An Efficient Intermodulation Product Computing Technique for Broadband Active Transmit Systems.

The determination of intermodulation products is an ever pervasive problem to the radar and satellite community—its eminence will continue to manifest itself as data rates increase, as transmitters are required to handle increasing numbers of multiple carriers, and as amplifiers are pushed to operate closer to their nonlinear regions (to circumvent the need of adding the extra weight and cost of a more linear amplifier). To study the intermodulation effects, fast Fourier transform (FFT)-based techniques are often employed. By definition, Δ = F₈/N defines the frequency resolution for the discrete Fourier transform (DFT). It is typically desired to design Δ to be as small as possible to allow for a very fine frequency resolution. Doing this requires that F₈, be minimized and/or N be selected as large as possible. Minimizing F₈, has its pitfalls since high-order harmonics may violate the Nyquist criteria. Selecting N to be as large as possible introduces unreasonably long simulation times. This may be the case when large values of N are very near or exceed the maximum allowable array size of a digital computer. Breaking this nexus is the focus of this paper, which introduces a computationally efficient technique that will allow the frequency and amplitude of the intermodulation terms to be precisely computed. [ABSTRACT FROM AUTHOR]