) to add layers which fully exploit both the signal and spatial domains. Layered spatial modulation (LSM) controls the preset rotated signals with normalized amplitudes to classify the layers and to identify the transmitted symbols at the transceiver by completely avoiding inter-channel interference (ICI). High spectral efficiency can be achieved with the addition of only one transmit antenna for each layer by the LSM scheme. Simulation results show that the proposed method has significant improvement in spectral efficiency and bit-error-rate (BER) performance as compared to other MIMO schemes.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125158177","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770560
Xifeng Chen, Jupeng Ding, Huicheng Lai
With increasing of high speed train (HST) in all over the world, users' requirements for communications are constantly improving in HST scenario. Although the fifth generation (5G) communication systems provided high rate data services in HST, yet it restricted for the signal bandwidth and spectrum license. In order to enhance the HST communication performances, optical communication technologies are applied in HST. Hence, in this work, we focus on the optical wave as transmission carrier in HST communication. In communication process, experienced frequent handover, limited coverage of base station and changed of environment factors are most challenges. For overcoming those obstacles, several optical communication systems are proposed in recent year. Several researchers are proposed optical wireless communication (OWC) systems which are optimized the communications barriers in HST. Beside, the optical fiber communication (OFC) systems are used to strengthen the capacity of HST communication. Moreover, plentiful experiments are established for studied the optical communication performances, which are set up consisted of car is replaced with train, simulated HST scenario in lab and tested in reality HST scenario.
{"title":"5G Oriented Optical Communications in Highspeed Trains: A Review","authors":"Xifeng Chen, Jupeng Ding, Huicheng Lai","doi":"10.1109/WOCC.2019.8770560","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770560","url":null,"abstract":"With increasing of high speed train (HST) in all over the world, users' requirements for communications are constantly improving in HST scenario. Although the fifth generation (5G) communication systems provided high rate data services in HST, yet it restricted for the signal bandwidth and spectrum license. In order to enhance the HST communication performances, optical communication technologies are applied in HST. Hence, in this work, we focus on the optical wave as transmission carrier in HST communication. In communication process, experienced frequent handover, limited coverage of base station and changed of environment factors are most challenges. For overcoming those obstacles, several optical communication systems are proposed in recent year. Several researchers are proposed optical wireless communication (OWC) systems which are optimized the communications barriers in HST. Beside, the optical fiber communication (OFC) systems are used to strengthen the capacity of HST communication. Moreover, plentiful experiments are established for studied the optical communication performances, which are set up consisted of car is replaced with train, simulated HST scenario in lab and tested in reality HST scenario.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130359757","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770668
Mi Yu, Xiaoming Wang, Youyun Xu, Dapeng Li, Jianping Chen
To meet the increasing demands of high data-rate service in high-speed railway (HSR) scenarios, the two-hop architecture with distributed antenna system (DAS) is a good choice. In this architecture, the transmit and receive sides of HSR system are both equipped with distributed antennas, then base stations (BSs) communicate with users via mobile relays (MRs) on the top of train. Thus, the data will be buffered when the wireless transmission rate can not satisfy the requirement of data arrival rate. In this paper, we focus on the efficient power allocation method with buffer and delay constraint on the basis of predicted path loss characteristic in downlink HSR communications. Firstly, the relationship between the data arrival rate and instantaneous wireless transmission rate is analyzed where BSs transmit with constant power. Secondly, the transmit power optimization problem with the requirement of buffer size and delay is formulated. The optimal power allocation scheme can be obtained through Lagrangian dual method. Simulation results demonstrate that the proposed power allocation scheme can achieve a lower consumed power and avoid data overflow and infinite delay as well.
