| dc.identifier.uri | http://hdl.handle.net/11401/78205 | |
| dc.description.sponsorship | This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree | en_US |
| dc.format | Monograph | |
| dc.format.medium | Electronic Resource | en_US |
| dc.language.iso | en_US | |
| dc.type | Dissertation | |
| dcterms.abstract | The performance of modern wireless networks depends on myriads of issues in different layers that interact in complex ways. They include applications, devices, network protocols and wireless channels. Their characteristics often vary widely even within the same network. Such complexity often makes traditional design analysis approaches inadequate. Inspired by the recent success of data-driven networking in handling complexity, this dissertation explores such approaches in modeling and optimization of wireless networked systems. Specifically, we look into a few emerging issues in two extreme ends of the wireless network protocol stack: (i) Managing Quality of Experience (QoE) of networked applications and (ii) Improving RF spectrum usage and RF signal-based device localization in the physical layer. Guaranteeing good QoE for the user is a challenge in mobile apps due to their diverse resource requirements and the resource-constrained, variable nature of wireless networks and heterogeneous mobile devices. Provisioning the network efficiently in the face of such constraints requires accurate modeling of the network's capacity. However, today's networks are surprisingly complex and highly non-transparent. Such factors make traditional whitebox modeling approaches infeasible. Instead, we use measurement-based approached to draw inferences about network capacity. We build two systems, Adapp and ExBox, that help devices connect to the best wireless network to maximize QoE for an individual user without disrupting the overall network's performance. Both Adapp and ExBox have been tested on real network testbeds. The second issue primarily relates to the problem of RF spectrum limitations. There is a growing realization that RF spectrum must be treated as a critical resource that is in limited supply in the face of growing demand for bandwidth from mobile applications. Just like any other resource with mismatched demand and supply, all steps towards better utilization have also increased the need for large scale spectrum monitoring. We address several problems in large scale spectrum monitoring. We use data-driven techniques to effectively combine model-driven and measurement-based approaches so that cost of monitoring can be optimized. We also contribute towards making the spectrum measurements themselves scalable by developing techniques to perform spectrum sensing on mobile devices. These efforts culminate building a spectrum database system called SpecSense that can schedule and collect measurements from a distributed system of spectrum sensors in order to estimate spatio-temporal patterns in spectrum availability. We also address a related issue concerning device localization from the network-side using passive, data-driven techniques. | |
| dcterms.available | 2018-03-22T22:39:18Z | |
| dcterms.contributor | Gupta, Himanshu | en_US |
| dcterms.contributor | Das, Samir R. | en_US |
| dcterms.contributor | Balasubranian, Aruna | en_US |
| dcterms.contributor | Djuric, Petar. | en_US |
| dcterms.creator | Chakraborty, Ayon | |
| dcterms.dateAccepted | 2018-03-22T22:39:18Z | |
| dcterms.dateSubmitted | 2018-03-22T22:39:18Z | |
| dcterms.description | Department of Computer Science. | en_US |
| dcterms.extent | 205 pg. | en_US |
| dcterms.format | Monograph | |
| dcterms.format | Application/PDF | en_US |
| dcterms.identifier | http://hdl.handle.net/11401/78205 | |
| dcterms.issued | 2017-08-01 | |
| dcterms.language | en_US | |
| dcterms.provenance | Made available in DSpace on 2018-03-22T22:39:18Z (GMT). No. of bitstreams: 1
Chakraborty_grad.sunysb_0771E_13266.pdf: 15037151 bytes, checksum: 8da4d3c2d88f1f1df2765dc2528e939b (MD5)
Previous issue date: 2017-08-01 | en |
| dcterms.subject | Computer science | |
| dcterms.subject | data-driven | |
| dcterms.subject | performance optimization | |
| dcterms.subject | wireless networks | |
| dcterms.title | Data-Driven Performance Optimization in Wireless Networks | |
| dcterms.type | Dissertation | |
