Pub Date : 2022-07-04DOI: 10.1109/spawc51304.2022.9833971
Sai Wang, Yi Gong
Federated edge learning has attracted great attention for edge intelligent networks. Due to the limited computation and energy, mobile devices usually need to offload data to nearby edge servers. Facing this scenario, we design a cloud-aided federated edge learning (CA-FEEL) framework where the edges cooperate with the cloud to train a federated learning model. Specifically, the edges adopt the gradient descent (GD) method in parallel to update the edge parameters and the cloud averages them to update the global parameter. By theoretical analysis, we find that the covariance of non-independent and identically distributed (non-IID) data sets hinders the convergence of the GD based FL. Thus, we propose a CA-FEEL algorithm by adding a simple judgment condition. It is proved to have a theoretical guarantee of convergence for convex and smooth problems. Experiment results indicate that the proposed algorithm outperforms the standard federated learning in terms of the convergence rate and accuracy.
{"title":"Convergence Analysis of Cloud-Aided Federated Edge Learning on Non-IID Data","authors":"Sai Wang, Yi Gong","doi":"10.1109/spawc51304.2022.9833971","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9833971","url":null,"abstract":"Federated edge learning has attracted great attention for edge intelligent networks. Due to the limited computation and energy, mobile devices usually need to offload data to nearby edge servers. Facing this scenario, we design a cloud-aided federated edge learning (CA-FEEL) framework where the edges cooperate with the cloud to train a federated learning model. Specifically, the edges adopt the gradient descent (GD) method in parallel to update the edge parameters and the cloud averages them to update the global parameter. By theoretical analysis, we find that the covariance of non-independent and identically distributed (non-IID) data sets hinders the convergence of the GD based FL. Thus, we propose a CA-FEEL algorithm by adding a simple judgment condition. It is proved to have a theoretical guarantee of convergence for convex and smooth problems. Experiment results indicate that the proposed algorithm outperforms the standard federated learning in terms of the convergence rate and accuracy.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122378033","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9834022
Rami Klaimi, Stefan Weithoffer, C. A. Nour
Non-binary forward error correction (FEC) codes have been getting more attention lately in the coding society thanks mainly to their improved error correcting capabilities. Indeed, they reveal their full potential in the case of a one-to-one mapping between the code symbols over Galois fields (GF) and constellation points of the same order. Previously, we proposed non-binary FEC code designs targeting a given classical constellation through the optimization of the minimum Euclidean distance between candidate codewords. To go a step further, a better Euclidean distance spectrum can be achieved through the joint optimization of code parameters and positions of constellation symbols. However, this joint optimization for high order GFs reveals to be intractable in number of cases to evaluate. Therefore in this work, we propose a solution based on the multi-agent Deep Q-Network (DQN) algorithm. Applied to non-binary turbo codes (NB-TCs) over GF(64), the proposal largely improves performance by significantly lowering the error floor region of the resulting coded modulation scheme.
