Pub Date : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786431
Sara Tayouri, I. Izadi, J. Ghaisari
Due to the development of technology, actuators possessing high precision performances are increasingly needed more than before. Because of their high precision, fast expansion, and independence from the magnetic field, piezoelectric actuators are superior in comparison to other actuators based on smart materials. However, hysteresis, creep, and vibrational dynamics are challenges faced while working with piezoelectric actuators. Among those characteristics, hysteresis, which is a nonlinear behavior, degrades precision, bandwidth, and accuracy. In this paper, a dynamical model for a piezoelectric actuator is obtained, and the parameters of the hysteresis, vibration, and creep models are calculated. The Levenberg-Marquardt algorithm is used for identification of the parameters of the Normalized Bouc-Wen, and the C-H model. The calculated models are tested and validated by applying a multi-sine wave as an input to the experimental setup. Also, a differential equation form for the C-H model is calculated which can be used to design a nonlinear controller. The obtained nonlinear hysteresis, linear creep, and linear vibration models can then be connected in series to provide a model to describe the main characteristics of a piezoelectric actuator.
{"title":"Modeling and Parameter Identification of Piezoelectric Actuator in Micropositioning Systems","authors":"Sara Tayouri, I. Izadi, J. Ghaisari","doi":"10.1109/IranianCEE.2019.8786431","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786431","url":null,"abstract":"Due to the development of technology, actuators possessing high precision performances are increasingly needed more than before. Because of their high precision, fast expansion, and independence from the magnetic field, piezoelectric actuators are superior in comparison to other actuators based on smart materials. However, hysteresis, creep, and vibrational dynamics are challenges faced while working with piezoelectric actuators. Among those characteristics, hysteresis, which is a nonlinear behavior, degrades precision, bandwidth, and accuracy. In this paper, a dynamical model for a piezoelectric actuator is obtained, and the parameters of the hysteresis, vibration, and creep models are calculated. The Levenberg-Marquardt algorithm is used for identification of the parameters of the Normalized Bouc-Wen, and the C-H model. The calculated models are tested and validated by applying a multi-sine wave as an input to the experimental setup. Also, a differential equation form for the C-H model is calculated which can be used to design a nonlinear controller. The obtained nonlinear hysteresis, linear creep, and linear vibration models can then be connected in series to provide a model to describe the main characteristics of a piezoelectric actuator.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"34 1","pages":"1193-1198"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84008759","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786570
S. F. Alem, I. Izadi, F. Sheikholeslam
Hysteresis phenomenon, which is prevalent in smart material actuators can affect control of the actuator, and should be compensated in the controller. In this paper its value is formulated as an internal state of the actuator, that is modeled by the Bouc-Wen model. A state observer, based on the position measurement is presented to estimate the hysteresis, which is not measurable by commercial sensors. The observer and convergence proof are presented using Lyapunov theorem, and the necessary constraints on design parameters are extracted. To facilitate these constraints, a lower order observer using velocity measurement is presented. The latter observer shows better transient and faster response. Both of the proposed observers are robust against model uncertainty. The effect of uncertainty in the estimated value is quantified and validated. An experimental setup using a piezoelectric actuator and an integrated strain gauge sensor is used as the test setup. Simulation and experimental results verify the effectiveness of the proposed schemes.
