Pub Date : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193230
B. Silva, M. Eldessouki, Y. F. Shen
Simulation-measurement correlation presents unique challenges that require co-simulation between SI and PI to capture all sources of noise. A new hybrid methodology using peak distortion analysis and realistic worst case SSN was devised to enable full channel simulation including both SI and PI effects. The method was verified against lab measurements on two different designs.
{"title":"LPDDR4 SIPI Co-Simulation and Measurement Correlation for IOT Computer Vision Application","authors":"B. Silva, M. Eldessouki, Y. F. Shen","doi":"10.1109/EPEPS47316.2019.193230","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193230","url":null,"abstract":"Simulation-measurement correlation presents unique challenges that require co-simulation between SI and PI to capture all sources of noise. A new hybrid methodology using peak distortion analysis and realistic worst case SSN was devised to enable full channel simulation including both SI and PI effects. The method was verified against lab measurements on two different designs.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"150 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131433174","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193215
S. Mittal, Swagato Chakraborty
This paper presents an approach to quantify the effective resistance of a trace at DC by taking metal roughness into account. Algorithms were developed to approximate the resistance variation at 0Hz due to roughness and results were compared with hand calculations, 3D model simulations and measurement. The existing models in the industry are valid only for higher frequencies where skin effect is fully developed. They do not hold true for modeling at DC. In flex designs where the traces are very thin, the surface roughness parameter can be relatively large and can have a significant impact on signal transmission and power delivery of the system at DC.
{"title":"Modeling Surface Roughness at DC","authors":"S. Mittal, Swagato Chakraborty","doi":"10.1109/EPEPS47316.2019.193215","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193215","url":null,"abstract":"This paper presents an approach to quantify the effective resistance of a trace at DC by taking metal roughness into account. Algorithms were developed to approximate the resistance variation at 0Hz due to roughness and results were compared with hand calculations, 3D model simulations and measurement. The existing models in the industry are valid only for higher frequencies where skin effect is fully developed. They do not hold true for modeling at DC. In flex designs where the traces are very thin, the surface roughness parameter can be relatively large and can have a significant impact on signal transmission and power delivery of the system at DC.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115210454","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193220
K. Roy, M. A. Dolatsara, H. Torun, R. Trinchero, M. Swaminathan
Design of microwave structures and tuning parameters have mostly relied on the domain expertise of circuit designers by doing many simulations, which can be prohibitively time consuming. An inverse problem approach suggests going in the opposite direction to determine design parameters from characteristics of the desired output. In this work, we propose a novel machine learning architecture that circumvents usual design method for given quality of eye characteristics by means of a Lifelong Learning Architecture. Our proposed machine learning architecture is a large-scale coupled training system in which multiple predictions and classifications are done jointly for inverse mapping of transmission line geometry from eye characteristics. Our model is trained in a guided manner by using intra-tasks results, common Knowledge Base (KB), and coupling constraints. Our method of inverse design is general and can be applied to other applications.
{"title":"Inverse Design of Transmission Lines with Deep Learning","authors":"K. Roy, M. A. Dolatsara, H. Torun, R. Trinchero, M. Swaminathan","doi":"10.1109/EPEPS47316.2019.193220","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193220","url":null,"abstract":"Design of microwave structures and tuning parameters have mostly relied on the domain expertise of circuit designers by doing many simulations, which can be prohibitively time consuming. An inverse problem approach suggests going in the opposite direction to determine design parameters from characteristics of the desired output. In this work, we propose a novel machine learning architecture that circumvents usual design method for given quality of eye characteristics by means of a Lifelong Learning Architecture. Our proposed machine learning architecture is a large-scale coupled training system in which multiple predictions and classifications are done jointly for inverse mapping of transmission line geometry from eye characteristics. Our model is trained in a guided manner by using intra-tasks results, common Knowledge Base (KB), and coupling constraints. Our method of inverse design is general and can be applied to other applications.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129701445","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193197
H. Barnes, J. Moreira
Analysis of improving simulation to measurement correlation of a microstrip patch antenna using as-fabricated PCB material properties. Prior work has shown the benefit of using a Beatty series resonant transmission line structure to estimate as-fabricated material properties. Results of this investigation show that surface roughness has a significant impact on the effective dielectric constant for the Beatty structure and for the patch antenna.
