. Energy is seen as a significant factor in wireless sensor networks (WSNs). It is a challenge to balance be-tween battery lifetime of the different sensors and network lifetime. The main contribution of the proposed approach is to decrease the energy consumption of each sensor node, overcome unbalanced energy usage among sensor nodes, reduce the data gathering time and enhance the network life-time. To achieve these goals, we combine the Hierarchical Agglomerative algorithm and an optimal path selection method. First, the suitable cluster heads (CHs) are elected based on the Euclidean distance and the residual energy of each sensor node. Then, the base station is situated at the center of the field, which will be partitioned into equal sub-areas, one for every mobile data collector (MDC). Second, the Kruskal algorithm is used to create an optimal data gathering path from each subset of elected cluster heads. Finally, each mobile data collector travels the optimal path to collect the data from the set of cluster heads of each subarea and returns periodically to the base station to upload gathered data. Computer simulation proves that the proposed approach outperforms existing ones in terms of data gathering time, residual energy and network lifetime.
{"title":"Energy-Efficient Path Construction for Data Gathering Using Mobile Data Collectors in Wireless Sensor Networks","authors":"W. Jlassi, R. Haddad, R. Bouallegue","doi":"10.13164/re.2023.0502","DOIUrl":"https://doi.org/10.13164/re.2023.0502","url":null,"abstract":". Energy is seen as a significant factor in wireless sensor networks (WSNs). It is a challenge to balance be-tween battery lifetime of the different sensors and network lifetime. The main contribution of the proposed approach is to decrease the energy consumption of each sensor node, overcome unbalanced energy usage among sensor nodes, reduce the data gathering time and enhance the network life-time. To achieve these goals, we combine the Hierarchical Agglomerative algorithm and an optimal path selection method. First, the suitable cluster heads (CHs) are elected based on the Euclidean distance and the residual energy of each sensor node. Then, the base station is situated at the center of the field, which will be partitioned into equal sub-areas, one for every mobile data collector (MDC). Second, the Kruskal algorithm is used to create an optimal data gathering path from each subset of elected cluster heads. Finally, each mobile data collector travels the optimal path to collect the data from the set of cluster heads of each subarea and returns periodically to the base station to upload gathered data. Computer simulation proves that the proposed approach outperforms existing ones in terms of data gathering time, residual energy and network lifetime.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138618807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. To address the challenges of increasing complexity and larger number of training samples required for high-accuracy spectrum prediction, we propose a novel lightweight model, leveraging a temporal convolutional network (TCN) and knowledge distillation. First, the prediction accuracy of TCN is enhanced via a self-transfer method. Then, we design a two-branch network which can extract the spectrum features efficiently. By employing knowledge distillation, we transfer the knowledge from TCN to the two-branch network, resulting in improved accuracy for spectrum prediction of the lightweight network. Experimental results show that the proposed model can improve accuracy by 19.5% compared to the widely-used LSTM model with sufficient historical data and reduces 71.1% parameters to be trained. Furthermore, the prediction accuracy is improved by 17.9% compared to Gated Recurrent Units (GRU) in the scenarios with scarce historical data.
{"title":"Lightweight Spectrum Prediction Based on Knowledge Distillation","authors":"R. Cheng, J. Zhang, J. Deng, Y. Zhu","doi":"10.13164/re.2023.0469","DOIUrl":"https://doi.org/10.13164/re.2023.0469","url":null,"abstract":". To address the challenges of increasing complexity and larger number of training samples required for high-accuracy spectrum prediction, we propose a novel lightweight model, leveraging a temporal convolutional network (TCN) and knowledge distillation. First, the prediction accuracy of TCN is enhanced via a self-transfer method. Then, we design a two-branch network which can extract the spectrum features efficiently. By employing knowledge distillation, we transfer the knowledge from TCN to the two-branch network, resulting in improved accuracy for spectrum prediction of the lightweight network. Experimental results show that the proposed model can improve accuracy by 19.5% compared to the widely-used LSTM model with sufficient historical data and reduces 71.1% parameters to be trained. Furthermore, the prediction accuracy is improved by 17.9% compared to Gated Recurrent Units (GRU) in the scenarios with scarce historical data.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138622936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. Coverless image steganography typically extracts feature sequences from cover images to map information. Once the extracted features have high similarity, it is challenging to construct a complete mapping sequence set, which places a heavy burden on the underlying storage and computation. In order to improve database utilization while increasing the data-hiding capacity, we propose a coverless steganography model based on low-similarity feature selection in the DCT domain. A mapping algorithm is presented based on an 8000-dimensional feature termed CS-DCTR extracted from each image to convert into binary sequences. The high feature dimension leads to a high capacity, ranging from 8 to 25 bits per image. Furthermore, scrambling is employed for feature mapping before building an inverted index tree, considerably enhancing security against steganal-ysis. Experimental results show that CS-DCTR features exhibit high diversity, averaging 49.3% complete mapping sequences, which indicates lower similarity among CS-DCTR features. The technique also demonstrates resistance to normal operations and benign attacks. The information extraction accuracy rises to 96.7% on average under typical noise attacks. Moreover, our technique achieves excellent performance in terms of hiding capacity, image utilization, and transmission security.
