Text

theses

The next-generation wireless network (5G/6G) aims to provide deep coverage, high capacity, and low latency for emerging applications ranging from Augmented Reality (AR), Virtual Reality (VR), emergency networks, and other mission critical 5G services. However, 5G/6G introduces new challenges at multiple layers of the protocol stack due to the lack of interoperability and heterogeneity. There is a need to fundamentally revisit and redesign the traditional protocol stack to support these newer services in the future network. In this thesis, we will address problems arising from 5G heterogeneous deployments specifically those related to spectrum management, mobility management, and the transport layer used for mmWave services. The proposed solutions are based on spectrum models, cross-layer analytics, and mobility protocols using the named-object network architecture and associated distributed algorithms. Next generation wireless services and applications will increasingly rely on dynamic spectrum access (DSA) methods that can manage spectrum resources rapidly and efficiently. In this context, chapter 2 discusses a novel spectrum management architecture and algorithm design that leverages Spectrum Consumption Models (SCMs) which offer a mechanism for RF devices to announce their intention to use the spectrum and their needs in terms of spectrum protection. In both our experimental setup and simulation environment, RF devices use SCMs to determine compatibility with existing devices and to dynamically configure their transmission parameters. A novel SCM based deconfliction algorithm and spectrum access methods are developed for large scale wireless network environments and evaluated using a custom simulation platform in terms of computation time, the efficiency of spectrum allocation, and the number of device reconfigurations due to interference. The simulation results and experimental evaluation on the ORBIT/COSMOS testbed validate the benefits of using SCMs and their capabilities to perform fine grained spectrum assignments in dynamic and dense communication environments. Chapter 3 describes a novel distributed mobility management (DMM) scheme based on the “named-object” information centric network (ICN) architecture in which the routers forward data based on unique identifiers which are dynamically mapped to the current network addresses of a device. The work proposes and evaluates two specific handover schemes namely, hard handoff with rebinding and soft handoff with multihoming intended to provide seamless data transfer with improved throughput during handovers. The proposed handover schemes are evaluated with respect to RTT and throughput using system simulation along with proof-of-concept implementation on the ORBIT testbed. Experimental results are presented to validate the proposed ICN inspired mobility management protocol. Finally, in chapter 4, we conclude with a focus on the transport layer necessary for future wireless networks. The proposed end-to-end protocol uses cross-layer feedback with in-network transport proxy for the fast delivery of data over mmWave channels arising in emerging 5G networks. Recent measurement studies of mmWave channels in urban micro cellular deployments show considerable fluctuations in received signal strength along with intermittent outages resulting from user mobility. This results in significant impairment of end-to-end data transfer speed when conventional TCP is used to transport data over such mmWave channels. To overcome this challenge, a new transport protocol capable of adapting to rapidly fluctuating MAC layer throughput whilst achieving high throughput efficiency is required for future large scale mmWave deployment. A specific transport layer protocol called “mmCPTP” is proposed and evaluated using emulation and experimental evaluation. Significant performance gains over TCP are demonstrated.

http://dissertations.umi.com/gsnb.rutgers:12469

Ph.D.

Includes bibliographical references

doi:10.7282/t3-2eh0-gc77

The author owns the copyright to this work.