Wenhang Ge, C. Qi, Y. Guo, L. Qian, R. Tong, P. Wei
. Massive multi-input multi-output (MIMO) has attracted significant interest in academia and industry, which can efficiently increase the transmission rate. However, the error rate of conventional channel equalizations in massive MIMO systems may be high owing to the dynamic channel states in practical conditions. To solve this problem, in this paper, we propose an improved channel equalization framework based on the deep neural network (DNN). Based on the analyzed relationship between the input and output of the DNN, the data can be recovered without the channel state information. Furthermore, aiming at reducing the convergence time and enhancing the learning ability of the DNN, a classification weighted algorithm is proposed to optimize the cost function of the DNN, which is named as classification weighted deep neural network (CW-DNN). Simulation results demonstrate that compared to conventional counterparts, the proposed CW-DNN based equalizer can achieve a better normalized mean square error (NMSE). Upon approximating the optimal neural network parameters with the significantly improved convergence speed and reduced training time of the network, under the condition of the fixed learning rate.
{"title":"Classification Weighted Deep Neural Network Based Channel Equalization for Massive MIMO-OFDM Systems","authors":"Wenhang Ge, C. Qi, Y. Guo, L. Qian, R. Tong, P. Wei","doi":"10.13164/re.2022.0346","DOIUrl":"https://doi.org/10.13164/re.2022.0346","url":null,"abstract":". Massive multi-input multi-output (MIMO) has attracted significant interest in academia and industry, which can efficiently increase the transmission rate. However, the error rate of conventional channel equalizations in massive MIMO systems may be high owing to the dynamic channel states in practical conditions. To solve this problem, in this paper, we propose an improved channel equalization framework based on the deep neural network (DNN). Based on the analyzed relationship between the input and output of the DNN, the data can be recovered without the channel state information. Furthermore, aiming at reducing the convergence time and enhancing the learning ability of the DNN, a classification weighted algorithm is proposed to optimize the cost function of the DNN, which is named as classification weighted deep neural network (CW-DNN). Simulation results demonstrate that compared to conventional counterparts, the proposed CW-DNN based equalizer can achieve a better normalized mean square error (NMSE). Upon approximating the optimal neural network parameters with the significantly improved convergence speed and reduced training time of the network, under the condition of the fixed learning rate.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48044307","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 deals with a method of designing PID controllers. Generalized Laguerre functions were used for this task. Generalized Laguerre functions generate an orthogonal base in the time domain and the operator domain. This property of generalized Laguerre functions is benefi-cially used for the design of the PID controller. Parameters for generalized Laguerre function PID controllers are computed from the Laguerre series of the open loop and the Laguerre series of the ideal open loop. To satisfy this goal, the plant transfer function, the controller transfer function, and the ideal open loop transfer function are transformed into a generalized Laguerre functions base. Three examples are shown to present this method.
{"title":"Design PID Controllers Using Generalized Laguerre Functions","authors":"V. Kárský, M. Tuma","doi":"10.13164/re.2022.0390","DOIUrl":"https://doi.org/10.13164/re.2022.0390","url":null,"abstract":". This paper deals with a method of designing PID controllers. Generalized Laguerre functions were used for this task. Generalized Laguerre functions generate an orthogonal base in the time domain and the operator domain. This property of generalized Laguerre functions is benefi-cially used for the design of the PID controller. Parameters for generalized Laguerre function PID controllers are computed from the Laguerre series of the open loop and the Laguerre series of the ideal open loop. To satisfy this goal, the plant transfer function, the controller transfer function, and the ideal open loop transfer function are transformed into a generalized Laguerre functions base. Three examples are shown to present this method.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49239185","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}
{"title":"A 2V, 32.13nA, fully MOSFET Voltage Limiter for Low Power Applications","authors":"H. Rayat, R. Dastanian","doi":"10.13164/re.2022.0323","DOIUrl":"https://doi.org/10.13164/re.2022.0323","url":null,"abstract":"","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47267351","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 novel high-sensitivity broadband rectifier is proposed aiming at ambient radio frequency (RF) energy harvesting. Traditionally, voltage doubling rectifying circuit is used to design high-sensitivity rectifier. But when the input power is lower, the rectifying efficiency is significantly reduced. Therefore, an improved parallel half-wave rectifying circuit is proposed in this article which can convert RF energy in the whole period. And the proposed rectifying circuit can work better in lower power environment and has a higher efficiency level. Besides, the impedance match is also important component of rectifier. Due to the nonlinearity and complexity of rectifying circuit, achieving wideband matching network is a challenge. Thus, a design approach of broadband impedance circuit is given in this study. Combining with the proposed high-sensitivity rectifying circuit, a high-sensitivity wideband rectifier can be generated, when the input power is –15 dBm, –20 dBm, –25 dBm, the efficiency is 43%, 32%, 20%, respectively. Finally, a second-order wideband rectifier with high sensitivity is realized, and the range of bandwidth can cover four main frequency bands of GSM 900 MHz, GSM 1800 MHz, UMTS 2100 MHz, WLAN 2400 MHz. To verify the validity, the rectifier is fabricated and measured, and the measurement has a good agreement with simulation results.
