Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

autoregressive model analysis

Status
Not open for further replies.

Biruntha

Junior Member level 1
Joined
May 26, 2015
Messages
16
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Location
Jaffna, Srilanka
Activity points
102
Hi,
can anyone briefly explain why we need to do autoregressive model analysis for audio files? give some introduction to autoregressive model analysis?
 

SanjKrish

Member level 3
Joined
Feb 6, 2011
Messages
67
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
India
Activity points
1,855
AR or All pole model is done to compress your data.

The traditional method of storing audio is to store each of its samples at all the sampling intervals.

In AR modeling you model the signal as thought it is the output of a LTI system(causal and stable of-course) with white noise as its input. LTI.jpg

The LTI system H(Z) = 1/(b0 + b1(z-1) + ... )

All you need is to determine the coeffs b0 .. bn and the variance of the white input x(n)

Then using the relation y(n) = var(x(n)) |H(e^jw)| you can construct the output audio.

Thereby instead of storing the samples you just store b0.. bn and var(x(n)) which is less data. This is how compression is achieved.
 

SanjKrish

Member level 3
Joined
Feb 6, 2011
Messages
67
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
India
Activity points
1,855
The book "Statistical signal processing" by Hayes gives a good idea of AR signal modelling.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top