{"title":"Coverless Steganography Based on Low Similarity Feature Selection in DCT Domain","authors":"L. Tan, J. Liu, Y. Zhou, R. Chen","doi":"10.13164/re.2023.0603","DOIUrl":"https://doi.org/10.13164/re.2023.0603","url":null,"abstract":". Coverless image steganography typically extracts feature sequences from cover images to map information. Once the extracted features have high similarity, it is challenging to construct a complete mapping sequence set, which places a heavy burden on the underlying storage and computation. In order to improve database utilization while increasing the data-hiding capacity, we propose a coverless steganography model based on low-similarity feature selection in the DCT domain. A mapping algorithm is presented based on an 8000-dimensional feature termed CS-DCTR extracted from each image to convert into binary sequences. The high feature dimension leads to a high capacity, ranging from 8 to 25 bits per image. Furthermore, scrambling is employed for feature mapping before building an inverted index tree, considerably enhancing security against steganal-ysis. Experimental results show that CS-DCTR features exhibit high diversity, averaging 49.3% complete mapping sequences, which indicates lower similarity among CS-DCTR features. The technique also demonstrates resistance to normal operations and benign attacks. The information extraction accuracy rises to 96.7% on average under typical noise attacks. Moreover, our technique achieves excellent performance in terms of hiding capacity, image utilization, and transmission security.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138607916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. This paper investigates a dual-hop satellite-marine communication network that employs mixed radio-frequency/underwater wireless optical communication (RF/UWOC). The study focuses on investigating the impacts of non-zero pointing errors and the additive white generalized Gaussian noise (AWGGN) on the dual-hop system. To address the challenge of computing the probability density function (PDF) for the UWOC system with non-zero bore-sight error, we apply the Laplace transformation and the generalized integro exponential function. Next, we utilize the generalized Gaussian noise to calculate the signal-to-noise ratio (SNR) and the conditional bit error rate (BER). Then, we present system performance metrics such as the outage probability (OP) and BER. We also calculate the asymptotic analysis of the OP and BER by considering poles coinciding, resulting in the proposal of four asymptotic formulas to gain additional insights into the diversity gain. Finally, we provide simulation results that analyze the performance of the proposed satellite-marine network with different system parameters, such as boresight displacements and bubble levels, and validate the accuracy of the numerical results.