{"title":"A Novel High-Sensitivity Broadband Rectifier for Ambient RF Energy Harvesting","authors":"X. Wang, B. Jin, L. Huang, M. Zhang, M. Fang","doi":"10.13164/re.2022.0331","DOIUrl":"https://doi.org/10.13164/re.2022.0331","url":null,"abstract":". In this paper, a novel high-sensitivity broadband rectifier is proposed aiming at ambient radio frequency (RF) energy harvesting. Traditionally, voltage doubling rectifying circuit is used to design high-sensitivity rectifier. But when the input power is lower, the rectifying efficiency is significantly reduced. Therefore, an improved parallel half-wave rectifying circuit is proposed in this article which can convert RF energy in the whole period. And the proposed rectifying circuit can work better in lower power environment and has a higher efficiency level. Besides, the impedance match is also important component of rectifier. Due to the nonlinearity and complexity of rectifying circuit, achieving wideband matching network is a challenge. Thus, a design approach of broadband impedance circuit is given in this study. Combining with the proposed high-sensitivity rectifying circuit, a high-sensitivity wideband rectifier can be generated, when the input power is –15 dBm, –20 dBm, –25 dBm, the efficiency is 43%, 32%, 20%, respectively. Finally, a second-order wideband rectifier with high sensitivity is realized, and the range of bandwidth can cover four main frequency bands of GSM 900 MHz, GSM 1800 MHz, UMTS 2100 MHz, WLAN 2400 MHz. To verify the validity, the rectifier is fabricated and measured, and the measurement has a good agreement with simulation results.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46374193","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 presents a comparative EMC susceptibility study of various integrated bandgap voltage reference cores. Conventional well-known bandgap references based on Kuijk, Brokaw and Tsividis concepts with reduced count of bipolar junction transistors in the core were analyzed. On top of the EMC susceptibility comparison, basic parameters like temperature drift, sensitivity to an operational amplifier input offset and line regulation are also discussed. The influence of a collector leakage current compensation at high temperatures is investigated as well.
