Adaptive Algorithm

From GNU Radio
Jump to navigation Jump to search

Creates an Adaptive Algorithm object to be used in Linear_Equalizer or Decision_Feedback_Equalizer to define how the error signal is calculated and how taps are updated, depending on the specified algorithm


(R): Run-time adjustable

Algorithm Type
enum to specify which adaptive algorithm will be used; LMS, NLMS, CMA are the valid choices
Digital Constellation Object
A Constellation_Object which specifies the modulation used to adapt using decision directed mode of the equalizer
Step Size
Specifies how quickly the adaptive algorithm will converge. Too high and the equalizer becomes unstable. The optimal value is dependent on the statistical properties of the input signal
(CMA only) Specifies the number of constellation points, e.g. for QPSK modulus = 4

Example Flowgraph

See Linear_Equalizer for a flowgraph utilizing the Adaptive Algorithm Object

Adaptive algorithm.png

Adaptive algorithm LMS.png

Adaptive algorithm CMA.png

Source Files