{"title":"Improved Non-Uniform Constellations for Non-Binary Codes Through Deep Reinforcement Learning","authors":"Rami Klaimi, Stefan Weithoffer, C. A. Nour","doi":"10.1109/spawc51304.2022.9834022","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9834022","url":null,"abstract":"Non-binary forward error correction (FEC) codes have been getting more attention lately in the coding society thanks mainly to their improved error correcting capabilities. Indeed, they reveal their full potential in the case of a one-to-one mapping between the code symbols over Galois fields (GF) and constellation points of the same order. Previously, we proposed non-binary FEC code designs targeting a given classical constellation through the optimization of the minimum Euclidean distance between candidate codewords. To go a step further, a better Euclidean distance spectrum can be achieved through the joint optimization of code parameters and positions of constellation symbols. However, this joint optimization for high order GFs reveals to be intractable in number of cases to evaluate. Therefore in this work, we propose a solution based on the multi-agent Deep Q-Network (DQN) algorithm. Applied to non-binary turbo codes (NB-TCs) over GF(64), the proposal largely improves performance by significantly lowering the error floor region of the resulting coded modulation scheme.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127488100","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9834027
Yacine Benatia, Romain Negrel, Anne Savard, E. Belmega
In this paper, the aim is to study the robustness against imperfect channel state information (CSI) of the power allocation policies maximizing the constrained and non-convex Shannon rate problem in a relay-aided cognitive radio network. The primary communication is protected by a Quality of Service (QoS) constraint and the relay only helps the secondary communication by performing complex and non-linear operations. First, we derive the optimal power allocation policies under Compress-and-Forward (CF) relaying under perfect CSI. Second, we investigate the robustness of this solution jointly with that of the deep learning existing solution for Decode-and-Forward (DF), which we exploit here for CF as well. For all these solutions that strongly rely on perfect CSI, our numerical results show that errors in the channel estimations have a damaging effect not only on the secondary rate, but most importantly on the primary QoS degradation, becoming prohibitive for poor quality estimations. Nevertheless, we show that the deep learning solutions can be made robust by adjusting the training process to rely on both perfect and imperfect CSI observations. Indeed, the resulting predictions are capable of meeting the primary QoS constraint at the cost of secondary rate loss, irrespective from the channel estimation quality.
{"title":"Robustness to imperfect CSI of power allocation policies in cognitive relay networks","authors":"Yacine Benatia, Romain Negrel, Anne Savard, E. Belmega","doi":"10.1109/spawc51304.2022.9834027","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9834027","url":null,"abstract":"In this paper, the aim is to study the robustness against imperfect channel state information (CSI) of the power allocation policies maximizing the constrained and non-convex Shannon rate problem in a relay-aided cognitive radio network. The primary communication is protected by a Quality of Service (QoS) constraint and the relay only helps the secondary communication by performing complex and non-linear operations. First, we derive the optimal power allocation policies under Compress-and-Forward (CF) relaying under perfect CSI. Second, we investigate the robustness of this solution jointly with that of the deep learning existing solution for Decode-and-Forward (DF), which we exploit here for CF as well. For all these solutions that strongly rely on perfect CSI, our numerical results show that errors in the channel estimations have a damaging effect not only on the secondary rate, but most importantly on the primary QoS degradation, becoming prohibitive for poor quality estimations. Nevertheless, we show that the deep learning solutions can be made robust by adjusting the training process to rely on both perfect and imperfect CSI observations. Indeed, the resulting predictions are capable of meeting the primary QoS constraint at the cost of secondary rate loss, irrespective from the channel estimation quality.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"1963 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129679107","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9834028
Hui Zhao, Antonio Bazco-Nogueras, P. Elia
The use of vector coded caching has been shown to provide important gains and, more importantly, to alleviate the impact of the file-size constraint, which prevents coded caching from obtaining its ideal gains in practical settings. In this work, we analyze the performance of vector coded caching in the massive MIMO regime, aiming at understanding the benefits that allowing users to cache a practical amount of data could bring to realistic settings in such massive MIMO regime. In particular, we separately consider two linear precoding schemes and analyze the corresponding throughput, for which we derive simple but precise upper and lower bounds. These bounds enable us to characterize the delivery speed-up gain over the uncoded caching setting when the CSI acquisition costs are taken into account. Numerical results demonstrate the tightness of the derived bounds and show a significant boost over uncoded caching and the standard cacheless setting.
