| Cloud radio access network(C-RAN)is a new wireless network architecture.The network architecture consists of several remote ratio heads(RRHs)and a Baseband Unit(BBU)pool.The RRHs transmit the received signals to the BBU pool through high-bandwidth backhaul links,and then carry out centralized processing.This architecture has become a potential wireless architecture because of its advantages such as reduced costs(capital expenditure,operating costs,energy consumption)and increased spectrum efficiency.On the other hand,Grant Free(GF)non-nonorthogonal multiple access(NOMA)technology is very suitable for massive machine type communications,which simultaneously serves multiple users in the same time/frequency resources in the way of non-orthogonal multiplexing.Therefore,this thesis studies the GF-NOMA technology in the C-RAN scenario.Firstly,a non-coherent NOMA transmission scheme based on index modulation(IM)in CRAN scenario is proposed.In this scheme,the system assigns multiple pilot sequences to each user,and only part of the pilot sequences are activated when the user information is transmitted,and the indexes of these activated pilot sequences are used to transmit data information.At the receiver,a non-Bayesian covariance-based maximum likelihood detection(CB-MLD)method is derived to identify active users and recover data information.Furthermore,we study the system implementation when amplifying forwarding(AF),quantize-and-forward(QF)and detect-and-forward(Det F)are used at RRHs.Numerical simulations show that the proposed scheme performs well in incoherent transmission scenarios.Secondly,the implementation of concatenated unsource random-access(URA)in C-RAN scenario is studied.Concatenated URA consists of inner compressed sensing(CS)codes and outer parity codes.For the inner CS code decoding in the C-RAN scenario,we present the CB-MLD implementation,where both AF and Det F are considered at RRHs.Finally,the effectiveness of the proposed algorithm is verified by computer simulation. |
