This book addresses the automatic sizing and layout of analog integrated circuits (ICs) using deep learning (DL) and artificial neural networks (ANN). It explores an innovative approach to automatic circuit sizing where ANNs learn patterns from previously optimized design solutions. In opposition to classical optimization-based sizing strategies, where computational intelligence techniques are used to iterate over the map from devices’ sizes to circuits’ performances provided by design equations or circuit simulations, ANNs are shown to be capable of solving analog IC sizing as a direct map from specifications to the devices’ sizes. Two separate ANN architectures are proposed: a Regression-only model and a Classification and Regression model. The goal of the Regression-only model is to learn design patterns from the studied circuits, using circuit’s performances as input features and devices’ sizes as target outputs. This model can size a circuit given its specifications for a single topology. The Classification and Regression model has the same capabilities of the previous model, but it can also select the most appropriate circuit topology and its respective sizing given the target specification. The proposed methodology was implemented and tested on two analog circuit topologies.
In this book, innovative research using artificial neural networks (ANNs) is conducted to automate the sizing task of RF IC design, which is used in two different steps of the automatic design process. The advances in telecommunications, such as the 5th generation broadband or 5G for short, open doors to advances in areas such as health care, education, resource management, transportation, agriculture and many other areas. Consequently, there is high pressure in today’s market for significant communication rates, extensive bandwidths and ultralow-power consumption. This is where radiofrequency (RF) integrated circuits (ICs) come in hand, playing a crucial role. This demand stresses out the problem which resides in the remarkable difficulty of RF IC design in deep nanometric integration technologies due to their high complexity and stringent performances. Given the economic pressure for high quality yet cheap electronics and challenging time-to-market constraints, there is an urgent need for electronic design automation (EDA) tools to increase the RF designers’ productivity and improve the quality of resulting ICs. In the last years, the automatic sizing of RF IC blocks in deep nanometer technologies has moved toward process, voltage and temperature (PVT)-inclusive optimizations to ensure their robustness. Each sizing solution is exhaustively simulated in a set of PVT corners, thus pushing modern workstations’ capabilities to their limits. Standard ANNs applications usually exploit the model’s capability of describing a complex, harder to describe, relation between input and target data. For that purpose, ANNs are a mechanism to bypass the process of describing the complex underlying relations between data by feeding it a significant number of previously acquired input/output data pairs that the model attempts to copy. Here, and firstly, the ANNs disrupt from the most recent trials of replacing the simulator in the simulation-based sizing with a machine/deep learning model, by proposing two different ANNs, the first classifies the convergence of the circuit for nominal and PVT corners, and the second predicts the oscillating frequencies for each case. The convergence classifier (CCANN) and frequency guess predictor (FGPANN) are seamlessly integrated into the simulation-based sizing loop, accelerating the overall optimization process. Secondly, a PVT regressor that inputs the circuit’s sizing and the nominal performances to estimate the PVT corner performances via multiple parallel artificial neural networks is proposed. Two control phases prevent the optimization process from being misled by inaccurate performance estimates. As such, this book details the optimal description of the input/output data relation that should be fulfilled. The developed description is mainly reflected in two of the system’s characteristics, the shape of the input data and its incorporation in the sizing optimization loop. An optimal description of these components should be such that the model should produce output data that fulfills the desired relation for the given training data once fully trained. Additionally, the model should be capable of efficiently generalizing the acquired knowledge in newer examples, i.e., never-seen input circuit topologies.
This book applies to the scientific area of electronic design automation (EDA) and addresses the automatic sizing of analog integrated circuits (ICs). Particularly, this book presents an approach to enhance a state-of-the-art layout-aware circuit-level optimizer (GENOM-POF), by embedding statistical knowledge from an automatically generated gradient model into the multi-objective multi-constraint optimization kernel based on the NSGA-II algorithm. The results showed allow the designer to explore the different trade-offs of the solution space, both through the achieved device sizes, or the respective layout solutions.
This book presents an innovative methodology for the automatic generation of analog integrated circuits (ICs) layout, based on template descriptions and on evolutionary computational techniques. A design automation tool, LAYGEN II was implemented to validate the proposed approach giving special emphasis to reusability of expert design knowledge and to efficiency on retargeting operations.
