An automatic step size lms time domain equalization filter and its realization method

Abstract

The present invention proposes an automatic step size LMS time-domain equalization filter and its implementation method, the equalization filter includes a filter, a main control mechanism unit, a secondary control mechanism unit and a first adder; the filter is used for Process the input vector according to the estimated tap output by the main control mechanism unit to obtain an estimated output vector and a decision output vector; the first adder is used to receive the estimated output vector and combine it with the expected response to obtain an estimated error; the The secondary control mechanism unit is used to receive and process the estimated error by using the QA-LMS algorithm to obtain the step size; the main control mechanism unit is used to receive and process the step size by using the tap gain algorithm, to get estimated taps. The invention reduces the steady-state error after convergence and the bit error rate of the output digital signal after equalization, and improves the convergence speed of the automatic step size algorithm at the same time.

Description

technical field [0001] The invention relates to the fields of digital communication and wireless communication, in particular to an automatic step size LMS time-domain equalization filter and a realization method thereof. Background technique [0002] In digital communication and wireless communication, the equalization filter can effectively eliminate the influence of intersymbol crosstalk. Common training-based time-domain equalizers are divided into the LMS algorithm family and the RLS algorithm family. The complexity of the LMS algorithm family equalization filter is linearly positively correlated with the RMS delay of the channel, the implementation is relatively simple, and it has very good steady-state error and robustness, so it is widely used in various wired communication systems In twisted wire voice communication technology, the transmission channels for such applications are mostly time-invariant and stable channels. Although the RLS algorithm has a good conve...

AI Technical Summary

Problems solved by technology

However, due to the nature of the exponential function, the step size after convergence cannot be very close to 0, resulting in a large step size after the A-LMS algorithm converges, and a large increment of the tap coefficient update, which makes the algorithm more sensitive to some mutation errors , resulting in the problem that the bit error rate of the judgment output signal is high

Images