Research On Resource Allocation With Statistical Delay QoS Guarantee In Full Duplex OFDMA Systems

With the continuous emergence of new network technologies,such as the mobile Internet and the Internet of Things,people's demand for communications services has grown exponentially,and wireless spectrum resources have become increasingly scarce.Wireless communication full-duplex technologies can support two-way communication at the same time with the same frequency,being an effective way to improve spectrum efficiency.Therefore,the wireless communication full-duplex technology has become one of the hot issues in the field of wireless communications at present.In the 5G era,wireless communication network services have the features of diversity and heterogeneity,different service types have different requirements for delay.One of the core issues of the next-generation wireless communication network is how to use the limited frequency spectrum resources to support different types of service QoS(Quality Of Service).Considering the time-varying random feature of wireless channels,statistical delay QoS guarantee have attracted widespread attention in the academic and industrial communities.Statistical delay QoS is introduced by the effective capacity theory,and it includes two elements: the delay bound and the probability of delay bound violation.The effective capacity theory constrains the delay QoS from the perspective of link layer,which reflects the network throughput performance under statistical delay QoS guarantee.Therefore,based on the theory of effective capacity,it is of great significance to study the statistical delay QoS guarantee in 5G wireless full-duplex communication networks.Based on the effective capacity theory,convex optimization theory,Hungarian algorithm,etc.,this paper studies the problem of subcarrier allocation and power allocation in OFDMA(Orthogonal Frequency Division Multiple Access)systems with and without nodes having full-duplex capability.Considering that the full-duplex capability is not perfect,this paper introduces the self-interference residual coefficient.This paper uses the effective capacity to characterize the throughput of the system,and derive the effective capacity model of the system based on the effective capacity theory,and furthermore construct the system's joint subcarrier allocation and powerallocation problem as an integer programming problem that maximizes the system's effective capacity.In order to make the integer planning problem easier to deal with and to reduce the complexity of the problem solving,this paper first executes initial power allocation for each subcarrier,reducing the dimension of the integer planning problem,and divide it into two sub-problems: subcarrier allocation and power allocation problem respectively.For power allocation problem,this paper proposes a QoS-driven power allocation strategy.As for the subcarrier allocation problem,for the OFDMA system with nodes having full-duplex capability,an improved Hungarian algorithm is proposed to solve the problem.For an OFDMA system where the node does not have full duplex capability,this paper models it as the three-dimensional matching problem among uplink nodes,subcarrier and downlink nodes.Since the interference between the uplink and downlink nodes is assumed to be the same,the dimensionality reduction can be further executed,divided into binary matching problems of uplink nodes and subcarriers and downlink nodes and subcarriers.For the matching of uplink nodes and subcarriers,limited by the "subcarrier continuity" of uplink resource allocation,this paper proposes an iterative maximum extension(IME)algorithm to solve it;for the matching of downlink nodes and subcarriers,the improved Hungarian algorithm is still used.The works of this paper provide research basis for resource allocation and statistical QoS guarantee in single cellular OFDMA systems.