This book describes a new pattern discovery approach based on the combination among rules between Perceptually Important Points (PIPs) and the Symbolic Aggregate approximation (SAX) representation optimized by Genetic Algorithm (GA). The proposed approach was tested with real data from S&P500 index and all the results obtained outperform the Buy&Hold strategy. Three different case studies are presented by the authors.
The management of financial portfolios or funds constitutes a widely known problematic in financial markets which normally requires a rigorous analysis in order to select the most profitable assets. This subject is becoming popular among computer scientists which try to adapt known Intelligent Computation techniques to the market’s domain. This book proposes a potential system based on Genetic Algorithms, which aims to manage a financial portfolio by using technical analysis indicators. The results are promising since the approach clearly outperforms the remaining approaches during the recent market crash.
This book presents a new computational finance approach combining a Symbolic Aggregate approximation (SAX) technique with an optimization kernel based on genetic algorithms (GA). While the SAX representation is used to describe the financial time series, the evolutionary optimization kernel is used in order to identify the most relevant patterns and generate investment rules. The proposed approach considers several different chromosomes structures in order to achieve better results on the trading platform The methodology presented in this book has great potential on investment markets.
The microelectronics market, with special emphasis to the production of complex mixed-signal systems-on-chip (SoC), is driven by three main dynamics, time-- market, productivity and managing complexity. Pushed by the progress in na- meter technology, the design teams are facing a curve of complexity that grows exponentially, thereby slowing down the productivity design rate. Analog design automation tools are not developing at the same pace of technology, once custom design, characterized by decisions taken at each step of the analog design flow, - lies most of the time on designer knowledge and expertise. Actually, the use of - sign management platforms, like the Cadences Virtuoso platform, with a set of - tegrated CAD tools and database facilities to deal with the design transformations from the system level to the physical implementation, can significantly speed-up the design process and enhance the productivity of analog/mixed-signal integrated circuit (IC) design teams. These design management platforms are a valuable help in analog IC design but they are still far behind the development stage of design automation tools already available for digital design. Therefore, the development of new CAD tools and design methodologies for analog and mixed-signal ICs is ess- tial to increase the designer’s productivity and reduce design productivitygap. The work presented in this book describes a new design automation approach to the problem of sizing analog ICs.
The management of financial portfolios or funds constitutes a widely known problematic in financial markets which normally requires a rigorous analysis in order to select the most profitable assets. This subject is becoming popular among computer scientists which try to adapt known Intelligent Computation techniques to the market’s domain. This book proposes a potential system based on Genetic Algorithms, which aims to manage a financial portfolio by using technical analysis indicators. The results are promising since the approach clearly outperforms the remaining approaches during the recent market crash.
This book applies to the scientific area of electronic design automation (EDA) and addresses the automatic sizing of analog integrated circuits (ICs). Particularly, this book presents an approach to enhance a state-of-the-art layout-aware circuit-level optimizer (GENOM-POF), by embedding statistical knowledge from an automatically generated gradient model into the multi-objective multi-constraint optimization kernel based on the NSGA-II algorithm. The results showed allow the designer to explore the different trade-offs of the solution space, both through the achieved device sizes, or the respective layout solutions.
This book addresses the automatic sizing and layout of analog integrated circuits (ICs) using deep learning (DL) and artificial neural networks (ANN). It explores an innovative approach to automatic circuit sizing where ANNs learn patterns from previously optimized design solutions. In opposition to classical optimization-based sizing strategies, where computational intelligence techniques are used to iterate over the map from devices’ sizes to circuits’ performances provided by design equations or circuit simulations, ANNs are shown to be capable of solving analog IC sizing as a direct map from specifications to the devices’ sizes. Two separate ANN architectures are proposed: a Regression-only model and a Classification and Regression model. The goal of the Regression-only model is to learn design patterns from the studied circuits, using circuit’s performances as input features and devices’ sizes as target outputs. This model can size a circuit given its specifications for a single topology. The Classification and Regression model has the same capabilities of the previous model, but it can also select the most appropriate circuit topology and its respective sizing given the target specification. The proposed methodology was implemented and tested on two analog circuit topologies.