{"title":"Vector Coded Caching Greatly Enhances Massive MIMO","authors":"Hui Zhao, Antonio Bazco-Nogueras, P. Elia","doi":"10.1109/spawc51304.2022.9834028","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9834028","url":null,"abstract":"The use of vector coded caching has been shown to provide important gains and, more importantly, to alleviate the impact of the file-size constraint, which prevents coded caching from obtaining its ideal gains in practical settings. In this work, we analyze the performance of vector coded caching in the massive MIMO regime, aiming at understanding the benefits that allowing users to cache a practical amount of data could bring to realistic settings in such massive MIMO regime. In particular, we separately consider two linear precoding schemes and analyze the corresponding throughput, for which we derive simple but precise upper and lower bounds. These bounds enable us to characterize the delivery speed-up gain over the uncoded caching setting when the CSI acquisition costs are taken into account. Numerical results demonstrate the tightness of the derived bounds and show a significant boost over uncoded caching and the standard cacheless setting.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128730364","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9833957
Xiao Meng, F. Liu, W. Yuan, Qixun Zhang
In this paper, we propose a sensing-assisted beam-forming design for integrated sensing and communication (ISAC) system in a vehicle-to-infrastructure (V2I) network, where a road side unit (RSU) provides localization and communication services to the vehicles on an arbitrarily shaped road. In our proposed scheme, the position and motion of the vehicles are decomposed into longitudinal and lateral directions to simplify the kinematic functions. We establish a curvilinear coordinate system based on the road geometry and employ an extended Kalman filter (EKF) to accurately estimate and predict the state of the vehicles. By employing such prediction, we construct a beamformer directing to the vehicles to acquire high array gain and corresponding high quality of service. Numerical results validate the feasibility of tracking and predicting the state of the vehicles by applying a curvilinear coordinate system. The superiority of the proposed algorithm in both communication and tracking metrics is also verified.
{"title":"Sensing Assisted Predictive Beamforming for V2I Networks: Tracking on the Complicated Road : (Invited Paper)","authors":"Xiao Meng, F. Liu, W. Yuan, Qixun Zhang","doi":"10.1109/spawc51304.2022.9833957","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9833957","url":null,"abstract":"In this paper, we propose a sensing-assisted beam-forming design for integrated sensing and communication (ISAC) system in a vehicle-to-infrastructure (V2I) network, where a road side unit (RSU) provides localization and communication services to the vehicles on an arbitrarily shaped road. In our proposed scheme, the position and motion of the vehicles are decomposed into longitudinal and lateral directions to simplify the kinematic functions. We establish a curvilinear coordinate system based on the road geometry and employ an extended Kalman filter (EKF) to accurately estimate and predict the state of the vehicles. By employing such prediction, we construct a beamformer directing to the vehicles to acquire high array gain and corresponding high quality of service. Numerical results validate the feasibility of tracking and predicting the state of the vehicles by applying a curvilinear coordinate system. The superiority of the proposed algorithm in both communication and tracking metrics is also verified.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122701013","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9834004
Lucas Ribeiro, Markus Leinonen, Isuru Rathnayaka, H. Al-Tous, M. Juntti
Serving a plethora of devices in massive machinetype communications (mMTC) can rely on spatial multiplexing enabled by massive multiple-input multiple-output (mMIMO) technology. To release the full potential, accurate channel estimation is needed. Due to the large numbers of devices it necessitates pilot reuse. We propose a pilot allocation algorithm based on multi-point channel charting (CC) to mitigate inevitable pilot contamination in a multi-cell multi-sector mMTC network with spatially correlated mMIMO channels. The generated CC represents an effective interference map from channel covariance matrices to capture the degree of pilot contamination caused by sharing the same pilot sequence among multiple users. The map is then fed into a greedy algorithm that aims at optimizing the reuse pattern of orthogonal pilot sequences to minimize the performance degradation caused by pilot contamination. The proposed CC-based method is empirically shown to obtain notable gains over a reuse-factor-aware random pilot allocation, yet leaving room for further improvements.
