Pub Date : 2017-06-01DOI: 10.1109/NAECON.2017.8268720
Ying Li, Sihao Ding, Yuan F. Zheng, D. Xuan
Robot follower, a robot following its human operator, has found its application in many areas such as senior care, manufacturing, transportation, and etc. Tracking the target person is a key technique for the follower. In this paper, we present a new method for partial human body tracking, namely human feet tracking. Human feet tracking suffers from weak visual features and appearance variations, making it more critical to continuously update the foot appearance model. We propose to utilize the human motion model to predict foot appearance. It is achieved by first defining a motion phase to each human foot appearance. Due to the fact that the foot appearance across different motion cycles with the same motion phase is similar, we can predict the target appearance using the current motion phase and the target images stored from previous walking cycles. A phase labeled exemplar pool is built to serve the motion phase indexed appearance searching. We combine this phase labeled exemplar pool into particle filtering and have achieved robust human feet tracking.
{"title":"Motion model enabled appearance prediction for partial human body tracking in robot follower","authors":"Ying Li, Sihao Ding, Yuan F. Zheng, D. Xuan","doi":"10.1109/NAECON.2017.8268720","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268720","url":null,"abstract":"Robot follower, a robot following its human operator, has found its application in many areas such as senior care, manufacturing, transportation, and etc. Tracking the target person is a key technique for the follower. In this paper, we present a new method for partial human body tracking, namely human feet tracking. Human feet tracking suffers from weak visual features and appearance variations, making it more critical to continuously update the foot appearance model. We propose to utilize the human motion model to predict foot appearance. It is achieved by first defining a motion phase to each human foot appearance. Due to the fact that the foot appearance across different motion cycles with the same motion phase is similar, we can predict the target appearance using the current motion phase and the target images stored from previous walking cycles. A phase labeled exemplar pool is built to serve the motion phase indexed appearance searching. We combine this phase labeled exemplar pool into particle filtering and have achieved robust human feet tracking.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132756476","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268761
Takumi Fujimori, Minora Watanabe
Recently, many studies of field programmable gate array (FPGA) hardware accelerators have been reported, in addition to studies of general-purpose computing on graphics processing units (GPGPUs), Xeon Phi, and so on. Since parallel processing is indispensable for such accelerating applications on FPGAs, implementing numerous parallel processing circuits is important to improve the performance of such FPGA hardware accelerators. When implementing a parallel operation for a conventional FPGA, some waste occurs: the same context is stored on numerous regions of configuration memory. This waste presents a critical issue because FPGAs used as accelerators perform parallel processing exclusively in most cases. This paper therefore proposes a parallel-operation-oriented FPGA exploiting a common configuration context. Herein, we describe the advantages of gate density, propagation delay, and compilation time in parallel-operation-oriented FPGAs.
{"title":"Gate density advantage of parallel-operation-oriented FPGA architecture","authors":"Takumi Fujimori, Minora Watanabe","doi":"10.1109/NAECON.2017.8268761","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268761","url":null,"abstract":"Recently, many studies of field programmable gate array (FPGA) hardware accelerators have been reported, in addition to studies of general-purpose computing on graphics processing units (GPGPUs), Xeon Phi, and so on. Since parallel processing is indispensable for such accelerating applications on FPGAs, implementing numerous parallel processing circuits is important to improve the performance of such FPGA hardware accelerators. When implementing a parallel operation for a conventional FPGA, some waste occurs: the same context is stored on numerous regions of configuration memory. This waste presents a critical issue because FPGAs used as accelerators perform parallel processing exclusively in most cases. This paper therefore proposes a parallel-operation-oriented FPGA exploiting a common configuration context. Herein, we describe the advantages of gate density, propagation delay, and compilation time in parallel-operation-oriented FPGAs.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122422295","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268789
Abdunaser Abdusamad, M. Wicks, H. Abdelbagi, Abdulhakim Daluom, Muftah Akroush, Turki M. Alanazi
Tomography is defined as an image reconstruction technique that exploits different points of view, and different observation points. Generally, tomographic images can be constructed by illuminating the target with a microwave signals and measuring the energy that passes through the target e.g. transmission tomography, or reflected from the target as in reflection tomography. Reflection tomography is used in some cases when the transmission tomography cannot be accomplished due to physical limitations, or high attenuation losses due to high material impedance. In this paper, we investigate the impact of surfaces placed beside the transmitter and/or the receiver, and quantify these effects via image analysis at high frequency domain. Our goal in this paper is to reduce the effect of side-lobes that may appear as result of these surfaces. We reduce the reflected waves from the surfaces by designing the surfaces using specific materials or by changing the placement of both the transmitter and/or receiver antenna. Moreover, the resulting tomographic image will be processed using the Winner filter to remove any remaining noise.
