EDA² | ISEDA 2024

Compact Modeling for Circuit Innovation

Abstract: A Verilog-A compact model for four-wave mixing (FWM) is proposed in this brief. It is fully compatible with existing EDA platforms, and can support the rapid electronic-photonic co-simulation. The model avoids the description of the complicated physical process of the FWM, and provides an easy way for system designers to monitor the changes of the key optical parameters, thus accelerating the co-design and co-optimization of the hybrid systems containing FWM process. The framework and the key derivations for modeling are presented, and the simulation results of our model agrees well with numerical full-map results.

Abstract: This paper summarizes the recent advances in compact modeling techniques based on the Quasi-Physical Zone Division (QPZD) theory mainly for AlGaN/GaN HEMTs. Firstly, the modeling principle of QPZD and the corresponding core model is introduced. A comprehensive comparison between the QPZD model and other well-known compact models is also presented. For high-power design applications, the modeling techniques of dispersion effects including self-heating effects, ambient temperature effects, and trapping effects are well developed. Furthermore, several efforts have been devoted to the extension of QPZD model capability. Details of the microwave noise model, switch model, and process fluctuation model are introduced, respectively. Owing to these efforts, the developed QPZD modeling techniques can contribute to the comprehensive and accurate designs of significant chips in RF front-ends.

Abstract: Recent advancements in neural network-based behavioral modeling approaches have profoundly influenced the modeling and simulation of radio frequency (RF) devices and circuit modules. This manuscript elucidates the constraints inherent in the existing body of literature with regard to accurately encapsulating the intricacies of nonlinear components. It introduces an innovative modeling paradigm designed to surmount these challenges. Moreover, the incorporation of neural ordinary differential equations into the modeling framework permits a more adaptable simulation step input, thereby augmenting the modeling process. The efficacy of the proposed model is validated through its application to PIN diodes within high-power microwave environments. Parameters of the model are meticulously extracted, and the inference mechanism is seamlessly integrated into circuit modules utilizing the VerilogA language. This integration substantially elevates the practical applicability of the model in circuit simulations.

Abstract: A compact model for organic light emitting diodes (OLED) is reported, which include current-voltage characteristics and capacitance-voltage characteristics for all regions. Compared with the previously published model, it not only expands the scope of application of the model, but also makes the model have better continuity and smoothness while ensuring accuracy. The model is continuous in the whole region through smooth functions and parameters, which provides convenience for circuit simulation. After parameter extraction, the model is in good agreement with the experimental measured data.

Abstract: The limit of Parasitic Effects in terms of number of synapses is estimated for phase-change-memory based neuromorphic circuit under scaling technology nodes following ITRS. Parasitic capacitance components are evaluated for a 55nm process and extrapolated to other nodes. A memory compact model is used to study the effects of the capacitance on firing and weight updating in synapses, which provides design constraint for posterior bitline load estimation. The estimated maximum number of synapses indicate decreasing bitline load capacity due to intensified impact of parasitic capacitance along scaling feature size. The proposed estimation methodology is applicable to advanced nodes to provide quick evaluation for neuromorphic circuit design.

Abstract: In this work, we present a comprehensive study of silicon diode current characteristics from room to cryogenic temperatures by experimental measurement, TCAD simulation, and physics model down. For the first time in the diode current model, we employ the Fermi-Dirac statistic to calculate the Fermi levels for solving the carrier degeneration problem at low temperatures, while considering the incomplete ionization on the series resistance effect. The temperature dependence of built-in potential, and the forward and reverse current of the silicon diode are analyzed and modeled. Our model prediction has good agreement with calibrated TCAD in the temperature range from 50K to 300K. This work advances the temperature-dependence diode models, while effectively evaluating the parasitic PN junction currents in CMOS technology at cryogenic temperature.