Channel estimation method for filtering multi-tone modulation technology

Abstract

The invention provides a channel estimation method for filtering multi-tone modulation technology. The interference introduced by the multipath channel to the pilot is gradually eliminated by iteration based on the time domain channel estimation algorithm of a PN sequence, thereby improving the estimation accuracy of the algorithm.

Description

technical field [0001] The invention relates to the field of wireless communication, in particular to a channel estimation method for filtering multi-tone modulation technology. Background technique [0002] Filtered multi-tone modulation (FMT) technology has the advantages of anti-multipath fading and high utilization of spectrum resources. Since it does not require the overhead of cyclic prefix, FMT has higher spectrum utilization efficiency than OFDM technology. and Doppler characteristics. This technology has a good application prospect in the marine broadband data communication system. When the signal is transmitted through the wireless channel, due to the influence of Doppler frequency shift and multipath fading, the signal will be distorted or deformed, resulting in an increase in the bit error rate, and in severe cases, normal communication may not be possible. Therefore, in order to overcome the influence of the channel and restore the modulated data to the greate...

AI Technical Summary

Problems solved by technology

[0003] In the field of frequency domain processing, the approximate interference cancellation (IAM) algorithm is commonly used. The basic idea of ​​the algorithm is to eliminate ISI by adding a 0 guard interval between the pilot and the data. The performance of the algorithm is improved compared with the LS algorithm, but the algorithm Will increase additional overhead and reduce spectrum utilization efficiency

In addition, the method of constructing a circular convolution formula between the channel and the cyclic sequence by inserting multiple cycles of the training sequence in the time domain, and then converting to the frequency domain to estimate the channel is also often cited. This algorithm requires the cycle of the training sequence If it is greater than the maximum multipath delay of the channel, it still has certain limitations

[0004] In the field of time-domain processing, the commonly used algorithm is the time-domain channel estimation algorithm based on PN sequences. This algorithm uses the autocorrelation characteristics of PN sequences to estimate the channel. Existence, the estimation accuracy of the algorithm cannot be further improved with the increase of the signal-to-noise ratio, and the algorithm has a floor effect

Another cited algorithm is to propose a time-domain channel estimation based on MUSIC (Multiple Signal Classification) and LS based on FMT characteristics. This algorithm eliminates the ceiling effect of the PN sequence, but ignores the noise covariance matrix pair algorithm performance impact

A time-domain channel estimation algorithm based on maximum likelihood (ML) has also been proposed. This algorithm does not need to insert a guard sequence, but due to the influence of data on pilot interference, the performance of the algorithm is limited.

In addition, there is another algorithm for channel estimation for each subcarrier, but when the number of subcarriers is large, the complexity of the algorithm is high, which is not conducive to practical implementation

Images