{"title":"Channel Charting Aided Pilot Allocation in Multi-Cell Massive MIMO mMTC Networks","authors":"Lucas Ribeiro, Markus Leinonen, Isuru Rathnayaka, H. Al-Tous, M. Juntti","doi":"10.1109/spawc51304.2022.9834004","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9834004","url":null,"abstract":"Serving a plethora of devices in massive machinetype communications (mMTC) can rely on spatial multiplexing enabled by massive multiple-input multiple-output (mMIMO) technology. To release the full potential, accurate channel estimation is needed. Due to the large numbers of devices it necessitates pilot reuse. We propose a pilot allocation algorithm based on multi-point channel charting (CC) to mitigate inevitable pilot contamination in a multi-cell multi-sector mMTC network with spatially correlated mMIMO channels. The generated CC represents an effective interference map from channel covariance matrices to capture the degree of pilot contamination caused by sharing the same pilot sequence among multiple users. The map is then fed into a greedy algorithm that aims at optimizing the reuse pattern of orthogonal pilot sequences to minimize the performance degradation caused by pilot contamination. The proposed CC-based method is empirically shown to obtain notable gains over a reuse-factor-aware random pilot allocation, yet leaving room for further improvements.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"75 2-3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123564340","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9833978
Jipeng Gan, Jun Wu, Pei Li, Zehao Chen, Zehao Chen, Jia Zhang, Jian-Duo He
Cooperative spectrum sensing (CSS) is crucial for cognitive radio (CR) to improve spectrum sensing performance. However, the cooperative paradigm is threatened by Byzantine attacks. To ensure the security and energy efficiency (EE) of CSS, in this paper, we propose a malicious exploitation algorithm. Firstly, we distinguish normal users (NUs) from malicious users (MUs) based on the historical performance of secondary users (SUs). Unlike most previous studies, we innovatively improve CSS detection performance by exploiting sensing information from MUs. In addition, we select specific SUs instead of all SUs in data fusion, which reduces the number of samples submitted by SUs to the fusion center (FC). Finally, we further introduce a sequential differential mechanism that substantially reduces samples to improve the EE of CSS. Finally, the numerical simulation results validate the effectiveness of our proposed algorithm.
{"title":"Malicious Exploitation of Byzantine Attack for Cooperative Spectrum Sensing","authors":"Jipeng Gan, Jun Wu, Pei Li, Zehao Chen, Zehao Chen, Jia Zhang, Jian-Duo He","doi":"10.1109/spawc51304.2022.9833978","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9833978","url":null,"abstract":"Cooperative spectrum sensing (CSS) is crucial for cognitive radio (CR) to improve spectrum sensing performance. However, the cooperative paradigm is threatened by Byzantine attacks. To ensure the security and energy efficiency (EE) of CSS, in this paper, we propose a malicious exploitation algorithm. Firstly, we distinguish normal users (NUs) from malicious users (MUs) based on the historical performance of secondary users (SUs). Unlike most previous studies, we innovatively improve CSS detection performance by exploiting sensing information from MUs. In addition, we select specific SUs instead of all SUs in data fusion, which reduces the number of samples submitted by SUs to the fusion center (FC). Finally, we further introduce a sequential differential mechanism that substantially reduces samples to improve the EE of CSS. Finally, the numerical simulation results validate the effectiveness of our proposed algorithm.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123017658","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9833965
Yuyang Wang, Shiva R. Iyer, D. Chizhik, Jinfeng Du, R. Valenzuela
Channel modeling is critical for coverage prediction, system level simulations, and wireless propagation characterization. Industry practice applies linear fit to the pathloss in decibels against the logarithm of the distance. Simple linear fit, however, cannot fully capture the shadowing effects in the channel, especially for a link with rich scatterings such as non-line-of-sight (NLOS) links in a complex propagation environment. In this paper, we propose an interpretable hybrid learning model with expert knowledge to predict the channel pathloss in desert-like environment using terrain profiles. We apply an autoencoder to extract compressed information from terrain profiles. The compressed representation of terrain, combined with features selected based on expert knowledge such as LOS/NLOS indicator and curvature of the terrain, are used to predict the pathloss. We show that a Random Forest regression model outperforms CNN/DNN models in generalizability of predicting unseen data by training and testing in disjoint sectors of the measured areas.