{"title":"Power Allocation with Buffer Constraint for Distributed Antenna System in High-Speed Railway Scenarios","authors":"Mi Yu, Xiaoming Wang, Youyun Xu, Dapeng Li, Jianping Chen","doi":"10.1109/WOCC.2019.8770668","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770668","url":null,"abstract":"To meet the increasing demands of high data-rate service in high-speed railway (HSR) scenarios, the two-hop architecture with distributed antenna system (DAS) is a good choice. In this architecture, the transmit and receive sides of HSR system are both equipped with distributed antennas, then base stations (BSs) communicate with users via mobile relays (MRs) on the top of train. Thus, the data will be buffered when the wireless transmission rate can not satisfy the requirement of data arrival rate. In this paper, we focus on the efficient power allocation method with buffer and delay constraint on the basis of predicted path loss characteristic in downlink HSR communications. Firstly, the relationship between the data arrival rate and instantaneous wireless transmission rate is analyzed where BSs transmit with constant power. Secondly, the transmit power optimization problem with the requirement of buffer size and delay is formulated. The optimal power allocation scheme can be obtained through Lagrangian dual method. Simulation results demonstrate that the proposed power allocation scheme can achieve a lower consumed power and avoid data overflow and infinite delay as well.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121525381","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770586
Liping Du, Long Chen, Xiaotian Hou, Yueyun Chen
In this paper, we proposed an Extended Kalman filter (EKF) method for multi-vehicle cooperative localization using Global Positioning System (GPS) data and inter-vehicle position information. Each cooperative vehicle uses its own GPS receiver to estimate its position. And inter-vehicle position information is obtained by the Dedicated Short-range Communication (DSRC). This proposed method includes two processes. Firstly, the GPS positioning information of cooperative vehicles are collected to get the positioning matrix. Then the EKF is applied to the matrix to further improve the positioning accuracy. In the simulation, we analyze the impact of different numbers of neighbor vehicles on positioning accuracy and the performance of the proposed method has been verified.
{"title":"Cooperative Vehicle Localization Base on Extended Kalman Filter In Intelligent Transportation System","authors":"Liping Du, Long Chen, Xiaotian Hou, Yueyun Chen","doi":"10.1109/WOCC.2019.8770586","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770586","url":null,"abstract":"In this paper, we proposed an Extended Kalman filter (EKF) method for multi-vehicle cooperative localization using Global Positioning System (GPS) data and inter-vehicle position information. Each cooperative vehicle uses its own GPS receiver to estimate its position. And inter-vehicle position information is obtained by the Dedicated Short-range Communication (DSRC). This proposed method includes two processes. Firstly, the GPS positioning information of cooperative vehicles are collected to get the positioning matrix. Then the EKF is applied to the matrix to further improve the positioning accuracy. In the simulation, we analyze the impact of different numbers of neighbor vehicles on positioning accuracy and the performance of the proposed method has been verified.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116971770","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770579
Xinyi Wang, Ce Sun, Jingxuan Huang, Dai Jia, Yifan Jiang, Deliang Song
As the first family of codes to achieve the channel capacity, polar codes are becoming popular in many applications, such as 5G communication. Motivated by the requirement of low complexity and latency in 5G, researchers are keen to study efficient construction of polar codes. In this paper, inspired by the idea of relaxed polar codes, we study the relaxation of polar codes, and propose an efficient construction method for relaxed polar codes. By relaxing part of polarization units, the proposed method can help polar codes significantly reduce the encoding and decoding complexity as well as decoding latency with no BLER performance loss. Meanwhile, the proposed relaxed polar codes shows better BER performance due to the reduction of the impact of error propagation.
{"title":"Relaxed Polar Codes under AWGN Channels with Low Complexity and Low Latency","authors":"Xinyi Wang, Ce Sun, Jingxuan Huang, Dai Jia, Yifan Jiang, Deliang Song","doi":"10.1109/WOCC.2019.8770579","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770579","url":null,"abstract":"As the first family of codes to achieve the channel capacity, polar codes are becoming popular in many applications, such as 5G communication. Motivated by the requirement of low complexity and latency in 5G, researchers are keen to study efficient construction of polar codes. In this paper, inspired by the idea of relaxed polar codes, we study the relaxation of polar codes, and propose an efficient construction method for relaxed polar codes. By relaxing part of polarization units, the proposed method can help polar codes significantly reduce the encoding and decoding complexity as well as decoding latency with no BLER performance loss. Meanwhile, the proposed relaxed polar codes shows better BER performance due to the reduction of the impact of error propagation.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117011753","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770520
Ye Xiao, Yi-jun Zhu, Dong-Fang Zhang, Hai-yong Zhang
This paper proposes a visible light communication (VLC) chipset based multi-color multiple-input multiple-output (MIMO) VLC system. The high-speed wideband VLC chipset we employed is independently researched and developed, it could afford higher than Gbps data rate transmission. With the utilization of multi-color LEDs (e.g., RGBY-LEDs), the data rate of VLC systems is further improved. In this paper we introduce the multi-color communication module and the architecture of our proposed multi-color MIMO system model. With the help of self-designed shape π resistance network and bridged-T based pre-equalizer, the bandwidth of each LED color chip is broaden. Finally, an experimental chipset based multi-color MIMO-VLC setup is constructed and tested. The proposed experimental system offers the aggregate physical layer (PHY) data rates up to 1 Gbps.
