Pub Date : 2022-08-17DOI: 10.1109/TSIPI.2022.3199331
Francesco de Paulis;Muhammet Hilmi Nisanci
The use of metallic pins covering the lid of a metallic cavity has been shown to effectively suppress the coupling mechanisms based on the cavity resonances. However, no clear evidence is available on the effectiveness of this solution for the signal transmission over microstrip interconnects routed on substrates inside the cavity. A comprehensive analysis is carried out to fill this gap by analyzing the signal propagation on single-ended and differential microstrip, thus demonstrating that the pins help to minimize the detrimental impact of the resonating cavity within the bandgap limits. The effectiveness of the pinned cavity to suppress the coupling among microstrips routed on the same substrate is demonstrated. Experimental data based on a pinned cavity with different pin lengths are provided to confirm that the intended bandgap is properly achieved and that the quality of the transmitting signal is ensured within the bandgap.
{"title":"Signal Integrity Assessment of Interconnects Routed Within Bandgap Metallic Cavities","authors":"Francesco de Paulis;Muhammet Hilmi Nisanci","doi":"10.1109/TSIPI.2022.3199331","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3199331","url":null,"abstract":"The use of metallic pins covering the lid of a metallic cavity has been shown to effectively suppress the coupling mechanisms based on the cavity resonances. However, no clear evidence is available on the effectiveness of this solution for the signal transmission over microstrip interconnects routed on substrates inside the cavity. A comprehensive analysis is carried out to fill this gap by analyzing the signal propagation on single-ended and differential microstrip, thus demonstrating that the pins help to minimize the detrimental impact of the resonating cavity within the bandgap limits. The effectiveness of the pinned cavity to suppress the coupling among microstrips routed on the same substrate is demonstrated. Experimental data based on a pinned cavity with different pin lengths are provided to confirm that the intended bandgap is properly achieved and that the quality of the transmitting signal is ensured within the bandgap.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"83-92"},"PeriodicalIF":0.0,"publicationDate":"2022-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/9745882/9770004/09858329.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-17DOI: 10.1109/TSIPI.2022.3199178
Herbert Hackl;David J. Pommerenke;Martin Ibel;Bernhard Auinger
Batteries are a fundamental part of many modern electric systems. As a result, battery modeling is increasingly important for the prediction of signal and power integrity (SIPI) as well as electromagnetic compatibility (EMC). Conventional battery module modeling requires knowledge of the integrated cells first, which is usually obtained by measurement on single cells. However, if individual cells are not accessible, the single cell's impedance needs to be extracted from measurement of the complete module. This work describes two solutions to this problem, which are both based on 3D electromagnetic (EM) simulation of the battery module with surrogate cell models to obtain S-parameters, which describe coupling effects inherent to the modules' geometry. By either fitting of the simulated module impedance to the measured data on circuit schematic level, or by numerical multiport de-embedding, the single cell impedance is extracted. Considered frequencies range from 9 kHz to 1 GHz.
{"title":"Extraction of Single Cell Impedance From Battery Module Measurement by Simulation-Based De-Embedding","authors":"Herbert Hackl;David J. Pommerenke;Martin Ibel;Bernhard Auinger","doi":"10.1109/TSIPI.2022.3199178","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3199178","url":null,"abstract":"Batteries are a fundamental part of many modern electric systems. As a result, battery modeling is increasingly important for the prediction of signal and power integrity (SIPI) as well as electromagnetic compatibility (EMC). Conventional battery module modeling requires knowledge of the integrated cells first, which is usually obtained by measurement on single cells. However, if individual cells are not accessible, the single cell's impedance needs to be extracted from measurement of the complete module. This work describes two solutions to this problem, which are both based on 3D electromagnetic (EM) simulation of the battery module with surrogate cell models to obtain S-parameters, which describe coupling effects inherent to the modules' geometry. By either fitting of the simulated module impedance to the measured data on circuit schematic level, or by numerical multiport de-embedding, the single cell impedance is extracted. Considered frequencies range from 9 kHz to 1 GHz.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"112-120"},"PeriodicalIF":0.0,"publicationDate":"2022-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842401","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 : 2022-08-08DOI: 10.1109/TSIPI.2022.3197110
Ihsan Erdin;Ramachandra Achar
A domain decomposition method is proposed to evaluate the effectiveness of decoupling capacitors in practical power delivery networks (PDNs). The proposed method is based on the separation of a PDN into its local and nonlocal domains. The local domain is constituted by circuit components with the highest impact on the impedance of a specified power pin on a planar PDN. The rest of the PDN makes up the nonlocal domain, which could be of any planar shape. The nonlocal domain is characterized as a distributed circuit, preferably using a numerical electromagnetic (EM) simulator. The self-impedance of the pin depends on the placement configuration of capacitors in its surroundings. Using the pin impedance as a figure of merit, the optimal placement configuration is then sought in the local domain. The impact of the stationary domain is included in calculations as an indefinite impedance and the optimal placement configuration of capacitors is computed using an iterative approach. The proposed method avoids the use of a computationally intensive EM simulation at each iteration step, which significantly speeds up the analysis process.