{"title":"The impact of reflected waves on the reconstruction of tomographic imaging","authors":"Abdunaser Abdusamad, M. Wicks, H. Abdelbagi, Abdulhakim Daluom, Muftah Akroush, Turki M. Alanazi","doi":"10.1109/NAECON.2017.8268789","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268789","url":null,"abstract":"Tomography is defined as an image reconstruction technique that exploits different points of view, and different observation points. Generally, tomographic images can be constructed by illuminating the target with a microwave signals and measuring the energy that passes through the target e.g. transmission tomography, or reflected from the target as in reflection tomography. Reflection tomography is used in some cases when the transmission tomography cannot be accomplished due to physical limitations, or high attenuation losses due to high material impedance. In this paper, we investigate the impact of surfaces placed beside the transmitter and/or the receiver, and quantify these effects via image analysis at high frequency domain. Our goal in this paper is to reduce the effect of side-lobes that may appear as result of these surfaces. We reduce the reflected waves from the surfaces by designing the surfaces using specific materials or by changing the placement of both the transmitter and/or receiver antenna. Moreover, the resulting tomographic image will be processed using the Winner filter to remove any remaining noise.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115497329","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268782
Ayesha Zaman, Weisong Wang, G. Subramanyam
A robust physical model governing trap assisted inelastic tunneling mechanism for the conducting area of a memristor device with an intermediate thickness dielectric (<100nm) has been developed. This model takes into account, the contribution of both the oxygen vacancies and the defects within the oxide responsible for the significant increase of charge carriers through this non-volatile memory device. A one dimensional simulator has been designed here that shows reasonable compatibility between the practically obtained values with that of the simulated ones. Here tunneling of electrons between variable energy states of the different layer materials within the device, dominates the switching mechanism that involves a strong non-linearity of the electric field dependence.
{"title":"Modeling of memristor device & analysis of stability issues","authors":"Ayesha Zaman, Weisong Wang, G. Subramanyam","doi":"10.1109/NAECON.2017.8268782","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268782","url":null,"abstract":"A robust physical model governing trap assisted inelastic tunneling mechanism for the conducting area of a memristor device with an intermediate thickness dielectric (<100nm) has been developed. This model takes into account, the contribution of both the oxygen vacancies and the defects within the oxide responsible for the significant increase of charge carriers through this non-volatile memory device. A one dimensional simulator has been designed here that shows reasonable compatibility between the practically obtained values with that of the simulated ones. Here tunneling of electrons between variable energy states of the different layer materials within the device, dominates the switching mechanism that involves a strong non-linearity of the electric field dependence.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114191725","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268766
A. Mohamed, Monish R. Chatterjee
Electromagnetic propagation through MVKS-type turbulence is examined assuming a slanted path using the Hufnagel-Valley model for the structure parameter. The propagation path consists of turbulent (Lt) and non-turbulent (LD) zones; using a split-step approach, the diffracted field characteristics for Gaussian-profile and Bessel beams are examined under varying turbulence strengths, and Lt/ Ld ratios. Cross-correlation (CC) products are computed as performance measures relative to non-turbulent recovery.
{"title":"Slanted electromagnetic wave propagation through atmospheric phase turbulence using altitude-dependent structure parameter","authors":"A. Mohamed, Monish R. Chatterjee","doi":"10.1109/NAECON.2017.8268766","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268766","url":null,"abstract":"Electromagnetic propagation through MVKS-type turbulence is examined assuming a slanted path using the Hufnagel-Valley model for the structure parameter. The propagation path consists of turbulent (Lt) and non-turbulent (LD) zones; using a split-step approach, the diffracted field characteristics for Gaussian-profile and Bessel beams are examined under varying turbulence strengths, and Lt/ Ld ratios. Cross-correlation (CC) products are computed as performance measures relative to non-turbulent recovery.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"29 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126935608","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268769
D. Megherbi, P. Mack, J. DiZoglio, M. I. Vakil, N. Limberopoulos, A. Urbas
In our prior work we have analyzed, proposed global performance metrics, and shown how MWIR microsphere lens-enhanced single detectors have exhibited an increase in detector sensitivity and reduction in Noise to Signal Ratio (NSR). We have shown that, in general, the microsphere lens reduces the detector NSR independently of the increase in the enhanced detector sensitivity. We have also shown how microsphere lens material absorptions and misalignments can affect and reduce the microsphere lens-enhanced MWIR detector sensitivity. When placing a microsphere lens on a photo-detector, usually an optical silicone-or rubber adhesive is used to adhere and hold the microsphere lens on the single photo-detector. In this paper, we present a novel hybrid model-based and experimental data driven-based technique for analysis and characterization of the effect of the adhesive material on microsphere-lens-enhanced SLS single photo-detector sensitivity. In this work these adhesive material characteristics are modeled and compared to experimental microsphere-enhanced single detector FTIR spectral data. As we demonstrate here, based on the data considered, the results show that the adhesive material appears to adversely affect and decrease the detector overall sensitivity, and in some instances, by more than 40% depending on the adhesive material index of refraction, even with a high adhesive material transmittance/transparency.