{"title":"Performance of Satellite UWOC Network with Generalized Boresight Error and AWGGN","authors":"Tao Teng, HE Ansu","doi":"10.13164/re.2023.0616","DOIUrl":"https://doi.org/10.13164/re.2023.0616","url":null,"abstract":". This paper investigates a dual-hop satellite-marine communication network that employs mixed radio-frequency/underwater wireless optical communication (RF/UWOC). The study focuses on investigating the impacts of non-zero pointing errors and the additive white generalized Gaussian noise (AWGGN) on the dual-hop system. To address the challenge of computing the probability density function (PDF) for the UWOC system with non-zero bore-sight error, we apply the Laplace transformation and the generalized integro exponential function. Next, we utilize the generalized Gaussian noise to calculate the signal-to-noise ratio (SNR) and the conditional bit error rate (BER). Then, we present system performance metrics such as the outage probability (OP) and BER. We also calculate the asymptotic analysis of the OP and BER by considering poles coinciding, resulting in the proposal of four asymptotic formulas to gain additional insights into the diversity gain. Finally, we provide simulation results that analyze the performance of the proposed satellite-marine network with different system parameters, such as boresight displacements and bubble levels, and validate the accuracy of the numerical results.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138616043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. Wireless Federated Learning (WFL) is an innovative machine learning paradigm enabling distributed devices to collaboratively learn without sharing raw data. WFL is particularly useful for mobile devices that generate massive amounts of data but have limited resources for training complex models. This paper highlights the significance of reducing delay for efficient WFL implementation through advanced multiple access protocols and joint optimization of communication and computing resources. We propose optimizing the WFL Compute-then-Transmit (CT) protocol using hybrid Non-Orthogonal Multiple Access (H-NOMA). To minimize and optimize latency for the transmission of local training data, we use the Successive Convex Optimization (SCA) method, which efficiently reduces the complexity of non-convex algorithms. Finally, the numerical results verify the effectiveness of H-NOMA in terms of delay reduction, compared to the benchmark that is based on Non-Orthogonal Multiple Acces (NOMA).
{"title":"Hybrid NOMA for Latency Minimization in Wireless Federated Learning for 6G Networks","authors":"P. Kavitha, K. Kavitha","doi":"10.13164/re.2023.0594","DOIUrl":"https://doi.org/10.13164/re.2023.0594","url":null,"abstract":". Wireless Federated Learning (WFL) is an innovative machine learning paradigm enabling distributed devices to collaboratively learn without sharing raw data. WFL is particularly useful for mobile devices that generate massive amounts of data but have limited resources for training complex models. This paper highlights the significance of reducing delay for efficient WFL implementation through advanced multiple access protocols and joint optimization of communication and computing resources. We propose optimizing the WFL Compute-then-Transmit (CT) protocol using hybrid Non-Orthogonal Multiple Access (H-NOMA). To minimize and optimize latency for the transmission of local training data, we use the Successive Convex Optimization (SCA) method, which efficiently reduces the complexity of non-convex algorithms. Finally, the numerical results verify the effectiveness of H-NOMA in terms of delay reduction, compared to the benchmark that is based on Non-Orthogonal Multiple Acces (NOMA).","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138624691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. This paper introduces a novel approach for hiding personal information in speech signals. The proposed approach applied a transform warping function, which is obtained from a long-term linear prediction spectrum individually for each speaker. The depersonalized speech was compared with the often used technique based on vocal tract length normalization. The proposed approach performs wider manipulation of fundamental frequency and provides higher intelligibility by 5% in clean speech and by 8% for signal-to-noise ratio 5 dB. It also significantly alters the derived glottal pulses, making them difficult to use for personality analysis. Speech intelligibility index and glottal pulse distortion are new aspects in the field of voice depersonalization.
{"title":"Depersonalization of Speech Using Speaker-Specific Transform Based on Long-Term Spectrum","authors":"M. Rujzl, M. Sigmund","doi":"10.13164/re.2023.0523","DOIUrl":"https://doi.org/10.13164/re.2023.0523","url":null,"abstract":". This paper introduces a novel approach for hiding personal information in speech signals. The proposed approach applied a transform warping function, which is obtained from a long-term linear prediction spectrum individually for each speaker. The depersonalized speech was compared with the often used technique based on vocal tract length normalization. The proposed approach performs wider manipulation of fundamental frequency and provides higher intelligibility by 5% in clean speech and by 8% for signal-to-noise ratio 5 dB. It also significantly alters the derived glottal pulses, making them difficult to use for personality analysis. Speech intelligibility index and glottal pulse distortion are new aspects in the field of voice depersonalization.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138616810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. A method for compensation of frequency-selective (FS) in-phase/quadrature (IQ) imbalance of a wideband transceiver is proposed in the paper. It is dedicated for implementation in software defined radio (SDR) cellular base stations. Both transmitter (TX) and receiver (RX) IQ impairments are corrected by complex valued finite impulse response (FIR) filters which are designed based on previously found imbalance correction models. The compensation performance is assessed after the method was implemented in the SDR platform capable of transmitting signals at different central frequencies. At frequencies higher than 3 GHz measured IQ gain and phase error functions exhibit asymmetrical characteristic. In order to reduce the level of asymmetry, adopted IQ gain correction model incorporates odd polynomial elements while the phase correction model includes even polynomial parts. Regardless of utilized central frequency IQ impairments are efficiently compensated. The advantage of the proposed method is low complexity. The method doesn't require specialized hardware for calibration, instead, it uses the RF loopback. At central frequency of 3.5 GHz, transmitter image rejection ratio (IRR) is increased from 20 dBc to 45–50 dBc by applying the proposed method. After receiver imbalance is compensated, the improvement in IRR of more than 25 dBc is achieved.
