Equalization method taking full advantage of channel multi-path characteristic

Abstract

The invention relates to an equalization method taking full advantage of a channel multi-path characteristic. The equalization method takes full advantage of delayed extension characteristics of multi-path channels to improve an equalization performance, and is characterized by successively performing matched filtering, sampling, receiving symbol grouping, channel factor matrix construction, symbol-to-message iterative computation in groups and judgment; wherein during message iterative computation, message of partial symbols of a next receiving group is used as external message of corresponding symbols of a current group to participate in computation; during the judgment, absolute values of confidence ratios of transmitting symbols at the rear part of the current group are compared with absolute values of confidence ratios of transmitting symbols at the front part of the next group one by one, if the latter absolute value is greater than the former absolute value, the confidence ratioof the current group is replaced by the confidence ratio of the next group; in this way, the confidence ratio of the transmitting symbol group is improved, judgment accuracy is raised, and a BER (BitError Rate) after equalization is lower than a BER after processing which does not take full advantage of the channel multi-path characteristic.

Description

technical field [0001] The invention relates to the equalization technology of the receiving end of the communication system, in particular to an equalization method which fully utilizes the multipath characteristic of the channel. Background technique [0002] In the wireless mobile communication system, the channel is multi-path and variable, so that it has dual selectivity in the frequency domain and time domain, and inter-symbol interference occurs during the transmission of the transmitted signal, resulting in a high reception error rate. Therefore, the receiving end needs more effective equalization and signal correction to reduce the bit error rate so that the next stage of channel decoding can play a better role. [0003] For the equalization of single-carrier communication, the traditional processing methods are linear transversal filtering method and nonlinear decision feedback method (DFE). Their computational complexity is low, and increases linearly with the inc...

AI Technical Summary

Problems solved by technology

For the MAP equalization method based on message delivery, the traditional method is to perform independent message calculation for each received packet to obtain the estimated value of the transmitted symbol for each packet. Such processing cannot make good use of the multipath characteristics of the channel.

Images