{"title":"Microstrip Antenna EM Model Tuning using the Beatty Structure","authors":"H. Barnes, J. Moreira","doi":"10.1109/EPEPS47316.2019.193197","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193197","url":null,"abstract":"Analysis of improving simulation to measurement correlation of a microstrip patch antenna using as-fabricated PCB material properties. Prior work has shown the benefit of using a Beatty series resonant transmission line structure to estimate as-fabricated material properties. Results of this investigation show that surface roughness has a significant impact on the effective dielectric constant for the Beatty structure and for the patch antenna.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126843369","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193199
Hanzhi Ma, Erping Li, A. Cangellaris, Xu Chen
We compare three different machine learning techniques for constructing predictive model for eye opening based on channel length and interconnect cross-sectional geometry. Surrogate model is constructed using sparse grids, support vector regression, and artificial neural networks. Models for training data are generated using quasi-TEM modeling of the interconnect, and eye opening training data is obtained from statistical high-speed link simulation using IBIS-AMI transmitter and receiver models. Numerical results illustrate that all three methods offer reasonable predictions of eye height, eye width and eye width at 10−12 bit error rate.
{"title":"Comparison of Machine Learning Techniques for Predictive Modeling of High-Speed Links","authors":"Hanzhi Ma, Erping Li, A. Cangellaris, Xu Chen","doi":"10.1109/EPEPS47316.2019.193199","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193199","url":null,"abstract":"We compare three different machine learning techniques for constructing predictive model for eye opening based on channel length and interconnect cross-sectional geometry. Surrogate model is constructed using sparse grids, support vector regression, and artificial neural networks. Models for training data are generated using quasi-TEM modeling of the interconnect, and eye opening training data is obtained from statistical high-speed link simulation using IBIS-AMI transmitter and receiver models. Numerical results illustrate that all three methods offer reasonable predictions of eye height, eye width and eye width at 10−12 bit error rate.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"106 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131043072","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193218
Shu Wang, Z. Peng
This paper investigates a space-time building block methodology for efficient time-domain analysis of multi-scale, locally periodic structures. By leveraging the principles of linear superposition and space-time causality in wave physics, the 4D simulation domain is represented by a few space-time building blocks, which are constructed upon 3D spatial unit cell and 1D time unit. The work results in novel time-evolution schemes, which exhibit high-order accuracy and achieve concurrency and parallelism in both spatial and temporal dimensions. The outcomes lead to significantly reduced computational resources. The capability and benefit of proposed algorithms are illustrated through representative applications.
{"title":"A Novel Space-Time Building Block Methodology for Transient Electromagnetic Analysis","authors":"Shu Wang, Z. Peng","doi":"10.1109/EPEPS47316.2019.193218","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193218","url":null,"abstract":"This paper investigates a space-time building block methodology for efficient time-domain analysis of multi-scale, locally periodic structures. By leveraging the principles of linear superposition and space-time causality in wave physics, the 4D simulation domain is represented by a few space-time building blocks, which are constructed upon 3D spatial unit cell and 1D time unit. The work results in novel time-evolution schemes, which exhibit high-order accuracy and achieve concurrency and parallelism in both spatial and temporal dimensions. The outcomes lead to significantly reduced computational resources. The capability and benefit of proposed algorithms are illustrated through representative applications.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129924633","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193221
P. Paladhi, P. Jayaraman, N. Pham, J. Hejase, Yanyan Zhang, Junyan Tang, J. Myers, S. Chun, W. Becker, D. Dreps
While equalization is usually a positively contributing factor towards opening eye diagrams further, sometimes over equalization can occur and degrade an eye opening as opposed to improve it. This paper explores the effectiveness of using post-cursors FFE transmitter equalization on 25.78Gb/s communication links with different transmitter and receiver equalization. The effect is observed on two topologies with different loss levels and impedance profiles. Simulation and experimental results are presented and discussed.
{"title":"Effect and Sensitivity of Postcursor FFE in a 25 Gb/s High Speed Bus Channel","authors":"P. Paladhi, P. Jayaraman, N. Pham, J. Hejase, Yanyan Zhang, Junyan Tang, J. Myers, S. Chun, W. Becker, D. Dreps","doi":"10.1109/EPEPS47316.2019.193221","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193221","url":null,"abstract":"While equalization is usually a positively contributing factor towards opening eye diagrams further, sometimes over equalization can occur and degrade an eye opening as opposed to improve it. This paper explores the effectiveness of using post-cursors FFE transmitter equalization on 25.78Gb/s communication links with different transmitter and receiver equalization. The effect is observed on two topologies with different loss levels and impedance profiles. Simulation and experimental results are presented and discussed.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117216122","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193201
N. Elumalai, Srinidhi Ganeshan, R. Achar
Vector Fitting (VF) algorithm has been widely used for system identification from multiport tabulated data. Particularly, it is of high interest to the design community focused on modeling of high-speed modules such as large number of coupled interconnects, packaging structures and variety of electromagnetic modules. This paper advances the applicability of VF to exploit the emerging massively parallel graphical processing Units (GPUs). Necessary parallelization strategies suitable for GPU platforms are developed. For large problem sizes (increasing number of ports and poles), numerical results demonstrate that the proposed method provides significant speedup compared to both the single CPU based VF as well as existing multi-CPU based parallel VF techniques depending on the number of cores used.