{"title":"Hysteresis Observation in Smart Material Actuators with Unstructured Uncertainty","authors":"S. F. Alem, I. Izadi, F. Sheikholeslam","doi":"10.1109/IranianCEE.2019.8786570","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786570","url":null,"abstract":"Hysteresis phenomenon, which is prevalent in smart material actuators can affect control of the actuator, and should be compensated in the controller. In this paper its value is formulated as an internal state of the actuator, that is modeled by the Bouc-Wen model. A state observer, based on the position measurement is presented to estimate the hysteresis, which is not measurable by commercial sensors. The observer and convergence proof are presented using Lyapunov theorem, and the necessary constraints on design parameters are extracted. To facilitate these constraints, a lower order observer using velocity measurement is presented. The latter observer shows better transient and faster response. Both of the proposed observers are robust against model uncertainty. The effect of uncertainty in the estimated value is quantified and validated. An experimental setup using a piezoelectric actuator and an integrated strain gauge sensor is used as the test setup. Simulation and experimental results verify the effectiveness of the proposed schemes.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"37 1","pages":"1035-1039"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80744654","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786410
Mostafa Fathi, S. Sheikhaei, Javad Tavakoli
In this work, one of local stereo vision algorithms named SAD approach, which is used in image depth estimation, has been surveyed, and an efficient and real-time new hardware implementation has been proposed. The proposed method has been verified and tested using C implementation. The acceptable simulation results along with the detailed explanation of numerous pre-processing steps are also presented. Our innovations could be divided into two sections: architecture and algorithm. In architecture section, by using a specific architecture, memory access has been lowered and therefore, speed has been increased. In algorithm section, a part of local algorithm, known as refinement, has been substituted with a simpler and more efficient algorithm. Cyclone IV has been utilized as our hardware platform. In this article, this point would demonstrate how to use an exact but less complicated controller, which results in less area, and how to use pipeline architecture and remove repetitive and redundant memory accesses, which conduct our stereo vision system to meet the real-time constraints. Suggested hardware implementation could reach to 53fps processing speed with 100MHz clock. No processing IP core has been used. In comparison with related work, our proposed method is more efficient in logic elements usage, and accordingly in power consumption.
{"title":"Low-cost and Real-time Hardware Implementation of Stereo Vision System on FPGA","authors":"Mostafa Fathi, S. Sheikhaei, Javad Tavakoli","doi":"10.1109/IranianCEE.2019.8786410","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786410","url":null,"abstract":"In this work, one of local stereo vision algorithms named SAD approach, which is used in image depth estimation, has been surveyed, and an efficient and real-time new hardware implementation has been proposed. The proposed method has been verified and tested using C implementation. The acceptable simulation results along with the detailed explanation of numerous pre-processing steps are also presented. Our innovations could be divided into two sections: architecture and algorithm. In architecture section, by using a specific architecture, memory access has been lowered and therefore, speed has been increased. In algorithm section, a part of local algorithm, known as refinement, has been substituted with a simpler and more efficient algorithm. Cyclone IV has been utilized as our hardware platform. In this article, this point would demonstrate how to use an exact but less complicated controller, which results in less area, and how to use pipeline architecture and remove repetitive and redundant memory accesses, which conduct our stereo vision system to meet the real-time constraints. Suggested hardware implementation could reach to 53fps processing speed with 100MHz clock. No processing IP core has been used. In comparison with related work, our proposed method is more efficient in logic elements usage, and accordingly in power consumption.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"1 1","pages":"258-263"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83059076","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786645
Abdolvahab Khalili Sadaghiani, Samad Sheikhai
Power spectrum analysis is a crucial matter in signal processing. In communication and biomedical signal processing, designing appropriate custom hardware for estimating signal is under everyday research. High frequency power spectrum estimator with embedded systems capabilities and for biomedical applications is the purpose of the current research. Nonparametric power spectral density estimation is the approach of this research in which no specific model is assumed in prior. This paper proposes a low power, high speed hardware with 125 MHz clock frequency rate, which its fast Fourier transform (FFT) engine is implemented with a radix-4 MDC structure. Bartlett method has been implemented in this research. The Artix-7 is the utilized FPGA in this research and VHDL is the used hardware design language. The proposed design computes Bartlett Power Spectral Density (PSD) estimation with only two added hardware units to a modified FFT engine. The offered limited power hardware performs well in small Word Length and high clock frequency.
