A Study On Spectrum Resource Allocation Algorithms For D2d Communication Underlaying Cellular Networks

Device-to-Device(D2D)communication is a new wireless communication technology which promises a number of attracting features such as offloaded BS traffic load,decreased communication delay,extended device battery lifetime,and increased spectral efficiency.Spectrum resource allocation for D2D communication underlaying cellular networks plays an important role in improving network transmission performance,a network's quality of service(QoS)and users' quality of experience(QoE).On the basis of the state-of the art in D2D communication technology,this thesis studies the spectrum resource allocation problem in D2D communication underlaying cellular networks,and proposes three spectrum resource allocation algorithms for D2D communication taking into account the QoS and QoE issues.Firstly,a QoS-aware resource allocation(QARA)algorithm is proposed for addressing the QoS issue.In performing spectrum resource allocation,the QARA algorithm estimates the channel gains and data rates for both cellular users and D2D users based on the users' location information.For each D2D users,it establishes a cellular user candidate set and determines its priority in resource allocation.Based on the priority,it allocates spectrum resource blocks for each D2D user and selects those sharable spectrum resource blocks from the D2D user's cellular user candidate set in allocation.The simulation results show that the proposed QARA algorithm can achieve a better network performance in terms of the satisfaction ratio,the average D2D user data rate,and the overall system throughput as compared with an existing resource allocation algorithm for D2D communication.Secondly,a QoE-aware resource allocation(EARA)algorithm is proposed for addressing the QoE issue.The EARA algorithm takes the service completion time as a main factor for evaluating users's QoE.In performing spectrum resource allocation,it divides time into a sequence of fixed-length timeslot to estimate the service completion time of a D2D user pair and introduces a parameter QoEjk to mearsure D2D user pair Dj's QoE in timeslot k.In each timeslot,the EARA algorithm allows a D2D user pair with a smaller QoEjk value to have a higher priority to obtain spectrum resources.Meanwhile,it introduces a QoE model to estimate a D2D user pair's satisfaction to its QoE.The simulation results show the proposed EARA algorithm can achieve a better performance in terms of the average service completion satisfaction ratio,the average QoE of D2D pairs,and the overall system throughput as compared with a max-throughput resource allocation(MTRA)algorithm for D2D communication.Finally,a battery energy and data rate aware resource allocation(BARA)algorithm is proposed for addressing QoE issue.The BARA algorithm takes the battery energy and data rates as the main factors for evaluating user's QoE.In performing spectrum resource allocation,it introduces a battery energy model based on the Peukert Law to predict the battery energy consumption of a user device.Meanwhile,it introduces a QoE model based on the classic MOS model and a logistics function to measure a D2D user pair's satisfaction to its QoE.In each timeslot k,the BARA algorithm introduces a satisfaction function q(k)to measure a D2D user pair's satisfaction to its QoE in the current timeslot and allows a D2D user pair with a smaller q(k)value to have a higher priority to obtain spectrum resources.The simulation results show the proposed BARA algorithm can achieve a better performance in terms of the average QoE of D2D pairs and the average battery energy consumption in the network without obviously affecting the overall system throughput.