{"title":"High-Speed Visible Light Communication Chipset Based Multi-Color MIMO System","authors":"Ye Xiao, Yi-jun Zhu, Dong-Fang Zhang, Hai-yong Zhang","doi":"10.1109/WOCC.2019.8770520","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770520","url":null,"abstract":"This paper proposes a visible light communication (VLC) chipset based multi-color multiple-input multiple-output (MIMO) VLC system. The high-speed wideband VLC chipset we employed is independently researched and developed, it could afford higher than Gbps data rate transmission. With the utilization of multi-color LEDs (e.g., RGBY-LEDs), the data rate of VLC systems is further improved. In this paper we introduce the multi-color communication module and the architecture of our proposed multi-color MIMO system model. With the help of self-designed shape π resistance network and bridged-T based pre-equalizer, the bandwidth of each LED color chip is broaden. Finally, an experimental chipset based multi-color MIMO-VLC setup is constructed and tested. The proposed experimental system offers the aggregate physical layer (PHY) data rates up to 1 Gbps.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128377703","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770609
Aghmaz Ul Haq, Kai Liu, Muhammad Bilal Latif
Contention-based IEEE 802.11p medium access control (MAC) protocol for vehicular ad-hoc networks (VANETs) has proven to be inefficient because it cannot ensure delivery of safety messages with a bounded delay due to and merge packet collisions, hidden terminal problem, inefficient resource utilization and unreliability in different traffic scenarios. To solve these problems, a conflict-free clustering-based MAC protocol is proposed in this paper which employs TDMA-based deterministic slot assignment to resolves access collisions and to ensure bounded delay and centralized access mechanism for efficient resource management. Mitigation of merge collisions is proposed through disjoint slot assignment and channel distribution with respect to direction of node mobility, whereas, location and mobility awareness of nodes helps in cluster formation and resolve packet conflicts due to hidden terminals. Our simulation results show that the proposed protocol performs better than IEEE 802.11p MAC and TCMAC protocols in both urban and highway traffic scenarios.
{"title":"A Location- and Mobility-Aware Clustering-Based TDMA MAC Protocol for Vehicular Ad-hoc Networks","authors":"Aghmaz Ul Haq, Kai Liu, Muhammad Bilal Latif","doi":"10.1109/WOCC.2019.8770609","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770609","url":null,"abstract":"Contention-based IEEE 802.11p medium access control (MAC) protocol for vehicular ad-hoc networks (VANETs) has proven to be inefficient because it cannot ensure delivery of safety messages with a bounded delay due to and merge packet collisions, hidden terminal problem, inefficient resource utilization and unreliability in different traffic scenarios. To solve these problems, a conflict-free clustering-based MAC protocol is proposed in this paper which employs TDMA-based deterministic slot assignment to resolves access collisions and to ensure bounded delay and centralized access mechanism for efficient resource management. Mitigation of merge collisions is proposed through disjoint slot assignment and channel distribution with respect to direction of node mobility, whereas, location and mobility awareness of nodes helps in cluster formation and resolve packet conflicts due to hidden terminals. Our simulation results show that the proposed protocol performs better than IEEE 802.11p MAC and TCMAC protocols in both urban and highway traffic scenarios.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122379127","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770550
Xiaofei Liu, Zhenyu Zhang, Xiyuan Wang, Jing Lian, Xiaoming Dai
In massive multiple-input multiple-output (MIMO) system, Neumann series (NS) expansion-based linear minimum mean square error (LMMSE) detection has been proposed due to its simple and efficient multi-stage pipeline hardware implementation. However, it suffers from poor performance and slow convergence as the number of the users grows. To address this issue, we proposed a novel weighted Neumann series (WNS)-based LMMSE detection to minimize the error between the exact matrix inversion and the WNS-based matrix inversion. Moreover, the optimal weights are obtained according to on-line learning basis. Numerical results indicate that the learning-based WNS detection outperforms the conventional NS-based detection and achieves near-LMMSE performance with a significantly lower computational complexity.
