Massive MIMO system signal detection algorithm

Base station distributes time$-$frequency resources to different users and provides services to a large

number of users. In the massive MIMO system of multi$-$cell and multi$-$user, when the cell terminal

sends the transmitted signal to the cell base station, the base station can detect the uplink signal

received by the space signature.

$7\mathrm{.}1$ Linear detection

When the cell base station is equipped with massive antenna array, the performance of MRC receiver

can reach the optimal linear receiver $($OLR$)$ performance if the condition of lower signal to

interference plus noise ratio $($SINR$)$ is satisfied, but it is lower than OLR under the condition of high

SINR. In the case of large interference, the performance of OLR will be optimized compared with that

of typical MMSE receiver system.

$7\mathrm{.}2$ Nonlinear detection

$(1)$Algorithm based on tree structure $($TB$).$ Spherical decoding $($SD$)[16]$ can be said to be a typical

nonlinear detection algorithm, and SD is actually a maximum likelihood $($ML$)$ decoder. The

disadvantage of the SD algorithm is that it only takes into account the points within a specific radius

and it should expand the radius to find any signalling point. In the existing low complexity TB$,$ adding

only the most valuable nodes can effectively reduce the search complexity.

$(2)$Random Step $($RS$)$ method. The principle of the algorithm is: select an initial vector, evaluate its

peripheral vector $N_{\text{N}\text{e}\text{i}\text{g}\text{h}},$ take MSE as the condition, select MSE as the minimum vector, repeat the

above process $N_{\text{i}\text{t}\text{e}\text{r}}$ times.