Integrated Circuit/EM Simulation and Design Technologies for Advanced Radio Systems-on-chip | |||||

- About Icestars
- ICESTARS Research Areas
- ICESTARS Achievements
- Time Domain
- Frequency Domain
- EM Analysis
- Validation
- Glossary
- Publications
- ICESTARS at ECMI 2010
- Press room

| Your position: Icestars > Frequency Domain > Harmonic Balance: Improved robustness |

Within ICESTARS existing frequency-domain algorithms are improved respectively new ones are developed – the development of which is based on scalable algorithms for RF design. Scalability in general is applied to enable RF simulations for large circuits.

The algorithms used in Harmonic Balance and its noise analysis are optimised for different types of circuits and available computational resources using the most accurate initial frequency estimates. Volterra-type analysis is used for detailed distortion analysis especially needed to design linear power amplifiers.

The new **ICESTARS **algorithms find the operating frequency of a free-running oscillator, thus improving both speed and robustness of harmonic balance analysis. One algorithm, using available pole-zero analysis methods to determine the dominant poles of the system, calculates a good initial estimate for the operating frequency. A different approach tries to find an initial frequency estimate from optimisation techniques.

HB is an analysis method that simulates a circuit under single- or multi-tone excitations. Nonlinear noise analysis is based on HB analysis and takes into account mixing phenomena from one frequency to another. HB, in principle, is a non-linear, steady-state simulation. Voltage and current sources create discrete frequencies resulting in a spectrum of discrete frequencies at every node in the circuit. Linear circuit components are solely modeled in frequency domain. Non-linear components are modeled in time domain and Fourier-transformed before each solving step.