{"title":"A Domain Decomposition Approach for Assessment of Decoupling Capacitors in Practical PDNs","authors":"Ihsan Erdin;Ramachandra Achar","doi":"10.1109/TSIPI.2022.3197110","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3197110","url":null,"abstract":"A domain decomposition method is proposed to evaluate the effectiveness of decoupling capacitors in practical power delivery networks (PDNs). The proposed method is based on the separation of a PDN into its local and nonlocal domains. The local domain is constituted by circuit components with the highest impact on the impedance of a specified power pin on a planar PDN. The rest of the PDN makes up the nonlocal domain, which could be of any planar shape. The nonlocal domain is characterized as a distributed circuit, preferably using a numerical electromagnetic (EM) simulator. The self-impedance of the pin depends on the placement configuration of capacitors in its surroundings. Using the pin impedance as a figure of merit, the optimal placement configuration is then sought in the local domain. The impact of the stationary domain is included in calculations as an indefinite impedance and the optimal placement configuration of capacitors is computed using an iterative approach. The proposed method avoids the use of a computationally intensive EM simulation at each iteration step, which significantly speeds up the analysis process.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"55-64"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842425","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 : 2022-07-27DOI: 10.1109/TSIPI.2022.3193871
Nobuo Kuwabara;Tohlu Matsushima;Yuki Fukumoto
Radiated emission from a differential-mode (DM) signal was calculated by using a chain parameter matrix for a shielded twisted pair (STP) cable. The series connection of the matrix represented the multiconductor transmission line inside the shield, and the common-mode current generated from the DM signal was obtained by solving the matrix equation. The matrix elements of the STP cable and the shield's transfer impedance were determined through both measurement and calculation, and these values were used to calculate the current outside the shield and the radiated electric field strength. The maximum current outside the shield and the maximum radiated electric field strength were then measured in the frequency range from 30 to 300 MHz, and these results were compared with the calculated values. The calculation results for the maximum current intensity agreed well with the measured values, and those for the maximum radiated electric field strength agreed fairly well with the measured values.
{"title":"Analysis of Differential-Mode Signal Emission From STP Cable by Using Chain Parameter Matrix","authors":"Nobuo Kuwabara;Tohlu Matsushima;Yuki Fukumoto","doi":"10.1109/TSIPI.2022.3193871","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3193871","url":null,"abstract":"Radiated emission from a differential-mode (DM) signal was calculated by using a chain parameter matrix for a shielded twisted pair (STP) cable. The series connection of the matrix represented the multiconductor transmission line inside the shield, and the common-mode current generated from the DM signal was obtained by solving the matrix equation. The matrix elements of the STP cable and the shield's transfer impedance were determined through both measurement and calculation, and these values were used to calculate the current outside the shield and the radiated electric field strength. The maximum current outside the shield and the maximum radiated electric field strength were then measured in the frequency range from 30 to 300 MHz, and these results were compared with the calculated values. The calculation results for the maximum current intensity agreed well with the measured values, and those for the maximum radiated electric field strength agreed fairly well with the measured values.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"74-82"},"PeriodicalIF":0.0,"publicationDate":"2022-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842427","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}
A frequency-dependent and per-port (FDPP) channel termination and renormalization method is presented as a more accurate substitute for the traditional method that uses frequency independent, uniform impedance as the serialization/deserialization transmit (TX)/receive (RX) termination impedances. Although the traditional method is employed in practically all the existing high-speed interface standards, the assumption of the constant, uniform termination impedance at TX and RX is not exactly correct and will lead to inaccuracy. The new method characterizes the mismatch conditions at the TX and the RX terminations on a per frequency point and per-port basis and, thus, produces more accurate results during the channel characterization and performance assessment. The mathematical computation procedure of the FDPP approach is introduced. Two representative TX/RX terminations and three typical channels representing mid/high/ultra-high losses are used to demonstrate the effects of the FDPP renormalization method. The obtained S -parameters, eye-diagrams, as well as the channel operational margin (COM) figure of merits are compared to that generated using the traditional method. The comparison shows that the FDPP method produces more realistic and accurate results than the traditional approach. In the end, a proposal of expanding the [Reference] section in the current Touchstone 2.0 format is presented.
