Radio Resource Management For Heterogeneous Networks Under Hybrid Access Of Human And Machine Type Devices

Compared with the fourth generation mobile communication system (4G),the fifth generation mobile communication system (5G) require performance improvement in network rate, user experienced rate, devices access density,user mobility and end-to-end latency. ITU has already confirmed three ma-jor communication scenarios in 5G system, i.e. enhanced mobile broadband(eMBB), ultra reliable and low latency (uRLLC) and massive machine-type communications (mMTC). Multiple paradigms exist in 5G system, including 1) diversified application requirements (e.g. data rate, reliability, latency, en-ergy efficiency, cost, etc.); 2) multiple radio access technology (e.g. 3G, LTE,WiFi, mm Wave, etc.); 3) various kinds of devices as users (e.g. mobile phones,sensors, smart meters, vehicles, drones, etc.); 4) multiple dimension available network resources (e.g. spectrum/computing/cache resources, etc.); 5) multiple types of communication scenarios (e.g. virtual /augmented reality, smart trans-portation/grid/home/health, etc.). Thus, human-type communication (HTC) de-vices and MTC devices will coexist in the network and transmission data in 5 G scenarios, Different type users with diversified communication requirements will challenge the service ability of the existing network.Cellular technology has been regarded as one of the most promising meth-ods to provide data service for hybrid access of HTC and MTC devices, which can provide high spectral efficiency, reliable and secure wireless links for users based on its comprehensive users access and resource scheduling scheme. How-ever, the design concept of traditional cellular network aims to serve HTC de-vices, which is not suitable for MTC services. Taking LTE for example, when a plethora of MTC devices require access to the eNode-B, access collision and heavy signalling overhead are inevitable because of the limitation of access re-sources. Moreover, for MTC devices with small burst data packets, resource scheduling for HTC will decrease spectral efficiency. Due to the spectrum re-source limitation, effective resource management is needed for HTC and MTC devices. In this paper, we analyze the radio resource management problem for HTC and MTC hybrid access in cellular network. By proposing effective and reasonable resource management methods, we aim to enhance the network per-formance, while ensuring the quality of services (QoS) of both HTC and MTC devices. The major contributions of the paper are as follows:For HTC in the heterogeneous cellular network, we novelly propose the optimal users association and resource allocation methods by applying down-link and uplink decoupling. The user association probability with base stations(BSs) in different tiers and the network ergodic rate is theoretically derived.Moreover, we use coalition game to jointly optimize user association and down-link/uplink resource partition for BSs. Evaluation results testify a 60% rate performance enhancement of our proposed method over other user association rules. Further, we proposed discrete location aware interference model and optimal resource allocation for D2D users underlaying cellular network. The total D2D rate is maximized, while satisfying cellular and D2D users data rate threshold. The performance of proposed scheme outperforms other location-aware D2D resource allocation schemes by 20%-50%.For HTC and MTC coexistence in the cellular network, we proposed a two-stage resource management scheme to maintain the HTC and MTC traffic stable. In the first stage, we novelly utilize the prey-predator model to analyze the HTC/MTC traffic dynamical variation, and derive the conditions of the sys-tem equilibriums and their local and global stabilities. The second stage scheme tackles the optimal resource allocation for group-based MTC devices. Closed-forms of the optimal results are derived, which shows that by only allocating resources to two MTC groups can maintain the total MTC traffic stable, while achieving the maximal system utility. Evaluation results indicate that a reason-able resource partition for HTC and MTC can make their traffic converge to equilibrium points or a limited cycle, which are global stable. Moreover, only 5% percent of resource partitioned for MTC can support 2 Mbit traffic.For vehicle to everything (V2X) communication in the cellular network,we analyze on the optimal vehicles association and resource allocation under latency and reliability requirements, which are transferred to outage probabil-ity constraint and the limited number of resources allocated to vehicle users per time slot. For vehicle-to-roadside unit (V2R) and vehicle-to-vehicle (V2V)using D2D, maximal matching and stable matching in graph theory is used to obtain the maximal number of vehicle users associated to network. Evalua-tion results show that compared with other popular association and resource allocation methods, under the reliability and latency constraints, our methods can provide extra 60% number of vehicles communicating with RSUs and cer-tain association performance for D2D vehicles, without influencing the cellular users' performance.Overall, the dissertation provides resource management methods for HTC and MTC hybrid access in multiple 5G communication scenarios, which im-prove the resource utilization efficiency and network performance, while both guaranteeing the QoS of MTC and HTC devices.