In this book, innovative research using artificial neural networks (ANNs) is conducted to automate the sizing task of RF IC design, which is used in two different steps of the automatic design process. The advances in telecommunications, such as the 5th generation broadband or 5G for short, open doors to advances in areas such as health care, education, resource management, transportation, agriculture and many other areas. Consequently, there is high pressure in today’s market for significant communication rates, extensive bandwidths and ultralow-power consumption. This is where radiofrequency (RF) integrated circuits (ICs) come in hand, playing a crucial role. This demand stresses out the problem which resides in the remarkable difficulty of RF IC design in deep nanometric integration technologies due to their high complexity and stringent performances. Given the economic pressure for high quality yet cheap electronics and challenging time-to-market constraints, there is an urgent need for electronic design automation (EDA) tools to increase the RF designers’ productivity and improve the quality of resulting ICs. In the last years, the automatic sizing of RF IC blocks in deep nanometer technologies has moved toward process, voltage and temperature (PVT)-inclusive optimizations to ensure their robustness. Each sizing solution is exhaustively simulated in a set of PVT corners, thus pushing modern workstations’ capabilities to their limits. Standard ANNs applications usually exploit the model’s capability of describing a complex, harder to describe, relation between input and target data. For that purpose, ANNs are a mechanism to bypass the process of describing the complex underlying relations between data by feeding it a significant number of previously acquired input/output data pairs that the model attempts to copy. Here, and firstly, the ANNs disrupt from the most recent trials of replacing the simulator in the simulation-based sizing with a machine/deep learning model, by proposing two different ANNs, the first classifies the convergence of the circuit for nominal and PVT corners, and the second predicts the oscillating frequencies for each case. The convergence classifier (CCANN) and frequency guess predictor (FGPANN) are seamlessly integrated into the simulation-based sizing loop, accelerating the overall optimization process. Secondly, a PVT regressor that inputs the circuit’s sizing and the nominal performances to estimate the PVT corner performances via multiple parallel artificial neural networks is proposed. Two control phases prevent the optimization process from being misled by inaccurate performance estimates. As such, this book details the optimal description of the input/output data relation that should be fulfilled. The developed description is mainly reflected in two of the system’s characteristics, the shape of the input data and its incorporation in the sizing optimization loop. An optimal description of these components should be such that the model should produce output data that fulfills the desired relation for the given training data once fully trained. Additionally, the model should be capable of efficiently generalizing the acquired knowledge in newer examples, i.e., never-seen input circuit topologies.
This book presents an innovative methodology for the automatic generation of analog integrated circuits (ICs) layout, based on template descriptions and on evolutionary computational techniques. A design automation tool, LAYGEN II was implemented to validate the proposed approach giving special emphasis to reusability of expert design knowledge and to efficiency on retargeting operations.
This book presents a new computational finance approach combining a Symbolic Aggregate approximation (SAX) technique with an optimization kernel based on genetic algorithms (GA). While the SAX representation is used to describe the financial time series, the evolutionary optimization kernel is used in order to identify the most relevant patterns and generate investment rules. The proposed approach considers several different chromosomes structures in order to achieve better results on the trading platform The methodology presented in this book has great potential on investment markets.
This book describes a new pattern discovery approach based on the combination among rules between Perceptually Important Points (PIPs) and the Symbolic Aggregate approximation (SAX) representation optimized by Genetic Algorithm (GA). The proposed approach was tested with real data from S&P500 index and all the results obtained outperform the Buy&Hold strategy. Three different case studies are presented by the authors.
This book presents a new methodology with reduced time impact to address the problem of analog integrated circuit (IC) yield estimation by means of Monte Carlo (MC) analysis, inside an optimization loop of a population-based algorithm. The low time impact on the overall optimization processes enables IC designers to perform yield optimization with the most accurate yield estimation method, MC simulations using foundry statistical device models considering local and global variations. The methodology described by the authors delivers on average a reduction of 89% in the total number of MC simulations, when compared to the exhaustive MC analysis over the full population. In addition to describing a newly developed yield estimation technique, the authors also provide detailed background on automatic analog IC sizing and optimization.
Thank you for visiting our website. Would you like to provide feedback on how we could improve your experience?
This site does not use any third party cookies with one exception — it uses cookies from Google to deliver its services and to analyze traffic.Learn More.