{"title":"Hardware Implementation of High Speed Bartlett Spectral Density Estimator Based on R4MDC FFT","authors":"Abdolvahab Khalili Sadaghiani, Samad Sheikhai","doi":"10.1109/IranianCEE.2019.8786645","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786645","url":null,"abstract":"Power spectrum analysis is a crucial matter in signal processing. In communication and biomedical signal processing, designing appropriate custom hardware for estimating signal is under everyday research. High frequency power spectrum estimator with embedded systems capabilities and for biomedical applications is the purpose of the current research. Nonparametric power spectral density estimation is the approach of this research in which no specific model is assumed in prior. This paper proposes a low power, high speed hardware with 125 MHz clock frequency rate, which its fast Fourier transform (FFT) engine is implemented with a radix-4 MDC structure. Bartlett method has been implemented in this research. The Artix-7 is the utilized FPGA in this research and VHDL is the used hardware design language. The proposed design computes Bartlett Power Spectral Density (PSD) estimation with only two added hardware units to a modified FFT engine. The offered limited power hardware performs well in small Word Length and high clock frequency.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"26 1","pages":"1518-1522"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76358706","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786657
Saeid Ghafouri, S. H. Khasteh
The problem of influence maximization is a classic subject to study in the field of network science. It is about finding the top-k important individuals in a network for message dissemination under a particular diffusion model. Each year a number of new research papers are published concerning the same issue. However, most of these methods can only operate in situations where the whole graph is visible to the algorithm which is an unrealistic assumption in many cases. There are many cases where the induced network model of a natural phenomenon is associated with missing links. Discarding these links will lead to serious drawbacks in the result. In this work, we have extended the current state of the art influence maximization algorithms by adding a link prediction heuristic step prior to the actual run of the algorithm. For the purpose of link prediction, we have used exponential random graph models also known as ERGM due to their probabilistic link prediction capabilities. We have shown that this heuristic can significantly improve the effectiveness of influence maximization algorithms and in a diffusion scenario we can have a larger number of infected nodes using the seed nodes of the influence maximization algorithm.
{"title":"Influence Maximization Under Partially Observable Environments","authors":"Saeid Ghafouri, S. H. Khasteh","doi":"10.1109/IranianCEE.2019.8786657","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786657","url":null,"abstract":"The problem of influence maximization is a classic subject to study in the field of network science. It is about finding the top-k important individuals in a network for message dissemination under a particular diffusion model. Each year a number of new research papers are published concerning the same issue. However, most of these methods can only operate in situations where the whole graph is visible to the algorithm which is an unrealistic assumption in many cases. There are many cases where the induced network model of a natural phenomenon is associated with missing links. Discarding these links will lead to serious drawbacks in the result. In this work, we have extended the current state of the art influence maximization algorithms by adding a link prediction heuristic step prior to the actual run of the algorithm. For the purpose of link prediction, we have used exponential random graph models also known as ERGM due to their probabilistic link prediction capabilities. We have shown that this heuristic can significantly improve the effectiveness of influence maximization algorithms and in a diffusion scenario we can have a larger number of infected nodes using the seed nodes of the influence maximization algorithm.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"14 1","pages":"1984-1988"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80933464","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786640
M. Rafaei-Booket, S. H. Taheri
In this paper, a wide-band and high-power array antenna with air substrate is proposed. Such an array antenna is designed for using in Tactical Air Navigation (TACAN). To do so, the elements of the array are excited such a way that the resultant antenna produces a cosecant-squared radiation pattern in elevation plane. To this end, it is utilized an iterative method for computing the optimized elements excitations to synthesis cosecant-squared pattern with reduced side lobe level (SLL). The obtained elements' excitations are also implemented using a metallic feed network. Finally, the designed array antenna along with feed network is fabricated and successfully tested. It is shown that the measurements are in good agreement with those obtained by simulations. In addition, the implemented array along with its feed network has 23% bandwidth, maximum gain 9.7dBi, and SLL<-23dB at central frequency (=1100MHz).
