Network Coding Empowered Transmission Scheduling In NOMA-based Multi-BTS Wireless Networks

Currently,the 5th Generation Mobile Networks(5G)is facing many new challenges.In order to provide users with better services,it must have the ability to carry a large number of wireless devices,and to provide latency performance guarantees simultaneously.On one hand,due to its supporting on parallel transmissions,PD-NOMA(Power Domain Non-Orthogonal Multiple Access)provides a feasible method to reduce the network delays.On the other hand,the network coding theory can compress the number of transmitted packets by effectively encoding the packets at the network layer,and thereby compresses the amount of network packets.How to effectively combine PDNOMA and network coding theory to minimize the number of transmitted packets so as to reduce transmission delay requires further studies.In this dissertation,under the premise of known user packet requirements,the downlink transmission scheduling problem of PD-NOMA wireless networks for minimizing network delays based on network coding in multi-BTS network is investigated.Firstly,based on the PD-NOMA and the network coding theory,we formulate the problem in the form of discrete optimization.Then,by presenting an original reduction proof,the classical 3DM(3-Dimension Matching)problem is reduced to the primitive problem and thereby the primitive problem is proved to be NP-Complete,which is considered as the main novelty of this dissertation.To obtain a highly efficient heuristic algorithm,it is revealed that the existence of MCUP(Mutual Coupled User Pair)is the root cause of performance enhancement brought by the network coding theory.while PD-NOMA can further lower the latency of any transmission strategy based on MCUP.So,based on PD-NOMA and by aiming at maximizing the number of phases and the number of MCUPs respectively,the heuristic strategy presented greedily selected transmission BTSs(Base Transmission Station)and then the BTS-message pairing strategy slot by slot,so as to decrease network delay.Performance evaluation reveals that,under multiple network topologies and typical communication parameters where the number and density of BTS vary,the average frame length decreases linearly with the increase of the channel attenuation factor.Besides,compared with the traditional TDMA(Time Division Multiple Access)scheme,the frame length is reduced by 24.5% on average and even 32.7% in the optimal case.It is concluded that the proposed algorithm realizes low-latency downlink network transmission in multiBTS scenarios.