Jilin Tan

Partitioned Latency Insertion Method With a Generalized Stability Criteria

This paper presents a modular approach to the high-frequency simulation of large networks. By utilizing the latency insertion method (LIM) and by studying the stability criteria of partitions of different latencies in the circuit, a robust method is formulated that is able to perform transient simulations significantly faster than the conventional LIM method. The LIM method is also extended to handle dependent sources, and a general stability criterion for selecting the time step, independent of the topology of the circuit, is proposed. The new method is verified with existing commercial tools for circuit simulations, and the improvement in run time is also depicted.

Comparative Study of Convolution and Order Reduction Techniques for Blackbox Macromodeling Using Scattering Parameters

In this paper, a fast convolution method using scattering parameters is presented for the macromodeling of blackbox multiport networks. The method is compared to model-order reduction passive macromodeling techniques in terms of robustness and computational efficiency. When scattering parameters are used as the transfer functions, convolution calculations can be accelerated to achieve superior performance and the resulting procedure leads to a robust, accurate, and efficient macromodel generation scheme. This paper examines the formulation of the convolution method. Model-order reduction techniques are reviewed and benchmark comparisons are performed.

Combining rational interpolation with the vector fitting method

In this paper, a method is proposed to improve the initial poles and convergence of the vector fitting method. The procedure takes advantage of the robustness of the rational interpolation method in extracting stable poles over partitioned bandwidth and the numerical stability and fast convergence of the vector fitting method in generating accurate high-order rational approximation. First, the stable poles of the entire frequency range are obtained by partitioning the domain and performing rational interpolation in each bandwidth. Then, the collected poles are used as initial poles to the vector-fitting algorithm to obtain the overall high-order rational approximation. The accuracy and convergence of the standard vector fitting and proposed methods are compared. Examples are provided to demonstrate the advantage of the proposed approach.

Partitioned latency insertion method (PLIM) with stability considerations

In this paper, we present a modular approach to the high frequency simulation of large networks by utilizing the latency insertion method (LIM) and considering the stability criteria of partitions of different latencies in the circuit. This results in a robust algorithm that is able to preserve the stability condition while improving the runtime of the overall transient simulation. An extension to the LIM to handle dependent sources is also presented, along with a generalized stability criteria for selecting a maximum stable time step.

Latency Insertion Method with Frequency Dependent Effect

This paper presents an extension of the latency insertion method (LIM) to the treatment of circuits with frequency-dependent parameters. In particular, the skin effect of conductors is modeled in the formulation and the simulation results are compared with actual measurements

Blackbox Macromodel with S-Parameters and Fast Convolution

In this paper, the scattering parameters of blackbox multi-port networks are pre-processed in the frequency domain to satisfy causality. Next, they are approximated to yield weighted delta functions in the time domain thus allowing the fast simulation. The resulting procedure leads to a robust, accurate and efficient macromodel generation scheme.

Application of the latency insertion method to circuits with blackbox macromodel representation

In this work the latency insertion method (LIM) is applied to the treatment of blackbox networks described by their frequency-domain scattering parameters. The method allows the simulation of passive macromodels in the LIM environment. This generalization allows LIM to simulate subnetworks with frequency-dependent parameters describing phenomena such as skin effect and substrate loss. The derivation of the algorithms is presented as well as simulation results to validate the method.

IBIS Simulation Using the Latency Insertion Method (LIM)

This paper presents an approach for the transient simulation of I/O buffers described by IBIS models. Using the latency insertion method, a formulation can be obtained for the transient behavior of IBIS models combined with external circuitry. The formulation offers better convergence than traditional Newton-Raphson methods and is therefore more robust. The method also implements the BIRD95 and BIRD98 updates that account for predriver current, simultaneous switching noise, and gate modulation effects. Several computer simulations are performed to validate the method.

A comparison of two latency insertion methods in dependent sources applications

This paper compares two different formulations of the latency insertion method (LIM) for the analysis of circuits with dependent sources. One is the scalar LIM and the other is the amplification matrix LIM. Numerical experiments demonstrate that the scalar LIM necessitates much smaller time step than the amplification matrix LIM when handling certain types of dependent sources indicating better performance of the latter technique without a sacrifice in accuracy.

Three finite-element time-domain-based numerical algorithms for high-frequency broadband PCB simulations

As the operating frequency and integration level of integrated circuits (IC) increase, full-wave analysis algorithms are needed to accurately simulate the arising electromagnetic phenomena. The finite-element time-domain (FETD) method has become an attractive candidate for this simulation due to its advantages in modeling complex geometries and materials, conducting transient analyses, and performing broadband characterizations. Three FETD-based numerical algorithms, including the original FETD method, the dual-field domain-decomposition (DFDD) method, and the discontinuous Galerkin time-domain (DGTD) method, are investigated and applied to the simulation of printed circuit board (PCB) structures to demonstrate their accuracies and capabilities.