Multiuser Detection Algorithm In The Direction Of Noise Enhanced Subspace In Massive MIMO Systems

Massive MIMO has become one of the key techniques in the 5th mobile communication, as it can exploit the frequency efficiently and improve the throughput of the system. When combined with multiuser MIMO, a base station in the massive MIMO system can communicate with a number of u sers. However, it is an urgent problem for the base station to perform uplink multiuser detection, as the user equipment can not perform joint coding, and it is difficult to obtain perfect channel state information.In this paper,the massive multiuser MIMO system and several multiuser detection algorithms are first introduced and summarized. In massive MIMO systems, the detection algorithms that are with low complexity and can be performed in parallel are suitable, so in this paper MMSE-based eigenvector subspace searching algorithm is proposed. The error of MMSE detection is first analyzed, then the noise vector is projected to the normal eigenvector space of the noise enhanced direction matrix. Parallel searching is performed to amend the detection error of MMSE, and log likelihood function is used to evaluate the accuracy of the detected result.To suit in the massive MIMO, the computational complexity of the MMSE-based eigenvector subspace searching algorithm is red uced. An approximated log likelihood function is employed, and only fixed number of candidates with small approximated log likelihood function values are used for further calculation, and it is called the noise enhanced subspace searching algorithm. Some computation can be ignored, thus reducing the complexity. To lessen the effect caused by the imperfect channel state information, the noise enhanced subspace searching algorithm is modified with the knowledge of the second statistics of the estimated channe l error.Computer simulations show that the enhanced parallel interference cancellation algorithm can perform good detection performance in the uplink multiuser MIMO system. The algorithm reduces the complexity in massive MIMO systems greatly. When the channel state information is imperfect, the enhanced parallel interference cancellation algorithm can still achieve better performance than some conventional detection algorithm.