{"title":"Frequency-Dependent Per-Port Renormalization","authors":"Sherman Shan Chen;Zhifei Xu;Armin Tajalli;Brian Holden","doi":"10.1109/TSIPI.2022.3188949","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3188949","url":null,"abstract":"A frequency-dependent and per-port (FDPP) channel termination and renormalization method is presented as a more accurate substitute for the traditional method that uses frequency independent, uniform impedance as the serialization/deserialization transmit (TX)/receive (RX) termination impedances. Although the traditional method is employed in practically all the existing high-speed interface standards, the assumption of the constant, uniform termination impedance at TX and RX is not exactly correct and will lead to inaccuracy. The new method characterizes the mismatch conditions at the TX and the RX terminations on a per frequency point and per-port basis and, thus, produces more accurate results during the channel characterization and performance assessment. The mathematical computation procedure of the FDPP approach is introduced. Two representative TX/RX terminations and three typical channels representing mid/high/ultra-high losses are used to demonstrate the effects of the FDPP renormalization method. The obtained \u0000<italic>S</i>\u0000-parameters, eye-diagrams, as well as the channel operational margin (COM) figure of merits are compared to that generated using the traditional method. The comparison shows that the FDPP method produces more realistic and accurate results than the traditional approach. In the end, a proposal of expanding the \u0000<bold>[Reference]</b>\u0000 section in the current Touchstone 2.0 format is presented.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"65-73"},"PeriodicalIF":0.0,"publicationDate":"2022-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842426","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 : 2022-06-01DOI: 10.1109/TSIPI.2022.3179250
Fabrizio Loreto;Daniele Romano;Martin Štumpf;Albert E. Ruehli;Giulio Antonini
The partial inductance is a very well known concept in electromagnetic modeling that allows us to ascribe the properties of inductance to an isolated piece of conductor and of a mutual inductance to a couple of finite-size conductors, not necessarily constituting a closed loop as it is required for the standard concept of inductance. Although, its computation has been widely studied in the static case and in the frequency domain for the dynamic case, its computation in the time domain (TD) has been only partially addressed. This article aims to fill this gap also pointing out their use in the framework of a TD solver. In particular, the modified numerical inversion of the Laplace transform (NILT) is adopted to compute the time samples of the partial inductance avoiding the cumbersome inverse Fourier transform (IFT). It will be shown that, in addition to the high accuracy, the delayed implementation of the NILT method strictly preserves the causality of the magnetic coupling. Furthermore, the use of Hermite interpolation allows us to significantly reduce the computational effort. The proposed method is tested by comparison with analytical formulas existing for coplanar zero-thickness regions and with IFT techniques for the both orthogonal and nonorthogonal geometries.
{"title":"Time-Domain Computation of Full-Wave Partial Inductances Based on the Modified Numerical Inversion of Laplace Transform Method","authors":"Fabrizio Loreto;Daniele Romano;Martin Štumpf;Albert E. Ruehli;Giulio Antonini","doi":"10.1109/TSIPI.2022.3179250","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3179250","url":null,"abstract":"The partial inductance is a very well known concept in electromagnetic modeling that allows us to ascribe the properties of inductance to an isolated piece of conductor and of a mutual inductance to a couple of finite-size conductors, not necessarily constituting a closed loop as it is required for the standard concept of inductance. Although, its computation has been widely studied in the static case and in the frequency domain for the dynamic case, its computation in the time domain (TD) has been only partially addressed. This article aims to fill this gap also pointing out their use in the framework of a TD solver. In particular, the modified numerical inversion of the Laplace transform (NILT) is adopted to compute the time samples of the partial inductance avoiding the cumbersome inverse Fourier transform (IFT). It will be shown that, in addition to the high accuracy, the delayed implementation of the NILT method strictly preserves the causality of the magnetic coupling. Furthermore, the use of Hermite interpolation allows us to significantly reduce the computational effort. The proposed method is tested by comparison with analytical formulas existing for coplanar zero-thickness regions and with IFT techniques for the both orthogonal and nonorthogonal geometries.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"32-42"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842423","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 : 2022-04-19DOI: 10.1109/TSIPI.2022.3167937
Yanming Zhang;Lijun Jiang
This article presents a novel hybrid knowledge-based and data-driven scheme to determine the governing partial differential equation (PDE) for the transmission line (TL). The two-dimensional (2-D) current and voltage distributions along the TL are used as the input data. The hypothetical functions, including the candidate terms that may appear in the telegraphic equations, are built based on prior knowledge of the TL system of interest. Through the spatial and temporal derivatives performed on 2-D current and voltage data, the governing PDEs of TLs are solely represented by the linear algebraic equations. The ridge regression is employed to ascertain the actual PDEs of TLs via extracting the active terms from hypothetical functions. The accuracy and effectiveness of this approach are demonstrated through three benchmarked examples, including the lossy, nonuniform, and nonlinear TLs. The results verify that the proposed scheme can inverse the per-unit-length (p.-u.-l.) parameters and identify the governing PDEs. Our work offers a helpful technique to establish connections between observation and the theoretical TL model.
