Joint Channel Estimation For MASSIVE MIMO Systems Based On Tensor Compressive Sensing

Massive Multiple-Input Multiple-Output(MIMO)is one of the key technologies for the next generation mobile communication systems.Compared to the existing MIMO systems,the Massive MIMO systems have the advantages of higher spectrum efficiency,more stable communication performance,and more users.However,the premise of these advantages is that channel estimation should be performed to obtain accurate channel state information(CSI).With the increase of the number of antennas,the complexity of the channel estimation for Massive MIMO systems continues to increase.The existing channel estimation algorithms have low reconstruction accuracy and consume a large amount of pilots occupying the bandwidth resources,which can no longer satisfy the requirements of accuracy and simplicity for Massive MIMO systems.In order to solve these problems,this paper proposes a joint-channel estimation algorithm based on tensor compressive sensing for Massive MIMO systems.The specific contents are as follows.Firstly,existing channel estimation algorithms based on traditional compressive sensing still use the method of channel-by-channel estimation for the current high-dimensional CSI in multi-user Massive MIMO systems.It not only results in a large amount of pilot consumption,but also more complex calculations.This paper researches the channel estimation problem from the perspective of high-dimensional data processing based on tensor theory.The model of channel estimation in tensor mode is established,and the mathematical model of compressive measurement for three-dimensional tensor by two-dimensional pilot sequences is proposed to realize the matching of pilot sequences in fixed transmission mode and three-dimensional CSI.In the downlink channel estimation of the multi-user Massive MIMO system,a corresponding CSI feedback model is proposed,and then the CSI is reconstructed using the pseudo inverse matrix of the measurement results in tensor compressive sensing.Secondly,an adaptive truncated tensor noise reduction algorithm is proposed to reduce the effect of noise on the channel estimation accuracy for the noise interference is common in wireless channels.In this algorithm,the singular value decomposition is performed on the nuclear tensor obtained from compressive measurement in the above-mentioned tensor mode channel estimation model,and the threshold is obtained through the mapping relationship between the noise variance and the singular value.Then the singular value threshold selection algorithm is extended to three dimensions to complete the optimal truncation of nuclear tensor.The noise reduction algorithm is embedded into the reconstruction part of the above channel estimation algorithm,for reducing the influence of noise on the channel estimation accuracy.Finally,in this paper,the system model and channel estimation model of multi-user Massive MIMO AF relay system in tensor mode are built,and the joint channel estimation algorithm is employed to this model to complete channel estimation for multi-user Massive MIMO AF relay communication system.In addition,the channel estimation algorithm proposed in this paper is compared with the two popular algorithms.The experimental results show that proposed algorithm has higher estimation accuracy under the condition of lower pilot overhead and feedback burden,and the proposed algorithm has low computational complexity because it is non-iterative.