{"title":"Low-cost Metallic Linear Array Antenna with Cosecant-Squared Radiation Pattern","authors":"M. Rafaei-Booket, S. H. Taheri","doi":"10.1109/IranianCEE.2019.8786640","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786640","url":null,"abstract":"In this paper, a wide-band and high-power array antenna with air substrate is proposed. Such an array antenna is designed for using in Tactical Air Navigation (TACAN). To do so, the elements of the array are excited such a way that the resultant antenna produces a cosecant-squared radiation pattern in elevation plane. To this end, it is utilized an iterative method for computing the optimized elements excitations to synthesis cosecant-squared pattern with reduced side lobe level (SLL). The obtained elements' excitations are also implemented using a metallic feed network. Finally, the designed array antenna along with feed network is fabricated and successfully tested. It is shown that the measurements are in good agreement with those obtained by simulations. In addition, the implemented array along with its feed network has 23% bandwidth, maximum gain 9.7dBi, and SLL<-23dB at central frequency (=1100MHz).","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"120 1","pages":"1452-1455"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91076351","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786729
A. Shadmani, M. Miri
Application of Silicon-based optical modulators in some wavelength ranges (such as O-band) is challenging because of the two photon absorption (TPA) effect of Si. Silicon nitride (Si3N4), on the other hand has a larger bandgap compared to Si, which minimizes the TPA effect in it. The lack of fast free carrier dispersion effects and relatively large footprint of Si3N4optical waveguides are the basic shortcomings of Si3N4-based devices. Here, integration of the phase change material of Ge2Sb2Te5within the Si3N4waveguides is proposed as a solution to overcome both these limitations. Phase transition of the GST provides a mechanism for optical modulation while large variation in its refractive index during the phase transitions, reduces the length of the modulators. A dual polarization optical modulator is designed in this manuscript, which according to our simulations provides respectively, the reasonable insertion loss and high extinction ratio of ~1.15dB, and 8.12dB for the TE, and low insertion loss of ~0.57dB, and the extinction of 1.75dB, for the TM polarized light. Furthermore, the proposed structure offers ultra-wide optical bandwidth of 500nm (covering O-band to U-band), small overall footprint 2.4µm2, and its bitrate and energy consumption is estimated to be about 0.4Gbit/sec and 12pJ, respectively.
{"title":"Design and Simulation of Dual Polarization GST-on-Silicon Nitride Optical Modulator","authors":"A. Shadmani, M. Miri","doi":"10.1109/IranianCEE.2019.8786729","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786729","url":null,"abstract":"Application of Silicon-based optical modulators in some wavelength ranges (such as O-band) is challenging because of the two photon absorption (TPA) effect of Si. Silicon nitride (Si3N4), on the other hand has a larger bandgap compared to Si, which minimizes the TPA effect in it. The lack of fast free carrier dispersion effects and relatively large footprint of Si3N4optical waveguides are the basic shortcomings of Si3N4-based devices. Here, integration of the phase change material of Ge2Sb2Te5within the Si3N4waveguides is proposed as a solution to overcome both these limitations. Phase transition of the GST provides a mechanism for optical modulation while large variation in its refractive index during the phase transitions, reduces the length of the modulators. A dual polarization optical modulator is designed in this manuscript, which according to our simulations provides respectively, the reasonable insertion loss and high extinction ratio of ~1.15dB, and 8.12dB for the TE, and low insertion loss of ~0.57dB, and the extinction of 1.75dB, for the TM polarized light. Furthermore, the proposed structure offers ultra-wide optical bandwidth of 500nm (covering O-band to U-band), small overall footprint 2.4µm2, and its bitrate and energy consumption is estimated to be about 0.4Gbit/sec and 12pJ, respectively.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"12 1","pages":"220-223"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87471420","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786628
Hamid Reza Shabani, M. Kalantar
In this paper, for Transient Stability Assessment (TSA), an approach based on two sets of intelligent units proposed that are using the appropriate inputs to quickly and accurately determine the Critical Clearing Time (CCT). The appropriate inputs are maximum potential energy, minimum kinetic energy and the slope of minimum kinetic energy. Initially, the effectiveness of these inputs is investigated using an intelligent unit in Set 1 in order to accurately identify CCT. Then to estimate CCT more quickly, the appropriate inputs of the intelligent unit in Set 1 are estimated using three other intelligent units (Set 2). Simulation results show that in this case CCT is determined through simulating only 0.05 seconds of the system response. This indicates the suitability of the proposed method for on-line pre-disturbance TSA.