{"title":"Modeling Transmission Lines Using a Hybrid Knowledge-Based and Data-Driven Approach","authors":"Yanming Zhang;Lijun Jiang","doi":"10.1109/TSIPI.2022.3167937","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3167937","url":null,"abstract":"This article presents a novel hybrid knowledge-based and data-driven scheme to determine the governing partial differential equation (PDE) for the transmission line (TL). The two-dimensional (2-D) current and voltage distributions along the TL are used as the input data. The hypothetical functions, including the candidate terms that may appear in the telegraphic equations, are built based on prior knowledge of the TL system of interest. Through the spatial and temporal derivatives performed on 2-D current and voltage data, the governing PDEs of TLs are solely represented by the linear algebraic equations. The ridge regression is employed to ascertain the actual PDEs of TLs via extracting the active terms from hypothetical functions. The accuracy and effectiveness of this approach are demonstrated through three benchmarked examples, including the lossy, nonuniform, and nonlinear TLs. The results verify that the proposed scheme can inverse the per-unit-length (p.-u.-l.) parameters and identify the governing PDEs. Our work offers a helpful technique to establish connections between observation and the theoretical TL model.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"12-21"},"PeriodicalIF":0.0,"publicationDate":"2022-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842420","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}
This article presents an efficient methodology based on boundary integration to calculate the dc and ac impedance of power distribution networks (PDNs) for arbitrary-shape and multilayer printed circuit boards (PCBs). The proposed method adopts a boundary element method to extract inductances for the arbitrary parallel-plane shapes. Subsequently, the equivalent circuit formed by the via inductances and plane capacitances is solved using the node voltage method. By merging the parallel and serial inductances and simplifying the equivalent circuit, the computation time can be significantly reduced for a large number of vias and layers. This matrix manipulation strategy can be applied to various PCB structures without human intervention or commercial circuit solvers. Moreover, a contour integral method is employed to calculate the dc resistances for arbitrary shape and multilayer PDNs. Therefore, the wideband PDN impedance from dc to ac frequencies can be efficiently calculated through 1-D boundary integration, which can be performed more quickly than full-wave simulations. The proposed method can aid in the development of electronic design automation tools for PDN design.
{"title":"Efficient DC and AC Impedance Calculation for Arbitrary-Shape and Multilayer PDN Using Boundary Integration","authors":"Ling Zhang;Jack Juang;Zurab Kiguradze;Bo Pu;Shuai Jin;Songping Wu;Zhiping Yang;Er-Ping Li;Jun Fan;Chulsoon Hwang","doi":"10.1109/TSIPI.2022.3164037","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3164037","url":null,"abstract":"This article presents an efficient methodology based on boundary integration to calculate the dc and ac impedance of power distribution networks (PDNs) for arbitrary-shape and multilayer printed circuit boards (PCBs). The proposed method adopts a boundary element method to extract inductances for the arbitrary parallel-plane shapes. Subsequently, the equivalent circuit formed by the via inductances and plane capacitances is solved using the node voltage method. By merging the parallel and serial inductances and simplifying the equivalent circuit, the computation time can be significantly reduced for a large number of vias and layers. This matrix manipulation strategy can be applied to various PCB structures without human intervention or commercial circuit solvers. Moreover, a contour integral method is employed to calculate the dc resistances for arbitrary shape and multilayer PDNs. Therefore, the wideband PDN impedance from dc to ac frequencies can be efficiently calculated through 1-D boundary integration, which can be performed more quickly than full-wave simulations. The proposed method can aid in the development of electronic design automation tools for PDN design.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"1-11"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842421","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 : 2022-03-23DOI: 10.1109/TSIPI.2022.3176592
Muqi Ouyang;Xiao-Ding Cai;Bo Pu;Qian Gao;Srinath Penugonda;Chaofeng Li;Bidyut Sen;Chulsoon Hwang;DongHyun Kim
Reflection theory has been long established for over decades targeted at microwave and radio frequency (RF) applications. With ultra-high-bandwidth applications emerging, such as 112 Gb/s and higher speed Ethernet protocols, discontinuities in high-speed channels negatively impact signal quality, where reflections become one of the most critical concerns in high-speed designs. In this article, for the first time, we analyzed the traditional reflection theory and proposed and verified a new formulation, which exhibits the reflection-related parameters explicitly, indicating where design optimization can be made for high-bandwidth applications using the backtracked propagation method. Our closed-form formulation is applied to high-speed channel examples, where effective mitigation of negative impact from reflections on signal integrity can be identified to be used as a prelayout channel design guide. Our proposed formulation of the reflection theory provides more accurate prediction of high-speed channel behavior to minimize the negative signal integrity impact from reflections.