{"title":"IQ Imbalance Correction in Wideband Software Defined Radio Transceivers","authors":"B. Jovanović, S. Milenković","doi":"10.13164/re.2023.0479","DOIUrl":"https://doi.org/10.13164/re.2023.0479","url":null,"abstract":". A method for compensation of frequency-selective (FS) in-phase/quadrature (IQ) imbalance of a wideband transceiver is proposed in the paper. It is dedicated for implementation in software defined radio (SDR) cellular base stations. Both transmitter (TX) and receiver (RX) IQ impairments are corrected by complex valued finite impulse response (FIR) filters which are designed based on previously found imbalance correction models. The compensation performance is assessed after the method was implemented in the SDR platform capable of transmitting signals at different central frequencies. At frequencies higher than 3 GHz measured IQ gain and phase error functions exhibit asymmetrical characteristic. In order to reduce the level of asymmetry, adopted IQ gain correction model incorporates odd polynomial elements while the phase correction model includes even polynomial parts. Regardless of utilized central frequency IQ impairments are efficiently compensated. The advantage of the proposed method is low complexity. The method doesn't require specialized hardware for calibration, instead, it uses the RF loopback. At central frequency of 3.5 GHz, transmitter image rejection ratio (IRR) is increased from 20 dBc to 45–50 dBc by applying the proposed method. After receiver imbalance is compensated, the improvement in IRR of more than 25 dBc is achieved.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138617020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. Modulation classification is a crucial technique to utilize the unconsumed spectrum in Cognitive Radio (CR) and Dynamic Spectrum Access (DSA) systems to meet the required traffic demands for future-generation cellular networks. This paper presents an end-to-end experimental setup as a generic methodology to implement various Transfer Learning (TL) models in an indoor environment. This allows us to learn the features from multiple modulation signals to train and test the model. The performance evaluation of proposed TL models such as Convolutional Neural Network - Random Forest (CNN-RF), and Convolutional Long Short Term Deep Neural Network (CLDNN) - Random Forest (CLDNN-RF) have been thoroughly discussed. The result shows that the proposed TL models yield more than 90% classification accuracy for various modulation types. A proposed framework for location-specific TL model selection based on the maximum classification accuracy has been investigated.
{"title":"Transfer Learning based Location-Aided Modulation Classification in Indoor Environments for Cognitive Radio Applications","authors":"K. Tamizhelakkiya, S. Gauni, P. Chandhar","doi":"10.13164/re.2023.0531","DOIUrl":"https://doi.org/10.13164/re.2023.0531","url":null,"abstract":". Modulation classification is a crucial technique to utilize the unconsumed spectrum in Cognitive Radio (CR) and Dynamic Spectrum Access (DSA) systems to meet the required traffic demands for future-generation cellular networks. This paper presents an end-to-end experimental setup as a generic methodology to implement various Transfer Learning (TL) models in an indoor environment. This allows us to learn the features from multiple modulation signals to train and test the model. The performance evaluation of proposed TL models such as Convolutional Neural Network - Random Forest (CNN-RF), and Convolutional Long Short Term Deep Neural Network (CLDNN) - Random Forest (CLDNN-RF) have been thoroughly discussed. The result shows that the proposed TL models yield more than 90% classification accuracy for various modulation types. A proposed framework for location-specific TL model selection based on the maximum classification accuracy has been investigated.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138620932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. In this paper, a modified vector fitting technique to extract coupling matrix from S-parameters is introduced. This work allows designers to extract the coupling matrix of different or any pre-defined topologies from the simulated or measured S-parameter data. A study on vector fitting (VF) equations that can extract the rational polynomial of bandpass filter responses is carried out. VF is a robust numerical method which is applied widely in rational approximations due to its fast convergence and able to apply for high order polynomials. The rational polynomials are formed by applying the VF process to S-parameter responses without having to remove the phase offset and de-embedding the transmission lines. Focus fitting as the first enhancement can avoid VF overfitting spurious as poles; Poles forcing as the second enhancement is able to ensure the poles of all S-parameters are the same. Finally, the desired coupling matrix configuration is generated directly from the extracted polynomials using unconstrained and finitely bounded non-linear polynomials (NLP) optimization. Without the need for matrix transformation, the matrix elements are still able to show a one-to-one relationship in coupling values of resonators. Two bandpass filters are shown as examples to illustrate the performance of the new variation of VF.
