By proving that the response’s power spectral densities (output spectra) of an arbitrary (causal, stable and of finite memory) nonlinear dynamic system to two different signals are equal if and only if the two signals have equal higher-order autocorrelations or power spectral density functions, the authors of this proposal have recently established the theoretical framework for the solution of a problem of great engineering significance: How to design a standard signal stimulus – a multi-sine – capable of mimicking the test of a nonlinear dynamic system under a pre-determined excitation [1, 2]. Then, they also proposed a numerical method for designing a multi-sine that shows such higher-order statistics, which established the methodology necessary to conceive an “ideal arbitrary waveform generator”.
Unfortunately, when these multi-sine signals are to be created by a laboratory instrument, several problems appear which can be traced back to the nonlinear path from the DSP generator to the output of the RF generator itself.
As the multi-sine flows through the mixers, amplifiers, and filters, their tones’ amplitudes and relative phases are modified which necessarily impairs the required higher-order signal statistics.
If those changes result in mere linear phase shift and amplitude attenuation, then they can be compensated by including that effect in the DSP as a linear equalizer. However, if these effects are nonlinear, then the referred phase shift and attenuation depend on the amplitude of the multi-sine envelope, and a “nonlinear equalizer” (some form of dynamic linearizer) is in order.
|Start Date: 01-08-2005|
|End Date: 01-02-2007|
|Team: Nuno Miguel Goncalves Borges de Carvalho|
|Groups: Radio Systems – Av|
|Local Coordinator: Nuno Miguel Goncalves Borges de Carvalho|