Mustapha C. E. Yagoub

A robust algorithm for automatic development of neural-network models for microwave applications

In this paper, we propose a robust algorithm for automating the neural network based RF/Microwave model development process. The algorithm can build a neural model starting with zero amount of training/test data, and then proceeding with neural network training in a stage-wise manner. In each stage, the algorithm utilizes neural network error criteria to determine additional training/test samples required and their location in model input space. The algorithm dynamically generates these new data samples during training, by automatic driving of simulation tools, e.g., OSA90, Ansoft-HFSS. Initially, fewer hidden neurons are used, and the algorithm adjusts the neural network size whenever it detects under-learning. Our technique integrates all the sub-tasks involved in neural modeling, thereby facilitating a more efficient and automated model building process. It significantly reduces the intensive human effort demanded by the conventional step-by-step neural modeling approach. The algorithm is demonstrated through MESFET and Embedded Capacitor examples.

Advanced microwave modeling framework exploiting automatic model generation, knowledge neural networks, and space mapping

In this paper, we propose an efficient Knowledge based Automatic Model Generation (KAMG) technique, aimed at generating microwave neural models of highest possible accuracy using fewest accurate data. The technique is comprehensively derived to integrate three distinct powerful concepts, namely, automatic model generation, knowledge neural networks and space mapping. We utilize two types of data generators - fine data generators that are accurate and slow (e.g., CPU-intensive 3D-EM simulators); coarse data generators that are approximate and fast (e.g., inexpensive 2D-EM). Motivated by the space-mapping concept, the KAMG utilizes extensive approximate data but fewest accurate data to generate neural models that accurately match fine data. Our formulation exploits a variety of knowledge network architectures to facilitate reinforced neural network learning from both coarse and fine data. During neural model generation by KAMG both coarse and fine data generators are automatically driven using adaptive sampling. The proposed technique is demonstrated through examples of MOSFET, and embedded passives used in multi-layer PCBs.

Neural based dynamic modeling of nonlinear microwave circuits

A neural network formulation for modeling nonlinear microwave circuits is achieved in the most desirable format, i.e., continuous time-domain dynamic system format. The proposed dynamic neural network (DNN) model can be developed directly from input-output data without having to rely on internal details of the circuit. An algorithm is developed to train the model with time or frequency domain information. A circuit representation of the model is proposed such that the model can be incorporated into circuit simulators for high-level design. Examples of dynamic-modeling of amplifiers, mixer and their use in system simulation are presented.

Efficient analytical formulation and sensitivity analysis of neuro-space mapping for nonlinear microwave device modeling

A new computer-aided design (CAD) method for automated enhancement of nonlinear device models is presented, advancing the concept of Neuro-space mapping (Neuro-SM). It is a systematic computational method to address the situation where an existing device model cannot fit new device data well. By modifying the current and voltage relationships in the model, Neuro-SM produces a new model exceeding the accuracy limit of the existing model. In this paper, a novel analytical formulation of Neuro-SM is proposed to achieve the same accuracy as the basic formulation of Neuro-SM (known as circuit-based Neuro-SM) with much higher computational efficiency. Through our derivations, the mapping between the existing (coarse) model and the overall Neuro-SM model is analytically achieved for dc, small-signal, and large-signal simulation and sensitivity analysis. The proposed analytical formulation is a significant advance over the circuit-based Neuro-SM, due to the elimination of extra circuit equations needed in the circuit-based formulation. A two-phase training algorithm utilizing gradient optimization is also developed for fast training of the analytical Neuro-SM models. Application examples on modeling heterojunction bipolar transistor (HBT), metal-semiconductor-field-effect transistor (MESFET), and high-electron mobility transmistor (HEMT) devices and the use of Neuro-SM models in harmonic balance simulations demonstrate that the analytical Neuro-SM is an efficient approach for modeling various types of microwave devices. It is useful for systematic and automated update of nonlinear device model library for existing circuit simulators.

