Adaptive Filter | ADPCM | ADSP | ADSP-2181 | Aliasing | AMR | Anti-Aliasing | ARMA | Autocorrelation | AutoCovariance | Beamforming | Bessel | Blackfin | Butterworth | C6713 | CCS | Chebyshev | CIC Filter | Circular Convolution | Code Composer Studio | Comb Filter | Compression | Convolution | Cross Correlation | DCT | Decimation | Deconvolution | Demodulation | DM642 | DSP Boards | DSP/BIOS | DTMF | Echo Cancellation | Equalization | Equalizer | ETSI | EZLITE (Ez-kit Lite) | FFT | FFTW | FIR Filter | Fixed Point | FSK | G.711 | G.723 | G.729 | Gaussian Noise | Goertzel | GPIO | Hilbert Transform | IFFT | IIR Filter | Interpolation | Invariance | JTAG | Kalman | Laplace Transform | Levinson | LPC | McBSP | MIPS | Modulation | MPEG | Multirate | Notch Filter | Nyquist | OFDM | Oversampling | Pink Noise | Pitch | PLL | Polyphase | QAM | QDMA | Quantization | Quantizer | Radar | Random Noise | Reed Solomon | Remez | Resampling | RTDX | Sampling | Sharc | TI C6711 | Undersampling | Viterbi | Wavelets | White Noise | Wiener Filter | Windowing | XDS510PP | Z Transform

> Consider a four-input, one-output nonlinear system, say an amplifier.
> 
> I have available the four power spectral density (PSD) curves for the
> four pseudo-random input signals, and also the PSD, P(f), of the
> output signal, where the independent variable f is frequency.
> 
> I would like to know if it is possible to determine the amount of
> power present in the output signal that is due to nonlinear effects. 
> So I would like to obtain the decomposition (if it even makes sense)
>    P(f) = P_1(f) + P_2(f) + P_3(f) + P_4(f) + Q(f)
> where P_k is the power due to linear amplification of the k'th input
> signal, and Q(f) is the power present in the output signal which is
> the sum of the noise contribution and all intermodulation products.
> 
> Is it possible to do this?  Is there an efficient algorithm?  The PSDs
> are sampled at about 20000 discrete frequencies.

You might be interested in the book 

Bendat: "Nonlinear Systems - Techniques and Applications"
     Wiley, 1998.

I have only browesd quickly through the book, but it seems to focus 
on exactly these types of questions. 

Rune

______________________________
DSPRelated.com's 50,000th member announced! Details Here.

"Rune Allnor" <a...@tele.ntnu.no> wrote in message
news:f...@posting.google.com...
> > Consider a four-input, one-output nonlinear system, say an amplifier.
> >
> > I have available the four power spectral density (PSD) curves for the
> > four pseudo-random input signals, and also the PSD, P(f), of the
> > output signal, where the independent variable f is frequency.
> >
> > I would like to know if it is possible to determine the amount of
> > power present in the output signal that is due to nonlinear effects.
> > So I would like to obtain the decomposition (if it even makes sense)
> >    P(f) = P_1(f) + P_2(f) + P_3(f) + P_4(f) + Q(f)
> > where P_k is the power due to linear amplification of the k'th input
> > signal, and Q(f) is the power present in the output signal which is
> > the sum of the noise contribution and all intermodulation products.
> >
> > Is it possible to do this?  Is there an efficient algorithm?  The PSDs
> > are sampled at about 20000 discrete frequencies.
>
> You might be interested in the book
>
> Bendat: "Nonlinear Systems - Techniques and Applications"
>      Wiley, 1998.
>
> I have only browesd quickly through the book, but it seems to focus
> on exactly these types of questions.
>
> Rune

Another possibility, if you are just simulating the amplifier and so can
replicate many identical copies, might be to sandwich several amplifiers in
a butler matrix then feed the input signals into different input ports.  The
output signals come out on different ports along with different sets of
intermod products so it may be that you can arrange to seperate out the bits
you are most interested in .

Best of luck - Mike

______________________________
DSPRelated.com's 50,000th member announced! Details Here.