Research On Radio Resource Management For SCMA Based Downlink Of Wireless Cellular Networks

The new generation (5G) wireless networks extremely demand massive connectivity, higher throughput and stability, lower latency as well as better QoS (quality of service) to meet the explosive data traffic growth, which are great challenges to the key technologies in wireless communications, especially to the multiple access technology. Recently, NMA (non-orthogonal multiple access) schemes are hot topics because they can be overloaded through introducing controllable interference. As one of the new NMA schemes, SCMA (sparse code multiple access) mainly proposed by Huawei company can achieve better SE (spectral efficiency) and different system overloading factors by using different unique sparse multi-dimensional complex codebook to share the same time-frequency radio resources such as OFDMA tones. In addition, with Grant-Free multiple access and transmission mechanism in the uplink of wireless networks, SCMA can greatly reduce system latency and signaling overhead. As a result, SCMA is a promising technology to meet the requirements of 5G networks.Since RRM (radio resource management) plays important role in various wireless communication networks and correspondence between SCMA and the requirements of 5G communications, it is of great theoretical and realistic meaning to study the RRM schemes in the wireless systems based on SCMA.In this thesis, we mainly study the RRM schemes in the downlink of SCMA networks. The basic radio resource element will be redefined by several dimensions such as time, frequency, SCMA codebook and so on. In addition, the SCMA codebooks are closely related to the time-frequency resources, which can be presented by the factor graph and indicator matrix F. In consideration of the incidence relation mentioned above and the frequency-selective fading characteristic of wireless channels, the gained system performances may be different when the same codebook is assigned to different downlink user. However, almost all the present researches on SCMA ignored the situation, in which they just randomly match SCMA codebooks to users. Thus in our thesis, a codebook selection scheme based on wireless channel state information is used for the SCMA wireless downlink communication system model. The key idea of the algorithm is to match the codebook with the best channel sets to the user, so as to improve the sum rate or capacity of the system. Moreover, as one of the important parts of RRM, power allocation scheme based on the classical water-falling algorithm is also studied. The similar OFDMA downlink with same resource configuration and approximate RRM algorithms is introduced as a counterpart comparison with SCMA system. All the simulation results show that the sum rate (capacity) under our proposed scheme is better than others when compared to the random codebook selection and equal power allocation scheme which is commonly used in current SCMA researches. Besides, all the SCMA RRM schemes can gain a better capacity performance compared to corresponding OFDMA RRM schemes, which is significantly benefited from the SCMA overloading characteristic. Finally, we analyze the universalities of our RRM schemes in other three SCMA downlink situations (MU-SCMA, single user with more than one SCMA layer/codebooks and user QoS guaranteed situation). According to the analysis, our schemes are of good universalities and thus are very efficient.