{"title":"A Low Complexity High Performance Weighted Neumann Series-based Massive MIMO Detection","authors":"Xiaofei Liu, Zhenyu Zhang, Xiyuan Wang, Jing Lian, Xiaoming Dai","doi":"10.1109/WOCC.2019.8770550","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770550","url":null,"abstract":"In massive multiple-input multiple-output (MIMO) system, Neumann series (NS) expansion-based linear minimum mean square error (LMMSE) detection has been proposed due to its simple and efficient multi-stage pipeline hardware implementation. However, it suffers from poor performance and slow convergence as the number of the users grows. To address this issue, we proposed a novel weighted Neumann series (WNS)-based LMMSE detection to minimize the error between the exact matrix inversion and the WNS-based matrix inversion. Moreover, the optimal weights are obtained according to on-line learning basis. Numerical results indicate that the learning-based WNS detection outperforms the conventional NS-based detection and achieves near-LMMSE performance with a significantly lower computational complexity.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132680344","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 : 2019-05-09DOI: 10.1109/WOCC.2019.8770700
Khalid Alshathri, Hongtao Xia, V. Lawrence, Yu-dong Yao
Deep learning (DL) is an effective tool in artificial intelligence (AI), especially in image based and human behavior recognition applications. However, there are many applications that are not very well explored using the DL tools. The telecommunications and networking applications are among those applications that can be explored more extensively using DL. In this paper, the neural network is utilized to identify different cellular communications signals including GSM, UMTS, and LTE. Our study results show that the cellular system identification method achieves very good identification performance without any necessity to select signal features manually.
{"title":"Cellular System Identification Using Deep Learning: GSM, UMTS and LTE","authors":"Khalid Alshathri, Hongtao Xia, V. Lawrence, Yu-dong Yao","doi":"10.1109/WOCC.2019.8770700","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770700","url":null,"abstract":"Deep learning (DL) is an effective tool in artificial intelligence (AI), especially in image based and human behavior recognition applications. However, there are many applications that are not very well explored using the DL tools. The telecommunications and networking applications are among those applications that can be explored more extensively using DL. In this paper, the neural network is utilized to identify different cellular communications signals including GSM, UMTS, and LTE. Our study results show that the cellular system identification method achieves very good identification performance without any necessity to select signal features manually.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131012422","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}
This paper proposes a human action recognition (HAR) algorithm based on convolutional neural network, which is used for human semaphore motion recognition. First, collecting datas in three scenarios and Deep Convolution Generative Adversarial Networks(DCGAN) is used to implement data enhancement to generate the dataset (DataSR). Then, the 1*1 and 3*3 convolution kernels are used to design the full convolution network and the model is further compressed using the group convolution to obtain the new model HARNET. Experiments show that the mAP of HARNET on the DataSR dataset is 94.36%, and the model size is 76M, which is 30% of the size of the YOLOv3 model.
{"title":"Research on Human Action Recognition Based on Convolutional Neural Network","authors":"Peng Wang, Yuliang Yang, Wanchong Li, Linhao Zhang, Mengyuan Wang, Xiaobo Zhang, Mengyu Zhu","doi":"10.1109/WOCC.2019.8770575","DOIUrl":"https://doi.org/10.1109/WOCC.2019.8770575","url":null,"abstract":"This paper proposes a human action recognition (HAR) algorithm based on convolutional neural network, which is used for human semaphore motion recognition. First, collecting datas in three scenarios and Deep Convolution Generative Adversarial Networks(DCGAN) is used to implement data enhancement to generate the dataset (DataSR). Then, the 1*1 and 3*3 convolution kernels are used to design the full convolution network and the model is further compressed using the group convolution to obtain the new model HARNET. Experiments show that the mAP of HARNET on the DataSR dataset is 94.36%, and the model size is 76M, which is 30% of the size of the YOLOv3 model.","PeriodicalId":285172,"journal":{"name":"2019 28th Wireless and Optical Communications Conference (WOCC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123821901","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：