Neural-based dynamic modeling of nonlinear microwave circuits

A neural network formulation for modeling nonlinear microwave circuits is achieved in the most desirable format, i.e., continuous time-domain dynamic system format. The proposed dynamic neural network (DNN) model can be developed directly from input-output data without having to rely on internal details of the circuit. An algorithm is developed to train the model with time or frequency domain information. A circuit representation of the model is proposed such that the model can be incorporated into circuit simulators for high-level design. Examples of dynamic-modeling of amplifiers, mixer and their use in system simulation are presented.

Advanced microwave modeling framework exploiting automatic model generation, knowledge neural networks and space mapping

In this paper, we propose an efficient Knowledge based Automatic Model Generation (KAMG) technique, aimed at generating microwave neural models of highest possible accuracy using fewest accurate data. The technique is comprehensively derived to integrate three distinct powerful concepts, namely, automatic model generation, knowledge neural networks and space mapping. We utilize two types of data generators - fine data generators that are accurate and slow (e.g., CPU-intensive 3D-EM simulators); coarse data generators that are approximate and fast (e.g., inexpensive 2D-EM). Motivated by the space-mapping concept, the KAMG utilizes extensive approximate data but fewest accurate data to generate neural models that accurately match fine data. Our formulation exploits a variety of knowledge network architectures to facilitate reinforced neural network learning from both coarse and fine data. During neural model generation by KAMG both coarse and fine data generators are automatically driven using adaptive sampling. The proposed technique is demonstrated through examples of MOSFET, and embedded passives used in multi-layer PCBs.

A new macromodeling approach for nonlinear microwave circuits based on recurrent neural networks

For the first time, recurrent neural networks (RNN) are trained to learn the dynamic responses of nonlinear microwave circuits. Once trained, the RNN macromodel provides fast prediction of the full analog behavior of the original circuit and can be used for high level simulation and optimization.

Neural-network approaches to electromagnetic-based modeling of passive components and their applications to high-frequency and high-speed nonlinear circuit optimization

In this paper, artificial neural-network approaches to electromagnetic (EM)-based modeling in both frequency and time domains and their applications to nonlinear circuit optimization are presented. Through accurate and fast EM-based neural models of passive components, we enable consideration of EM effects in high-frequency and high-speed computer-aided design, including components geometrical/physical parameters as optimization variables. Formulations for standard frequency-domain neural modeling approach, and recent time-domain neural modeling approach based on state-space concept, are described. A new EM-based time-domain neural modeling approach combining existing knowledge in the form of equivalent circuits (ECs), with state-space equations (SSEs) and neural networks (NNs), called the EC-SSE-NN, is proposed. The EC-SSE-NN models allow EM behaviors of passive components in the circuit to interact with nonlinear behaviors of active devices, and facilitate nonlinear circuit optimization in the time domain. An automatic mechanism for EM data generation, which can lead to efficient training of neural models for EM components, is presented. Demonstration examples including EM-based frequency-domain optimization of a three-stage amplifier, time-domain circuit optimization in a multilayer printed circuit board, including geometrical/physical-oriented neural models of power-plane effects, and EM-based optimization of a high-speed interconnect circuit with embedded passive terminations and nonlinear buffers in the time domain are presented.

Exact adjoint sensitivity analysis for neural-based microwave modeling and design

For the first time, an adjoint neural network method is introduced for sensitivity analysis in neural-based microwave modeling and design. Exact first and second order sensitivities are systematically calculated for generic microwave neural models including variety of knowledge based neural models embedding microwave empirical information. A new formulation allows the models to learn both the input/output behavior of the modeling problem and its derivative data simultaneously. Examples for passive and active microwave modeling and simulation are presented.

ULTRA WIDEBAND CPW-FED APERTURE ANTENNA WITH WLAN BAND REJECTION

In this paper, we present a new ultra wideband antenna design with band rejection for UWB applications. A CPW-fed circular patch radiates through a circular aperture, which ensures wideband impedance matching and stable omnidirectional pattern over an UWB frequency range, from 3GHz to 10.6GHz. In order to avoid interference