Combined channel and carrier frequency offset estimation method for millimeter wave MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) system

Abstract

The invention relates to the field of wireless communication systems, in particular to a combined channel and carrier frequency offset estimation method for a millimeter wave MIMO-OFDM (Multiple InputMultiple Output-Orthogonal Frequency Division Multiplexing) system. The method mainly comprises the following steps of: S1, converting a combined channel and carrier frequency offset matrix into a form which conforms to the atomic expression of a constructed two-dimensional sequence and is shown in the specification; S2, estimating a joint channel and carrier frequency offset matrix shown in thespecification by adopting a sparse signal reconstruction algorithm based on sequence atom norm minimization; S3, giving an estimated value shown in the specification, based on compressibility of carrier frequency offset, decomposing to obtain a transceiving direction, channel gain and carrier frequency offset component of the channel, and then directly obtaining a channel matrix by utilizing the estimated channel transceiving direction and channel gain; and S4, performing performance evaluation on the estimated channel matrix and the carrier frequency offset vector by adopting a normalized mean square error evaluation standard. According to the method, the sparsity of channel parameters can be highly and closely approximated in a continuous angle space domain, the base error matching problem is avoided, and then the channel estimation accuracy is improved.

Description

Technical field [0001] The present invention relates to the field of wireless communication systems, and in particular to a method for estimating a joint channel and carrier frequency offset of a millimeter wave MIMO-OFDM system. Background technique [0002] Millimeter wave large-scale multi-antenna technology will be applied to next-generation 5G communications. In a multi-antenna system, channel information is a necessary condition for reliable signal transmission and reception, which is especially important for beamforming in millimeter wave communication systems. The transceiver antenna is usually driven by an oscillator to up / down-convert the carrier signal, but the inevitable jitter of the oscillator may cause carrier frequency offset and phase noise, resulting in phase error. [0003] In the millimeter wave multi-antenna system, the signal-to-noise ratio of the received signal before beamforming is low and the channel matrix dimension is high, all of which bring huge chall...

AI Technical Summary

Problems solved by technology

In view of the high complexity of the previous two methods extended to broadband systems, N.J.Myers et al. also published Message passing-based joint CFO and channel estimation in millimeter wave systems with one-bit on arXiv preprint arXiv:1803.09012, 2018. The matrix stretching technique is no longer applicable in ADCs (1-bit analog-to-digital conversion based on message passing and channel estimation), and the joint estimation problem is modeled as a quantized bilinear optimization problem and A parallel bilinear Gaussian approximation message passing algorithm is used to solve

[0005] Existing tensor-based and l 1 The two joint estimators based on norm minimization and approximate message passing algorithm are essentially established by matrix stretching technology. When the joint channel and carrier frequency offset estimator is applied to broadband systems, the complexity of the algorithm will increase greatly. In turn, it is easy to cause intolerable training delay and energy consumption

Moreover, even existing estimators that model the joint estimation problem as a quantized bilinear optimization problem and employ a parallel bilinear Gaussian approximation message-passing algorithm to solve it have reduced computational complexity

However, it is worth noting that the above three joint estimators are all based on discrete compressed sensing theory, which discretizes the channel sending and receiving direction space domain and the carrier frequency offset frequency domain into a large number of grid points, while the actual channel sending and receiving directions take values ​​in the continuous angle domain , the carrier frequency offset also takes values ​​continuously within the frequency range, which can easily cause base mismatch and seriously reduce the estimation accuracy

Images