Cyclostationary noise analysis of large RF circuits with multitone excitations

This paper introduces a new, efficient technique for analyzing noise in large RF circuits subjected to true multitone excitations. Noise statistics in such circuits are time-varying, hence cyclostationary stochastic processes, characterized by harmonic power spectral densities (HPSDs), are used to describe noise. HPSDs are used to devise a harmonic-balance-based noise algorithm with the property that required computational resources grow almost linearly with circuit size and nonlinearity. Device noises with arbitrary spectra (including thermal, shot, and flicker noises) are handled, and input and output correlations, as well as individual device contributions, can be calculated. HPSD-based analysis is also used to establish the nonintuitive result that bandpass filtering of cyclostationary noise can result in stationary noise. Results from the new method are validated against Monte Carlo simulations. A large RF integrated circuit (>300 nodes) driven by a local oscillator (LO) tone and a strong RF signal is analyzed in less than two hours. The analysis predicts correctly that the presence of the RF tone leads to noise folding, affecting the circuit's noise performance significantly.