{"title":"Using FTIR spectral response signals to separate, characterize and quantify the effects of silicone or rubber-based adhesive materials on microsphere lens enhanced MWIR SLS photo detectors","authors":"D. Megherbi, P. Mack, J. DiZoglio, M. I. Vakil, N. Limberopoulos, A. Urbas","doi":"10.1109/NAECON.2017.8268769","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268769","url":null,"abstract":"In our prior work we have analyzed, proposed global performance metrics, and shown how MWIR microsphere lens-enhanced single detectors have exhibited an increase in detector sensitivity and reduction in Noise to Signal Ratio (NSR). We have shown that, in general, the microsphere lens reduces the detector NSR independently of the increase in the enhanced detector sensitivity. We have also shown how microsphere lens material absorptions and misalignments can affect and reduce the microsphere lens-enhanced MWIR detector sensitivity. When placing a microsphere lens on a photo-detector, usually an optical silicone-or rubber adhesive is used to adhere and hold the microsphere lens on the single photo-detector. In this paper, we present a novel hybrid model-based and experimental data driven-based technique for analysis and characterization of the effect of the adhesive material on microsphere-lens-enhanced SLS single photo-detector sensitivity. In this work these adhesive material characteristics are modeled and compared to experimental microsphere-enhanced single detector FTIR spectral data. As we demonstrate here, based on the data considered, the results show that the adhesive material appears to adversely affect and decrease the detector overall sensitivity, and in some instances, by more than 40% depending on the adhesive material index of refraction, even with a high adhesive material transmittance/transparency.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"57 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114131380","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268765
Patrick R. Kennedy, T. Laurvick
The objective for this research is to determine a relationship between plasmonic grating geometries and the wavelength-dependent focus depth. This research is focused on enhancing the signal collected by infrared detectors by using a metal grating as a planar lens to focus light in the detecting region of the substrate. This can be used to maintain a thinner absorbing region and possibly to create multi-color imaging in a single pixel.
{"title":"Passive and active sensing — Plasmonic grating geometries and wavelength-dependent focus depth in IR detectors","authors":"Patrick R. Kennedy, T. Laurvick","doi":"10.1109/NAECON.2017.8268765","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268765","url":null,"abstract":"The objective for this research is to determine a relationship between plasmonic grating geometries and the wavelength-dependent focus depth. This research is focused on enhancing the signal collected by infrared detectors by using a metal grating as a planar lens to focus light in the detecting region of the substrate. This can be used to maintain a thinner absorbing region and possibly to create multi-color imaging in a single pixel.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"2014 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130066270","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268793
Turki M. Alanazi, James Reed, Hamdi Abdelbaagi, Abdunaser Abdusamad, Abdulhakim Daluom, Mufatah Akroush, Hamza Ekraash, M. Wicks
The goal of this research is to develop a method that allows for processing of bistatic marine radar signals, in order to demonstrate an improvement in target detection (Pd) and false alarm control (Pfa) in systems limited by cross-correlated interference. In this work, a method is presented for coherent processing of signals from a bistatic magnetron oscillator based marine radar. The feasibility of this approach was previously demonstrated for a monostatic radar through a hardware modification that allowed for capture of data via a Xilinx ADC and processing in FPGAs. It is demonstrated here that operating two of radars in this manner and combining their resulting signals allows for an improvement in overall detection and track. Our approach works by sampling the transmitted and received signals at each radar. Cross-correlations between all four combinations of transmitted and received signals are used to demonstrate the limits due to mutual interference in a bistatic/multistatic system of radars. This processing is successfully demonstrated in software, showing the potential for coherency between two marine radars. In general, bistatic coherent radars are very expensive, and this work provides a method for achieving the equivalent coherent performance using a two modified non-coherent radar systems.
