A deep learning approach for volterra kernel extraction for time domain simulation of weakly nonlinear circuits

Keyphrases

• Deep Learning Methods 100%
• Time-domain Simulation 100%
• Volterra Kernels 100%
• Weakly Nonlinear Circuits 100%
• Kernel Extraction 100%
• Linear Time-varying Systems 66%
• Pseudorandom Sequences 66%
• Feedforward Neural Network 66%
• Time-domain Response 66%
• Neural Network 33%
• Impulse Response 33%
• Numerical Examples 33%
• Flat Region 33%
• Simulation Method 33%
• Gaussian Noise 33%
• Simulation Program 33%
• Frequency Spectrum 33%
• Model Order Reduction 33%
• Performance Metrics 33%
• Training Phase 33%
• Accurately Model 33%
• Memory Length 33%
• System Framework 33%
• Spices 33%
• Computer-aided Design Tools 33%
• Power Amplifier 33%
• Model Frequencies 33%
• Weakly Nonlinear 33%
• Long Memory 33%
• Transceiver 33%
• Recurrent Neural Network 33%
• Nonlinear Circuit Simulation 33%
• High-speed Links 33%
• IBIS® 33%
• Polynomial Nonlinearity 33%
• Multidimensional Convolution 33%
• 28Gbps 33%
• Higher-order Kernels 33%
• Multidimensional Extension 33%
• 4-level Pulse Amplitude Modulation (PAM-4) 33%
• Phase Inference 33%
• Neural Network Topology 33%
• White-box Model 33%

Computer Science

• Learning Approach 100%
• Neural Network 100%
• Deep Learning 100%
• linear time invariant system 100%
• Nonlinear Circuit 100%
• Feedforward Neural Network 100%
• Feed Forward Neural Networks 100%
• Time Domain Simulation 100%
• Signal Identification 50%
• Frequency Domain 50%
• Gaussian White Noise 50%
• Numerical Example 50%
• Simulation Program 50%
• Training Phase 50%
• Computer Aided Design 50%
• Network Topology 50%
• Performance Metric 50%
• Impulse Response 50%
• Power Amplifier 50%
• Level Simulation 50%
• Circuit Simulation 50%
• Recurrent Neural Network 50%
• Frequency Spectrum 50%
• Link Buffer 50%
• Dimensional Convolution 50%

Engineering

• Time Domain 100%
• Nonlinear Circuit 100%
• Deep Learning 100%
• Learning Approach 100%
• Invariant System 66%
• Linear Time Invariant 66%
• Feedforward 66%
• Domain Response 66%
• Impulse Response 33%
• Frequency Domain 33%
• Shallower 33%
• Signal Identification 33%
• Numerical Example 33%
• Metrics 33%
• SPICE 33%
• Circuit Simulation 33%
• Simulation Method 33%
• Nonlinearity 33%
• Frequency Spectrum 33%
• Computer Aided Design 33%
• Data Rate 33%
• Simulation Program 33%
• Design Tool 33%
• Gaussian White Noise 33%
• Box Model 33%
• Network Topology 33%
• Power Amplifier 33%
• Transceiver 33%
• Recurrent Neural Network 33%
• Memory Length 33%
• Dimensional Convolution 33%