{"title":"GVF: GPU based Vector Fitting","authors":"N. Elumalai, Srinidhi Ganeshan, R. Achar","doi":"10.1109/EPEPS47316.2019.193201","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193201","url":null,"abstract":"Vector Fitting (VF) algorithm has been widely used for system identification from multiport tabulated data. Particularly, it is of high interest to the design community focused on modeling of high-speed modules such as large number of coupled interconnects, packaging structures and variety of electromagnetic modules. This paper advances the applicability of VF to exploit the emerging massively parallel graphical processing Units (GPUs). Necessary parallelization strategies suitable for GPU platforms are developed. For large problem sizes (increasing number of ports and poles), numerical results demonstrate that the proposed method provides significant speedup compared to both the single CPU based VF as well as existing multi-CPU based parallel VF techniques depending on the number of cores used.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125002621","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193217
Ammar Aljamal, R. Gholami, Shucheng Zheng, V. Okhmatovski
A delta-gap source model is proposed for Surface-Volume-Surface Electric Field Integral Equation (SVS-EFIE) which is a new class of single source integral equation (SSIE). The proposed excitation model enables the use of SVS-EFIE for full-wave analysis of interconnects which accurately accounts for the loss in the conductors and allows for rigorous handling of the substrate multilayered medium. The delta-gap source model of SVS-EFIE is derived based on the conventional model that was formulated for the classic surface Electric Field Integral Equation (EFIE). The computation of the network parameters, however, requires computation of the electric fields at the ports volumetric cross-sections. The SVS-EFIE and the proposed source model are discretized using Method of Moments (MoM). Preliminary numerical results are provided for current distribution and the input impedance calculations of a dipole antenna. The input impedance of the dipole antenna computed using proposed excitation model is shown to agree well with that computed using classical surface EFIE provided the latter uses appropriate surface impedance model to account for the conductor loss.
{"title":"Delta-Gap Source Excitation Model in Surface-Volume-Surface Electric Field Integral Equation for 3-D Interconnect Characterization","authors":"Ammar Aljamal, R. Gholami, Shucheng Zheng, V. Okhmatovski","doi":"10.1109/EPEPS47316.2019.193217","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193217","url":null,"abstract":"A delta-gap source model is proposed for Surface-Volume-Surface Electric Field Integral Equation (SVS-EFIE) which is a new class of single source integral equation (SSIE). The proposed excitation model enables the use of SVS-EFIE for full-wave analysis of interconnects which accurately accounts for the loss in the conductors and allows for rigorous handling of the substrate multilayered medium. The delta-gap source model of SVS-EFIE is derived based on the conventional model that was formulated for the classic surface Electric Field Integral Equation (EFIE). The computation of the network parameters, however, requires computation of the electric fields at the ports volumetric cross-sections. The SVS-EFIE and the proposed source model are discretized using Method of Moments (MoM). Preliminary numerical results are provided for current distribution and the input impedance calculations of a dipole antenna. The input impedance of the dipole antenna computed using proposed excitation model is shown to agree well with that computed using classical surface EFIE provided the latter uses appropriate surface impedance model to account for the conductor loss.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121602385","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 : 2019-10-01DOI: 10.1109/EPEPS47316.2019.193205
G. Hernández‐Sosa, Se-jung Moon, X. Ye
In this work, we apply electromagnetic domain decomposition (EDD) to separate through-hole mounting (THM) connector models into two sub-domains: the connector body and the PCB section. The sub-domain models were simulated and then concatenated in order to get the whole connector response. Frequency and time-domain validation was performed by comparing the S-parameters and time-domain impedance profile corresponding to full-domain and concatenated models respectively, obtaining excellent correlation results.
{"title":"Electromagnetic Domain Decomposition for Through-Hole Mounting Connectors Models","authors":"G. Hernández‐Sosa, Se-jung Moon, X. Ye","doi":"10.1109/EPEPS47316.2019.193205","DOIUrl":"https://doi.org/10.1109/EPEPS47316.2019.193205","url":null,"abstract":"In this work, we apply electromagnetic domain decomposition (EDD) to separate through-hole mounting (THM) connector models into two sub-domains: the connector body and the PCB section. The sub-domain models were simulated and then concatenated in order to get the whole connector response. Frequency and time-domain validation was performed by comparing the S-parameters and time-domain impedance profile corresponding to full-domain and concatenated models respectively, obtaining excellent correlation results.","PeriodicalId":304228,"journal":{"name":"2019 IEEE 28th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126402045","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：