{"title":"A Modified Vector Fitting Technique to Extract Coupling Matrix from S-parameters","authors":"C. Ng, S. Soeung, S. Cheab, K. Y. Leong","doi":"10.13164/re.2023.0325","DOIUrl":"https://doi.org/10.13164/re.2023.0325","url":null,"abstract":". In this paper, a modified vector fitting technique to extract coupling matrix from S-parameters is introduced. This work allows designers to extract the coupling matrix of different or any pre-defined topologies from the simulated or measured S-parameter data. A study on vector fitting (VF) equations that can extract the rational polynomial of bandpass filter responses is carried out. VF is a robust numerical method which is applied widely in rational approximations due to its fast convergence and able to apply for high order polynomials. The rational polynomials are formed by applying the VF process to S-parameter responses without having to remove the phase offset and de-embedding the transmission lines. Focus fitting as the first enhancement can avoid VF overfitting spurious as poles; Poles forcing as the second enhancement is able to ensure the poles of all S-parameters are the same. Finally, the desired coupling matrix configuration is generated directly from the extracted polynomials using unconstrained and finitely bounded non-linear polynomials (NLP) optimization. Without the need for matrix transformation, the matrix elements are still able to show a one-to-one relationship in coupling values of resonators. Two bandpass filters are shown as examples to illustrate the performance of the new variation of VF.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45332975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
. In this paper, the Direction of Arrival - DoA estimation for two mobile sources was performed by using the Single Multilayer Perceptron (MLP) neural network model (SMLP-DoA) and the Cascade MLP model(CMLP). The latter model consists of two neural networks connected in a cascade where the outputs of the first MLP that rejects noise represent the inputs to the second network in a cascade. The outputs of the neural network models determine the direction of arrival of the incoming signals. Two cases were considered, in the first case the neural networks were trained on the samples that were without noise, and in the second with samples containing noise. Both considered neural network models were tested with noisy samples. The results of these two neural models are compared to the results achieved by the RootMUSIC algorithm. The presented results show that the proposed CMLP model has a higher accuracy in determining the angular positions of sources compared to the classical SMLP-DoA model and the RootMUSIC algorithm. Moreover, the CMLP model executes significantly faster compared to the model based on the RootMUSIC algorithm.
{"title":"Spatial Localization of Electromagnetic Radiation Sources by Cascade Neural Network Model with Noise Reduction","authors":"Milan Ilic, Z. Stanković, N. M. Ilić","doi":"10.13164/re.2023.0381","DOIUrl":"https://doi.org/10.13164/re.2023.0381","url":null,"abstract":". In this paper, the Direction of Arrival - DoA estimation for two mobile sources was performed by using the Single Multilayer Perceptron (MLP) neural network model (SMLP-DoA) and the Cascade MLP model(CMLP). The latter model consists of two neural networks connected in a cascade where the outputs of the first MLP that rejects noise represent the inputs to the second network in a cascade. The outputs of the neural network models determine the direction of arrival of the incoming signals. Two cases were considered, in the first case the neural networks were trained on the samples that were without noise, and in the second with samples containing noise. Both considered neural network models were tested with noisy samples. The results of these two neural models are compared to the results achieved by the RootMUSIC algorithm. The presented results show that the proposed CMLP model has a higher accuracy in determining the angular positions of sources compared to the classical SMLP-DoA model and the RootMUSIC algorithm. Moreover, the CMLP model executes significantly faster compared to the model based on the RootMUSIC algorithm.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":null,"pages":null},"PeriodicalIF":1.1,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41490174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","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：