{"title":"An EMC Susceptibility Study of Integrated Basic Bandgap Voltage Reference Cores","authors":"D. Krolák, P. Horský","doi":"10.13164/re.2022.0413","DOIUrl":"https://doi.org/10.13164/re.2022.0413","url":null,"abstract":". This paper presents a comparative EMC susceptibility study of various integrated bandgap voltage reference cores. Conventional well-known bandgap references based on Kuijk, Brokaw and Tsividis concepts with reduced count of bipolar junction transistors in the core were analyzed. On top of the EMC susceptibility comparison, basic parameters like temperature drift, sensitivity to an operational amplifier input offset and line regulation are also discussed. The influence of a collector leakage current compensation at high temperatures is investigated as well.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44653630","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}
Ee Meng Cheng, Kim Yee Lee, S. F. Khor, Nashrul Fazli Mohd Nasir, C. W. S. R. Mohamad, Nor Azah Abdul Aziz, E. Tarmizi, S. A. Baharuddin
. Honey adulteration is common in the food industry as it provides a cheaper alternative for the user to con-sume honey. However, it has been abused by industry runners with unsavory practices. As a result, it leads to busi-ness fraudulency. Pure honey is very precious due to its powerful health-giving properties. It raises the attention of beekeepers, wholesalers, food manufacturers, retailers, and consumers because this issue has been sensationally reported in mass media. Enforcement of the law is initiated to mitigate the abuse and fraudulency. It also motivates scientists, technologists, and engineers to strive for an effective solution. The microwave sensing method is well known in agricultural products and food. Hence, dielectric and reflection responses are explored to study the potential of the development of an instrumentation system for gaug-ing edible honey. In this work, the dielectric and reflection measurement was conducted using Agilent E8362B PNA Network Analyzer in conjunction with Agilent 85070E Performance Probe from 0.5 GHz to 4.5 GHz. Dielectric and reflection measurements were conducted to investigate dielectric behavior and mismatch impedance due to water and sucrose content in honey. It can be noticed that the dielectric constant, ε ', decreases as frequency increases. In the meantime, ε ' decreases with the decrement of water and the increment of sucrose content for Honey Gold and Trigona Honey. Meanwhile, for water adulterated Honey Gold and Trigona Honey, the loss factor, ε " decreases when frequencies increase. In addition, ε " decreases when the water content is < 36% and < 43% for Honey Gold and Trigona Honey, respectively. It can be found that at 1 GHz to 4 GHz, ε " increases when sucrose content increases which is applicable for Honey Gold and Trigona Honey. In reflection measurement, the magnitude of reflection coefficient, | Γ | decreases when frequency increases for all percentage of water and sucrose content for both kinds of honey. Withal, phase, – φ increases as frequency increases for water adulterated honey. – φ varies insignificantly when sucrose content increases for sucrose adulterated honey.
{"title":"Microwave Dielectric and Reflection Analysis on Pure and Adulterated Trigona Honey and Honey Gold","authors":"Ee Meng Cheng, Kim Yee Lee, S. F. Khor, Nashrul Fazli Mohd Nasir, C. W. S. R. Mohamad, Nor Azah Abdul Aziz, E. Tarmizi, S. A. Baharuddin","doi":"10.13164/re.2022.0281","DOIUrl":"https://doi.org/10.13164/re.2022.0281","url":null,"abstract":". Honey adulteration is common in the food industry as it provides a cheaper alternative for the user to con-sume honey. However, it has been abused by industry runners with unsavory practices. As a result, it leads to busi-ness fraudulency. Pure honey is very precious due to its powerful health-giving properties. It raises the attention of beekeepers, wholesalers, food manufacturers, retailers, and consumers because this issue has been sensationally reported in mass media. Enforcement of the law is initiated to mitigate the abuse and fraudulency. It also motivates scientists, technologists, and engineers to strive for an effective solution. The microwave sensing method is well known in agricultural products and food. Hence, dielectric and reflection responses are explored to study the potential of the development of an instrumentation system for gaug-ing edible honey. In this work, the dielectric and reflection measurement was conducted using Agilent E8362B PNA Network Analyzer in conjunction with Agilent 85070E Performance Probe from 0.5 GHz to 4.5 GHz. Dielectric and reflection measurements were conducted to investigate dielectric behavior and mismatch impedance due to water and sucrose content in honey. It can be noticed that the dielectric constant, ε ', decreases as frequency increases. In the meantime, ε ' decreases with the decrement of water and the increment of sucrose content for Honey Gold and Trigona Honey. Meanwhile, for water adulterated Honey Gold and Trigona Honey, the loss factor, ε \" decreases when frequencies increase. In addition, ε \" decreases when the water content is < 36% and < 43% for Honey Gold and Trigona Honey, respectively. It can be found that at 1 GHz to 4 GHz, ε \" increases when sucrose content increases which is applicable for Honey Gold and Trigona Honey. In reflection measurement, the magnitude of reflection coefficient, | Γ | decreases when frequency increases for all percentage of water and sucrose content for both kinds of honey. Withal, phase, – φ increases as frequency increases for water adulterated honey. – φ varies insignificantly when sucrose content increases for sucrose adulterated honey.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43570092","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}
. Non-Orthogonal Multiple Access (NOMA) tech-nique is a remarkable component of 5G wireless networks; since NOMA immensely augments the spectral efficiency and serves all users fairly. To accomplish these, efficient power allocation is crucial for improving the NOMA system’s performance. Accordingly, in this article, we formulate a power allocation optimization issue, which concentrates on enrich-ing the system sum-throughput, by realizing the transmitted power constraint and also fulfilling the minimum throughput for each user. However, to tackle this mentioned optimization problem, a Modified Artificial Bee Colony (MABC) algorithm is proposed. Besides, the designed MABC algorithm obtains optimal powers among multiplexed users on every sub-channel. Further, simulation results illustrate that the presented power allocation scheme-based NOMA system’s sum throughput is higher than the original ABC-based power allocation and other state-of-the-art power allocation schemes. Moreover, the MABC method swiftly converges to optimal solutions compared to the original ABC algorithm under selected control parameters.