{"title":"Demonstrating the limitations on target detection (Pd) and false alarm control (Pfa) due to cross-correlated interference in a bistatic / multistatic radar","authors":"Turki M. Alanazi, James Reed, Hamdi Abdelbaagi, Abdunaser Abdusamad, Abdulhakim Daluom, Mufatah Akroush, Hamza Ekraash, M. Wicks","doi":"10.1109/NAECON.2017.8268793","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268793","url":null,"abstract":"The goal of this research is to develop a method that allows for processing of bistatic marine radar signals, in order to demonstrate an improvement in target detection (Pd) and false alarm control (Pfa) in systems limited by cross-correlated interference. In this work, a method is presented for coherent processing of signals from a bistatic magnetron oscillator based marine radar. The feasibility of this approach was previously demonstrated for a monostatic radar through a hardware modification that allowed for capture of data via a Xilinx ADC and processing in FPGAs. It is demonstrated here that operating two of radars in this manner and combining their resulting signals allows for an improvement in overall detection and track. Our approach works by sampling the transmitted and received signals at each radar. Cross-correlations between all four combinations of transmitted and received signals are used to demonstrate the limits due to mutual interference in a bistatic/multistatic system of radars. This processing is successfully demonstrated in software, showing the potential for coherency between two marine radars. In general, bistatic coherent radars are very expensive, and this work provides a method for achieving the equivalent coherent performance using a two modified non-coherent radar systems.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116296797","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268798
T. Schultz, R. Jha
This paper discusses attack vulnerabilities in Resistive Random Access Memory (ReRAM) devices that can be potentially triggered by Trojan circuits. Systematic experiments were performed on Ru/MgO/Ti/W based ReRAM devices in 1R and 1T1R configurations. Our observations indicate that vulnerabilities in ReRAM include destabilization of filament due to current overshoot, local heating, introduction of parasitic capacitances, and glitches in the source voltages. These studies are critical to design more robust ReRAM devices that are immune to hardware attacks.
{"title":"Understanding vulnerabilities in ReRAM devices for trust in semiconductor designs","authors":"T. Schultz, R. Jha","doi":"10.1109/NAECON.2017.8268798","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268798","url":null,"abstract":"This paper discusses attack vulnerabilities in Resistive Random Access Memory (ReRAM) devices that can be potentially triggered by Trojan circuits. Systematic experiments were performed on Ru/MgO/Ti/W based ReRAM devices in 1R and 1T1R configurations. Our observations indicate that vulnerabilities in ReRAM include destabilization of filament due to current overshoot, local heating, introduction of parasitic capacitances, and glitches in the source voltages. These studies are critical to design more robust ReRAM devices that are immune to hardware attacks.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131512870","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 : 2017-06-01DOI: 10.1109/NAECON.2017.8268804
Daniel W. Steeneck, Trevor J. Bihl
Herein the authors apply the stochastic approximation method of Kiefer and Wolfowitz to optimize learning rate selection for Generalized Relevance Learning Vector Quantization — Improved (GRLVQI) neural networks with application to Z-Wave cyber-physical device identification. Recent work on full factorial models for GRLVQI optimal settings has shown promise, but is computationally costly and not feasible for large datasets. Results using stochastic optimization illustrate show fast convergence to high classification rates.
{"title":"Stochastic approximation for learning rate optimization for generalized relevance learning vector quantization","authors":"Daniel W. Steeneck, Trevor J. Bihl","doi":"10.1109/NAECON.2017.8268804","DOIUrl":"https://doi.org/10.1109/NAECON.2017.8268804","url":null,"abstract":"Herein the authors apply the stochastic approximation method of Kiefer and Wolfowitz to optimize learning rate selection for Generalized Relevance Learning Vector Quantization — Improved (GRLVQI) neural networks with application to Z-Wave cyber-physical device identification. Recent work on full factorial models for GRLVQI optimal settings has shown promise, but is computationally costly and not feasible for large datasets. Results using stochastic optimization illustrate show fast convergence to high classification rates.","PeriodicalId":306091,"journal":{"name":"2017 IEEE National Aerospace and Electronics Conference (NAECON)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116337764","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：