{"title":"Channel Prediction over Irregular Terrains: Deep Autoencoder with Random Forest","authors":"Yuyang Wang, Shiva R. Iyer, D. Chizhik, Jinfeng Du, R. Valenzuela","doi":"10.1109/spawc51304.2022.9833965","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9833965","url":null,"abstract":"Channel modeling is critical for coverage prediction, system level simulations, and wireless propagation characterization. Industry practice applies linear fit to the pathloss in decibels against the logarithm of the distance. Simple linear fit, however, cannot fully capture the shadowing effects in the channel, especially for a link with rich scatterings such as non-line-of-sight (NLOS) links in a complex propagation environment. In this paper, we propose an interpretable hybrid learning model with expert knowledge to predict the channel pathloss in desert-like environment using terrain profiles. We apply an autoencoder to extract compressed information from terrain profiles. The compressed representation of terrain, combined with features selected based on expert knowledge such as LOS/NLOS indicator and curvature of the terrain, are used to predict the pathloss. We show that a Random Forest regression model outperforms CNN/DNN models in generalizability of predicting unseen data by training and testing in disjoint sectors of the measured areas.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121437833","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9833926
B. Manoj, P. M. Santos, Meysam Sadeghi, E. Larsson
Deep learning (DL) is a powerful technique for many real-time applications, but it is vulnerable to adversarial attacks. Herein, we consider DL-based modulation classification, with the objective to create DL models that are robust against attacks. Specifically, we introduce three defense techniques: i) randomized smoothing, ii) hybrid projected gradient descent adversarial training, and iii) fast adversarial training, and evaluate them under both white-box (WB) and black-box (BB) attacks. We show that the proposed fast adversarial training is more robust and computationally efficient than the other techniques, and can create models that are extremely robust to practical (BB) attacks.
{"title":"Toward Robust Networks against Adversarial Attacks for Radio Signal Modulation Classification","authors":"B. Manoj, P. M. Santos, Meysam Sadeghi, E. Larsson","doi":"10.1109/spawc51304.2022.9833926","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9833926","url":null,"abstract":"Deep learning (DL) is a powerful technique for many real-time applications, but it is vulnerable to adversarial attacks. Herein, we consider DL-based modulation classification, with the objective to create DL models that are robust against attacks. Specifically, we introduce three defense techniques: i) randomized smoothing, ii) hybrid projected gradient descent adversarial training, and iii) fast adversarial training, and evaluate them under both white-box (WB) and black-box (BB) attacks. We show that the proposed fast adversarial training is more robust and computationally efficient than the other techniques, and can create models that are extremely robust to practical (BB) attacks.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"289 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115218843","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 : 2022-07-04DOI: 10.1109/spawc51304.2022.9834021
Jaakko Pihlajasalo, D. Korpi, T. Riihonen, J. Talvitie, M. Uusitalo, M. Valkama
With wireless networks evolving towards mmWave and sub-THz frequency bands, hardware impairments such as IQ imbalance, phase noise (PN) and power amplifier (PA) nonlinear distortion are increasingly critical implementation challenges. In this paper, we describe deep learning based physical-layer receiver solution, with neural network layers in both time- and frequency-domain, to efficiently demodulate OFDM signals under coexisting IQ, PN and PA impairments. 5G NR standard-compliant numerical results are provided at 28 GHz band to assess the receiver performance, demonstrating excellent robustness against varying impairment levels when properly trained.
{"title":"Detection of Impaired OFDM Waveforms Using Deep Learning Receiver","authors":"Jaakko Pihlajasalo, D. Korpi, T. Riihonen, J. Talvitie, M. Uusitalo, M. Valkama","doi":"10.1109/spawc51304.2022.9834021","DOIUrl":"https://doi.org/10.1109/spawc51304.2022.9834021","url":null,"abstract":"With wireless networks evolving towards mmWave and sub-THz frequency bands, hardware impairments such as IQ imbalance, phase noise (PN) and power amplifier (PA) nonlinear distortion are increasingly critical implementation challenges. In this paper, we describe deep learning based physical-layer receiver solution, with neural network layers in both time- and frequency-domain, to efficiently demodulate OFDM signals under coexisting IQ, PN and PA impairments. 5G NR standard-compliant numerical results are provided at 28 GHz band to assess the receiver performance, demonstrating excellent robustness against varying impairment levels when properly trained.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132786840","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：