{"title":"Novel Formulations of Multireflections and Their Applications to High-Speed Channel Design","authors":"Muqi Ouyang;Xiao-Ding Cai;Bo Pu;Qian Gao;Srinath Penugonda;Chaofeng Li;Bidyut Sen;Chulsoon Hwang;DongHyun Kim","doi":"10.1109/TSIPI.2022.3176592","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3176592","url":null,"abstract":"Reflection theory has been long established for over decades targeted at microwave and radio frequency (RF) applications. With ultra-high-bandwidth applications emerging, such as 112 Gb/s and higher speed Ethernet protocols, discontinuities in high-speed channels negatively impact signal quality, where reflections become one of the most critical concerns in high-speed designs. In this article, for the first time, we analyzed the traditional reflection theory and proposed and verified a new formulation, which exhibits the reflection-related parameters explicitly, indicating where design optimization can be made for high-bandwidth applications using the backtracked propagation method. Our closed-form formulation is applied to high-speed channel examples, where effective mitigation of negative impact from reflections on signal integrity can be identified to be used as a prelayout channel design guide. Our proposed formulation of the reflection theory provides more accurate prediction of high-speed channel behavior to minimize the negative signal integrity impact from reflections.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"43-54"},"PeriodicalIF":0.0,"publicationDate":"2022-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842424","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 : 2022-03-10DOI: 10.1109/TSIPI.2022.3173592
Hanzhi Ma;Da Li;Er-Ping Li;Andreas C. Cangellaris;Xu Chen
A fast constrained design optimization technique is introduced for the inverse design of high-speed links. The proposed method makes use of the support-vector-regression-based active subspace (SVR-AS) method to generate a lower dimensional space of active variables as a linear weighted combination of the higher dimensional design space to help simplify the optimization problem. This newly developed technique successfully transforms the complex nonlinear constraint optimization problem into a linear constraint minimization problem and provides a directly solvable function for the optimal results associated with each active variable. Furthermore, the proposed optimization algorithm can be utilized for the optimization problem with equality and inequality constraints, which contain one- and multidimensional active variables generated from the SVR-AS method. Studies of two high-speed links with channel design parameters for nonreturn-to-zero pulse and IBIS-AMI equalization settings for four-level pulse amplitude modulation, respectively, are utilized to verify the efficiency of the method. The sensitivity analysis ability of the SVR-AS method is also presented for both the one- and multidimensional active variable cases.
{"title":"A Fast Optimization Method for High-Speed Link Inverse Design With SVR-AS Algorithm","authors":"Hanzhi Ma;Da Li;Er-Ping Li;Andreas C. Cangellaris;Xu Chen","doi":"10.1109/TSIPI.2022.3173592","DOIUrl":"https://doi.org/10.1109/TSIPI.2022.3173592","url":null,"abstract":"A fast constrained design optimization technique is introduced for the inverse design of high-speed links. The proposed method makes use of the support-vector-regression-based active subspace (SVR-AS) method to generate a lower dimensional space of active variables as a linear weighted combination of the higher dimensional design space to help simplify the optimization problem. This newly developed technique successfully transforms the complex nonlinear constraint optimization problem into a linear constraint minimization problem and provides a directly solvable function for the optimal results associated with each active variable. Furthermore, the proposed optimization algorithm can be utilized for the optimization problem with equality and inequality constraints, which contain one- and multidimensional active variables generated from the SVR-AS method. Studies of two high-speed links with channel design parameters for nonreturn-to-zero pulse and IBIS-AMI equalization settings for four-level pulse amplitude modulation, respectively, are utilized to verify the efficiency of the method. The sensitivity analysis ability of the SVR-AS method is also presented for both the one- and multidimensional active variable cases.","PeriodicalId":100646,"journal":{"name":"IEEE Transactions on Signal and Power Integrity","volume":"1 ","pages":"22-31"},"PeriodicalIF":0.0,"publicationDate":"2022-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67842416","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：