Channel estimation method and device

Abstract

The invention discloses a channel estimation method and device, so as to avoid generation of a time domain mirror path when DFT (Discrete Fourier Transform) channel estimation is adopted in a multi-user equipment scene and improve the channel estimation accuracy. The channel estimation method comprises steps: the frequency domain receiving data of the symbol of the demodulated reference signal ineach time slot are determined, the base sequence in a demodulated reference signal transmitting sequence of the multiple user equipment is used for initial channel estimation, and the channel estimation value of the frequency domain is obtained; a virtual carrier generation matrix is used to determine virtual carriers, and virtual carriers are added to two sides of the channel estimation value ofthe frequency domain, wherein the virtual carrier generation matrix is generated according to the position relationship between the virtual carriers and the channel estimation value of the frequency domain; and the joint time domain channel estimation value of the multiple user equipment is determined by using the channel estimation value of the frequency domain with the virtual carriers added attwo sides.

Description

technical field [0001] The present application relates to the field of communication technologies, and in particular to a channel estimation method and device. Background technique [0002] The Long Term Evolution (LTE) base station needs to use the uplink (Up Link, UL) demodulation reference signal (Demodulation Reference Signal, DMRS) when receiving symbol equalization on the Physical Uplink Shared Channel (PUSCH) channel estimation. In a multi-user multiple input multiple output (Most In Most Out, MIMO) (Multiple Users MIMO, MU-MIMO) scenario, multiple user equipment (User Equipment, UE) multiplex the same time-frequency resource, and the DMRS sequences of different UEs pass through the frequency In the traditional channel estimation based on Discrete Fourier Transform (DFT) method, the receiver uses the characteristics of cyclic displacement of CIR of multiple UEs in the time domain to distinguish UEs and completes them separately. channel estimation. [0003] When th...

AI Technical Summary

Problems solved by technology

[0005] To sum up, in the MU-MIMO scenario, the CIRs of multiple UEs are arranged at multiple positions in the time domain. Observation in the frequency domain shows that the phase difference between the actual subcarrier and the virtual carrier is larger than that in the SU-MIMO scenario. In the channel estimation method of domain mirroring, there is a phase mutation between the mirrored part and the original signal part, which will lead to phase discontinuity in the symbol in the frequency domain (or interpreted as destroying the orthogonality of IDFT), and it is observed as a false path in the time domain (image path), which destroys the original multipath resolution and window noise suppression rules, and increases the mean square error (Mean Squared Error, MSE) of DFT-based channel estimation

In order to avoid this problem, it is necessary to cyclically shift the CIR of each UE to the edge of the time domain axis and then perform the multipath search of the UE. This method causes each UE to complete the IDFT / DFT operation separately, and the calculation complexity is high.

Images