Rui Tang

User-Centric Joint Admission Control and Resource Allocation for 5G D2D Extreme Mobile Broadband: A Sequential Convex Programming Approach

Device-to-Device (D2D) communication has been recognized as a promising remedy to offload 5G extreme mobile broadband services by utilizing the favorable channel condition in proximity. In contrast to most literature emphasizing system-centric metrics, we optimize user-centric ones, including the number of admitted D2D links with satisfied quality-of-service requirement and power consumption, by integrating resource allocation with admission control. Due to the intractability of the above problem, we apply the sequential convex programming technique to achieve a reasonable approximation to the original optimum. Simulations verify the advantages of conducting user-centric optimizations and show the efficacy of our scheme.

Resource Allocation for Underlaid Device-to-Device Communication by Incorporating Both Channel Assignment and Power Control

Device-to-Device (D2D) communication can greatly enhance the system throughput (ST) and individual quality-of-service (QoS) experience by reusing the spectrum of cellular network. But due to dense spectrum sharing, severe co-channel interference may be imposed between cellular links (C-links) and D2D links (D-links). So, to fully utilize the gain of D2D communication, we attempt to maximize the ST of D-links with guaranteed individual QoS requirement for both types of links by joint consideration of channel assignment (CA) and power control (PC). Specifically, particle swarm optimization (PSO) is explicitly adopted to optimize CA, but when deriving the fitness value for each CA particle, underlaid PC optimization will be implicitly performed for throughput maximization on each channel, together with a gradual removal (GR) scheme to maintain individual QoS requirement and admit the maximum number of D-links into the system. Thanks to the numerical results, we demonstrate the superiority of our scheme in terms of ST and system capacity (SC), i.e., The number of admitted D-links, when comparing with the ones in literature.

Joint optimization of channel allocation, link assignment and power control for device-to-device communication underlaying cellular network

Device-to-Device (D2D) communication has been proposed to facilitate cellular network with system capacity (SC) and quality of service (QoS). We consider the design of link assignment (LA), channel allocation (CA) and power control (PC) in D2D-aided content delivery scenario for both user fairness (UF) and system throughput (ST) under QoS requirement. Due to the complexity of the problem, we decompose it into two components: CA is formulated from graph perspective to mitigate severe co-channel interference, which turns out to be the Max K-cut problem; LA and PC are jointly optimized to utilize the gain achieved from CA for supreme performance, and specifically, genetic algorithm (GA) is adopted to optimize LA, but when deriving the fitness of each chromosome, PC optimization will be involved. Thanks to numerical results, we elucidate the efficacy of our scheme.

User-Centric QoS Provisioning for Heterogeneous D2D Multimedia Flows with Dense Spectral Reuse

With the increasing of heterogeneous short-range multimedia services in cellular networks, device-to- device (D2D) communication is viewed as a promising remedy to offload the above traffic volumes by taking full advantage of the favorable channel condition in proximity. Unfortunately, severe co-channel interference would be introduced among D2D links reusing the same resource. Besides, each D2D link is in need of a stringent quality-of-service (QoS) requirement, and the system capacity may not accommodate all of the above traffic volumes. Different to the majority of the existing works, we address the user-centric metrics and aim to maximize the number of admitted D2D links with satisfied QoS requirements within each scheduling period by QoS provisioning, including admission control (AC) and slot allocation (SA). Specifically, the above two components maintain a hierarchical structure: The AC concerns the spectral reuse at each slot and brings in the definition of maximal feasible pattern, i.e., the maximal number of D2D links reusing a same resource under a given bit- error rate constraint. The above problem can be transferred into a mixed-integer linear programming (MILP) which is generally NP-hard. Instead of using branch-and-bound method, we view it from the perspective of graph theory and put forward an iterative interference graph based scheme to strike a balance between performance and complexity. Upon the above results, the SA can be relaxed into a convex- cardinality problem and is also equivalent to an MILP. To get a tractable solution, we approximate the cardinality with l1-norm heuristic, and apply the iterative weighted l1-norm heuristic to maintain polynomial-time complexity at minor performance loss. Thanks to the simulations, we illustrate the necessity to conduct user-centric optimizations and elucidate the efficacy of our QoS provisioning scheme.

Energy-Efficient Resource Allocation for 5G Full-Duplex Enabled Device-to-Device Communication

Full-duplex (FD) transmission and device-to-device (D2D) communication underlaying cellular networks are viewed as key components in the fifth generation (5G) systems, and their benefit can be magnified if they are effectively combined together. However, the self- interference at FD-enabled mobile terminals and the mutual interference between D2D links and cellular links are inevitable due to spectral reuse. Furthermore, there is also a need to strike a balance between spectral efficiency and energy conservation. Therefore, we aim to optimize energy efficiency for FD D2D underlaying system by incorporating power control (PC) and channel assignment (CA) under additional quality-of-service requirements of the existing cellular links. With the property of fractional programming, we transfer the original problem into solving a sequence of parametric programs in subtractive form. Besides, for each of the above parametric programs, it can be decomposed into the PC subproblem and the CA subproblem. Moreover, the former turns out to be a difference-of-concave (D-C) programming, which can be well addressed by the sequential convex optimization technique. Based on the above optimum, the latter reduces to the assignment problem, which can be perfectly resolved by Hungary algorithm. In addition, we discuss the time complexity and implementation issue of our scheme. Thanks to the simulations, we observe the benefit of FD D2D underlaying system compared to its half-duplex counterpart and elucidate the efficacy of our scheme.