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

Hi 
Can anyone explain how one choses the order of the AR process when trying
to estimate it's coefficient (Via levinson durbin algorithm). Is this a
try and error?

Thanks

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

dreamerg wrote:
> Hi
> Can anyone explain how one choses the order of the AR process when trying
> to estimate it's coefficient (Via levinson durbin algorithm). Is this a
> try and error?
>
> Thanks

Hello Dreamerg,

If you have a mathematical model that fits your data well, then you
have a mathematical way to know the order of the process.

But if you have some data without knowledge of what model, if any, fits
your data, then try higher and higher order processes until the
residual drops below an acceptable error. A lot of speech work uses a
five tube vocal tract model, so 10 coefs work pretty well for these
cases.

Basically model your process and subtract the predicted values from the
raw values and see what it takes to get the error small.

IHTH,

Clay

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

Clay wrote:
> dreamerg wrote:
> > Hi
> > Can anyone explain how one choses the order of the AR process when trying
> > to estimate it's coefficient (Via levinson durbin algorithm). Is this a
> > try and error?
> >
> > Thanks
>
> Hello Dreamerg,
>
> If you have a mathematical model that fits your data well, then you
> have a mathematical way to know the order of the process.
>
> But if you have some data without knowledge of what model, if any, fits
> your data, then try higher and higher order processes until the
> residual drops below an acceptable error. A lot of speech work uses a
> five tube vocal tract model, so 10 coefs work pretty well for these
> cases.
>

There is a more Scientific approach Akaike Final Prediction Error
approach.

Naebad
> Basically model your process and subtract the predicted values from the
> raw values and see what it takes to get the error small.
> 
> IHTH,
> 
> Clay

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

dreamerg wrote:

> Hi 
> Can anyone explain how one choses the order of the AR process when trying
> to estimate it's coefficient (Via levinson durbin algorithm). Is this a
> try and error?


On every iteration of the Levinson-Durbin, you have the current residual 
error. Observe how the error is reduced with the number of iterations, 
and when there is too little of improvement or no improvement at all, 
then you have found the right order for the process.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.