NOMA in Cell-Free mMIMO Systems with AP Selection

In this paper we proposed access point (AP) tiered in cell-free (CF) massive multiple input multiple output (mMIMO) non-orthogonal multiple access (NOMA) system with AP selection. Based on AP tiered, we proposed the new method of user grouping and power allocation in the system. In current works on CF mMIMO systems, AP selection is getting more and more attention. AP selection means the system is user centric and the central processing unit (CPU) needs to determine whether one AP can serve a user by considering the channel condition between them. So users in the system can choose the corresponding AP subsets to provide service. As a result, different users may be served by different AP subsets and the traditional user grouping and power allocation method in NOMA can't adapt to these users with inconsistent serving APs. As each user chooses an AP subset to provide service, the APs are clustered and these APs must be in different intersections of the AP clusters, we can note these relations as AP tier and group. Based on AP tiered, the user grouping and power allocation problem in the CF mMIMO NOMA system with AP selection can be clearly solved. Through simulation, we can prove that, by performing superposition coding (SC) as the corresponding AP group, NOMA can better adapt to the proposed system. And the optimization of power allocation in the proposed system can be simplified.