{"title":"Transient Stability Detection via Quick Estimation of the Critical Clearing Time Using Intelligent Systems with Energy Functions Based Inputs","authors":"Hamid Reza Shabani, M. Kalantar","doi":"10.1109/IranianCEE.2019.8786628","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786628","url":null,"abstract":"In this paper, for Transient Stability Assessment (TSA), an approach based on two sets of intelligent units proposed that are using the appropriate inputs to quickly and accurately determine the Critical Clearing Time (CCT). The appropriate inputs are maximum potential energy, minimum kinetic energy and the slope of minimum kinetic energy. Initially, the effectiveness of these inputs is investigated using an intelligent unit in Set 1 in order to accurately identify CCT. Then to estimate CCT more quickly, the appropriate inputs of the intelligent unit in Set 1 are estimated using three other intelligent units (Set 2). Simulation results show that in this case CCT is determined through simulating only 0.05 seconds of the system response. This indicates the suitability of the proposed method for on-line pre-disturbance TSA.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"5 1","pages":"614-618"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78915803","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786669
Zahra Motazedian, A. Safavi
This paper presents a novel Adaptive Fully Connected Recurrent Wavelet Network (AFCRWN) for online identification of nonlinear dynamic and time varying systems. The AFCRWN inherits the architecture of fully connected recurrent neural network proposed by Williams & Zipser. Since the AFCRWN incorporates translated and dilated versions of scaling function and wavelet instead of global functions as activation functions of hidden neurons, this would lead to a significant improvement of network performance. An adaptive gradient based algorithm is used to adjust the shapes and weights of scaling functions and wavelets. Simulation results for modeling of different dynamic nonlinear and dynamic nonlinear and time varying systems are presented. Comparisons with a network of neurons with wavelets and a network of neurons with sigmoid functions are provided. Computer simulation results have successfully validated the superior performance of AFCRWN.
{"title":"Nonlinear and Time Varying System Identification Using a Novel Adaptive Fully Connected Recurrent Wavelet Network","authors":"Zahra Motazedian, A. Safavi","doi":"10.1109/IranianCEE.2019.8786669","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786669","url":null,"abstract":"This paper presents a novel Adaptive Fully Connected Recurrent Wavelet Network (AFCRWN) for online identification of nonlinear dynamic and time varying systems. The AFCRWN inherits the architecture of fully connected recurrent neural network proposed by Williams & Zipser. Since the AFCRWN incorporates translated and dilated versions of scaling function and wavelet instead of global functions as activation functions of hidden neurons, this would lead to a significant improvement of network performance. An adaptive gradient based algorithm is used to adjust the shapes and weights of scaling functions and wavelets. Simulation results for modeling of different dynamic nonlinear and dynamic nonlinear and time varying systems are presented. Comparisons with a network of neurons with wavelets and a network of neurons with sigmoid functions are provided. Computer simulation results have successfully validated the superior performance of AFCRWN.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"25 1","pages":"1181-1187"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77181310","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 : 2019-04-01DOI: 10.1109/IranianCEE.2019.8786761
Nazanin Ahmadi Dastgerdi, Hossein Hoseini-Nejad, H. Amiri
This paper reports a novel spike compression approach for implantable intra-cortical neural recording microsystems based on Split Vector Quantization (SVQ). The proposed method presents a spike compression ratio 14.8 at the cost of classification accuracy (CA). The average value of CA is 94% over a wide range (7 to 15) of signal to noise ratios (SNR) of the neural signal.
{"title":"Neural Spike Compression Based on Split Vector Quantization for Implantable BMIs","authors":"Nazanin Ahmadi Dastgerdi, Hossein Hoseini-Nejad, H. Amiri","doi":"10.1109/IranianCEE.2019.8786761","DOIUrl":"https://doi.org/10.1109/IranianCEE.2019.8786761","url":null,"abstract":"This paper reports a novel spike compression approach for implantable intra-cortical neural recording microsystems based on Split Vector Quantization (SVQ). The proposed method presents a spike compression ratio 14.8 at the cost of classification accuracy (CA). The average value of CA is 94% over a wide range (7 to 15) of signal to noise ratios (SNR) of the neural signal.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"24 1","pages":"1779-1782"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86183530","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：