{"title":"Innovative Power Allocation Strategy for NOMA Systems by Employing the Modified ABC Algorithm","authors":"S. Sreenu, N. Kalpana","doi":"10.13164/re.2022.0312","DOIUrl":"https://doi.org/10.13164/re.2022.0312","url":null,"abstract":". Non-Orthogonal Multiple Access (NOMA) tech-nique is a remarkable component of 5G wireless networks; since NOMA immensely augments the spectral efficiency and serves all users fairly. To accomplish these, efficient power allocation is crucial for improving the NOMA system’s performance. Accordingly, in this article, we formulate a power allocation optimization issue, which concentrates on enrich-ing the system sum-throughput, by realizing the transmitted power constraint and also fulfilling the minimum throughput for each user. However, to tackle this mentioned optimization problem, a Modified Artificial Bee Colony (MABC) algorithm is proposed. Besides, the designed MABC algorithm obtains optimal powers among multiplexed users on every sub-channel. Further, simulation results illustrate that the presented power allocation scheme-based NOMA system’s sum throughput is higher than the original ABC-based power allocation and other state-of-the-art power allocation schemes. Moreover, the MABC method swiftly converges to optimal solutions compared to the original ABC algorithm under selected control parameters.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45622504","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}
. Interleaved polar (I-Polar) codes, a new facet of polar codes to achieve better channel capacity, is designed by placing the interleaver and deinterleaver blocks midway between the stages of the polar codes. Low latency hardware optimization makes their implementation even more suitable for ultra-reliable low latency applications. This study pro-poses an optimal hardware design for low latency interleaved polar codes by reframing the last stage of the interleaved successive cancellation decoder. A high-speed adder-subtractor is used to reduce the latency further, thus increasing the speed of operation. Interleaving data in the proposed polar codes augment BER performance compared to conventional (n, k) polar codes. The proposed I-Polar codes are synthesized using Synopsys design compiler (SDC) in CMOS 65-nm technology. Results show that the latency is reduced by 50.5% on average compared to the conventional polar codes as high-speed adder and merged processing elements are used. Moreover, the average gate count and power are reduced by 14% and 40.56%, respectively.
{"title":"Low Latency SC Decoder Architecture for Interleaved Polar Codes","authors":"N. Jali, P. Muralidhar, S. Patri","doi":"10.13164/re.2022.0398","DOIUrl":"https://doi.org/10.13164/re.2022.0398","url":null,"abstract":". Interleaved polar (I-Polar) codes, a new facet of polar codes to achieve better channel capacity, is designed by placing the interleaver and deinterleaver blocks midway between the stages of the polar codes. Low latency hardware optimization makes their implementation even more suitable for ultra-reliable low latency applications. This study pro-poses an optimal hardware design for low latency interleaved polar codes by reframing the last stage of the interleaved successive cancellation decoder. A high-speed adder-subtractor is used to reduce the latency further, thus increasing the speed of operation. Interleaving data in the proposed polar codes augment BER performance compared to conventional (n, k) polar codes. The proposed I-Polar codes are synthesized using Synopsys design compiler (SDC) in CMOS 65-nm technology. Results show that the latency is reduced by 50.5% on average compared to the conventional polar codes as high-speed adder and merged processing elements are used. Moreover, the average gate count and power are reduced by 14% and 40.56%, respectively.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42051053","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 proposes a novel wideband, high-gain, compact Vivaldi antenna operating in the 14.7–20.5 GHz frequency range, where the antenna is fed by the substrate integrated waveguide (SIW). A Negative Index Metamaterial (NIM) has been designed, with its parameters extracted using MATLAB software. The NIM was developed to address the deficiencies of conventional Antipodal Vivaldi antennas. With a measured gain of over 5.3 dBi, the fabricated antenna performs satisfactorily across the entire bandwidth. Due to its low profile and short transversal dimension, the proposed antenna is suitable for antenna arrays; consequently, the proposed antenna design could be a viable option for modern communication and radar systems.
