Signal processing + parameter estimation

[JamesGoh]

Is parameter estimation in DSP concerned with things such as bit error rates, maximum likelihood detection etc ... ?

If not, can someone point me in the right direction so I can find some learning material.

thanks

 

[Proton Soup]

when i used parameter estimation (many many years ago), it was part of an adaptive controller model. that is, estimate parameters to adjust the control law, and over time get tighter control and a tighter confidence interval on the error of your control law estimates. our modelling was not related to communications, but the theory for the parameter estimation should be covered in standard graduate-level digital controls courses, i think.

 

[OptimalDesign]

Is parameter estimation in DSP concerned with things such as bit error rates, maximum likelihood detection etc ... ?

Yes those are parameters that some may like to play with; i.e. tweak. If you have a math model with any parameters that you want to tweak then these are parameter estimations. In http://www.digitalCalculus.com/misc/calculus-programming.html" most math models have parameters that are tweated by a 'find' statement.

For DSP, there are a dozen http://www.digitalCalculus.com/demo/rainbow.html" in one program to compare results. Maximum likelihood is one of the methods available. Results differ dramatically between estimators. See effects of zero padding. Some estimators can detect signals 50 to 100 dB from main signal.