École de technologie supérieure
Resource allocation designs for 5G wireless backhaul networks
Abstract
dc:descriptionIn recent years, network densification has progressively been shown as a prominent technological candidate to effectively serve the ever-increasing demands of wireless devices. Despite greatly inheriting many benefits from the conventional small cell architecture, dense network prompts a tremendous requirement of excessive fiber backhaul connections between the small cell base stations and core networks. This leads to a costly and impractical network deployment perspective. To overcome this issue, wireless backhaul (WB) technology is proposed as a cost-effective and viable replacement of wired backhaul to wirelessly transport data to the intermediate small cell nodes and subsequently serve local users on the access links. Contrarily, WB must guarantee the quality of small cell’s backhaul rate as comparatively high as the wired one in order to sustain the reliable wireless access (WA) communications between small cells and the local users. Technically, it is required that WB rate must always exceed the WA rate. However, WB rates are generally limited since the backhaul signals, when propagating through wireless channel, spatially attenuate and suffer severe interference from many other concurrent transmissions. This creates a certain challenge to sustain the newly introduced WB HetNets at their expected performance, which calls for an appropriate resource allocation design in WB HetNets. This dissertation aims at designing and performing resource allocation in WB HetNets under an explicit consideration of WB and WA transmission relationship. This relationship requires that the WB rate must always be greater than or equal to the WA rate so as to maintain the viability ofWBHetNets. Via jointly optimizing the multi-dimensional resource allocation such as transmit (or/and) receive beamforming, power allocation, spectrum partitioning, small cell or user scheduling, and data management, etc., this dissertation investigates the WB HetNets performance via the following two parts. In the first part, Chapters 2 and 3 mainly study the resource allocation of the WB small cell two-tier HetNets operated under the reverse time devision duplexing (RTDD) system. In particular, Chapter 2 highlights the coupled behavior of uplink and downlink transmissions resulting from the RTDD to design a joint radio resource allocation of beamforming, power, and spectrum partitioning which maximizes small cell sum rate on both uplink and downlink sides. On the other hand, Chapter 3 only focuses on the downlink design of transmit beamforming and power allocation which maximizes the access energy-efficiency, where spectrum partitioning is given. In both work, a unified solution framework based on successive convex approximation (SCA) and second order cone programming (SOCP) techniques are invoked to develop an iterative low-complexity algorithm to attain a high-quality sub-optimal solution efficiently. Extensive numerical results are presented to corroborate the superiority of the proposed designs compared to conventional work in terms of system performance gain and algorithm’s convergence behavior. In the second part, Chapters 4, 5, and 6 concentrate on deploying novel ideas to improve the WB small cell two-tier HetNets performance on top of formulating a resource allocation problem to optimize the objective of interest. Specifically, Chapter 4 considers a local buffer at each small cell which can flexibly store and release backhaul data to transmit to the users when needed. With many concurrent backhaul and access transmissions occurring through multiple time slots and assuming channel condition varies after each time slot, the problem is how to optimally control the dynamic of buffer queues via jointly optimizing the radio resource allocation together with data management in the buffer which maximizes the total throughput of small cell user over time. On the other hand, Chapter 5 approaches the WB HetNets enhancement by proposing a novel cooperative non-orthogonal multiple access (NOMA) technology to leverage both the achievable rates of WB and WA communications. To approach the solution, this work solves for a joint optimal solution of NOMA’s decoding order, small cell cooperation policy, transmit beamforming at the macro cell base station and power allocation at the small cell base station which maximizes the total small cell and macro cell user sum rate. Here, a similar efficient solution approaches from Chapters 3 and 4 are properly exploited to compute the solution of the formulated problems. Numerical results show the improvement and effectiveness of WB HetNets under the novel proposed scheme compared to the conventional designs. Towards this end, Chapter 6 presents a novel idea of replacing the traditional small cell base stations in WB small cell HetNets by unmanned aerial vehicles (UAVs). Beside the capability of transmitting and receiving signal to/from intended receivers like a small cell, UAVs can flexibly fly in space at any three-dimensional (3D) coordinate to increase the signal strength. This potentially provides more degrees of freedom to improve the overall performance with WB UAV networks compared to the conventional WB small cell HetNets. Therefore, Chapter 6 aims at jointly optimizing the radio resource allocation together with the UAVs’ locations which maximizes the user sum rate. Finally, a novel, more advanced with lower complexity algorithm based on the difference of convex (DC) programming are developed to compute the solution of the formulated problem. Achieved results show that the new system of WB UAV networks, though optimizing the UAVs’ location, can significantly boost the network performance.
Degree
thesis:*- Name thesis:degree_name
- Thèse de doctorat électronique
- Level thesis:degree_level
- phd
- Grantor dc:publisher
- École de technologie supérieure
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Nguyen, Tri Minh
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- Identifier
- INSERT-YOUR-LIBRARY-ID-HERE2198