{"title":"Gain and Bandwidth Enhancement of a Metamaterial Loaded Antipodal Vivaldi Antenna Fed by Substrate Integrated Waveguide","authors":"Amir Bordbar, F. Mohajeri, Z. Ghorbani","doi":"10.13164/re.2022.0193","DOIUrl":"https://doi.org/10.13164/re.2022.0193","url":null,"abstract":". This paper proposes a novel wideband, high-gain, compact Vivaldi antenna operating in the 14.7–20.5 GHz frequency range, where the antenna is fed by the substrate integrated waveguide (SIW). A Negative Index Metamaterial (NIM) has been designed, with its parameters extracted using MATLAB software. The NIM was developed to address the deficiencies of conventional Antipodal Vivaldi antennas. With a measured gain of over 5.3 dBi, the fabricated antenna performs satisfactorily across the entire bandwidth. Due to its low profile and short transversal dimension, the proposed antenna is suitable for antenna arrays; consequently, the proposed antenna design could be a viable option for modern communication and radar systems.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44278481","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 proposed a parametric modeling technique for the microwave structures with a customization magnitude response by combining the RBF neural network and pole-residue-based transfer functions. The Latin hypercube sampling method is used for sampling given physical ranges and obtaining the EM behaviors of the microwave structures. A pole sorting process and a modified pole-residues splitting process are proposed to solve the pole sequence chaos and order-changing problems which occur in the modeling process. The pole-residues parameters after the above pre-processing steps are used as the inputs of the RBF neural network and the physical parameters are used as the outputs of RBF network. Then, the known magnitude response of the microwave structure are used as the prior knowledge to guide obtaining the goal pole-residues values corresponding to the giving magnitude response specification. After the training process of the RBF model, the goal pole-residues are input into the trained RBF network and the goal physical parameters corresponding to the customization responses is obtained. Finally, this model technique is illustrated by the two examples of microwave structures.
{"title":"Parametric Modeling of Microwave Structure with Customization Responses by Combining RBF Neural Network and Pole-Residue-Based Transfer Functions","authors":"Y. Ma, S. Wu, Y. Yuan, N. Yuan","doi":"10.13164/re.2022.0185","DOIUrl":"https://doi.org/10.13164/re.2022.0185","url":null,"abstract":". This paper proposed a parametric modeling technique for the microwave structures with a customization magnitude response by combining the RBF neural network and pole-residue-based transfer functions. The Latin hypercube sampling method is used for sampling given physical ranges and obtaining the EM behaviors of the microwave structures. A pole sorting process and a modified pole-residues splitting process are proposed to solve the pole sequence chaos and order-changing problems which occur in the modeling process. The pole-residues parameters after the above pre-processing steps are used as the inputs of the RBF neural network and the physical parameters are used as the outputs of RBF network. Then, the known magnitude response of the microwave structure are used as the prior knowledge to guide obtaining the goal pole-residues values corresponding to the giving magnitude response specification. After the training process of the RBF model, the goal pole-residues are input into the trained RBF network and the goal physical parameters corresponding to the customization responses is obtained. Finally, this model technique is illustrated by the two examples of microwave structures.","PeriodicalId":54514,"journal":{"name":"Radioengineering","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41328346","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：