Publication:
Service Competition and Data-Centric Protocols for Internet Access

Simple item page
dc.contributor.advisorTilman Wolf
dc.contributor.advisorMichael Zink
dc.contributor.advisorJay Taneja
dc.contributor.advisorAnna Nagurney
dc.contributor.authorTeixeira, Thiago
dc.contributor.departmentUniversity of Massachusetts Amherst
dc.date2024-03-28T16:42:54.000
dc.date.accessioned2024-04-26T15:35:08Z
dc.date.available2024-04-26T15:35:08Z
dc.date.submittedSeptember
dc.date.submitted2019
dc.description.abstractThe Internet evolved in many aspects, from the application to the physical layers. However, the evolution of the Internet access technologies, most visible in dense urban scenarios, is not easily noticeable in sparsely populated and rural areas. In the United States, for example, the FCC identified that 50% of the census blocks have access to up to two broadband providers; however, these providers do not necessarily compete. Additionally, due to the methodology of the study, there is evidence that the number of actual customers without broadband access is higher since the FCC considers the entire block to have broadband if any customer in a block has broadband. Moreover, the average downstream connection bandwidth in the United States is 18.7 Mbps, according to the Akamai State of the Internet report, which places the US in the 10th position in the global rank. It’s worth noting that modern applications such as Ultra High Definition (UHD) video streaming requires a bandwidth of at least 25 Mbps. Newer applications such as virtual reality streaming require at least a 50 Mbps bandwidth. Additionally, urban scenarios are dominated by monopolistic and duopolistic markets, whereby network providers have little incentives to offer innovative services. In this work, we propose an open access network infrastructure along with a novel Internet architecture that allows dynamic economic relationships between users and providers through a marketplace of network services. These economic relationships have a finer granularity than today’s coarse and lengthy contracts, allowing higher competition and promoting innovation in the access market. We develop an agent-based simulator to evaluate our proposed network model and its various competition scenarios. Our simulations show that competition greatly benefits users and applications, creating the necessary incentives for providers to innovate while also benefiting consumers. The trend that resulted in sparsely populated areas lagging of the latest innovations in the access networks is also observed in wireless access networks, where the investments are focused on densely populated areas. Moreover, the rapidly increasing number of mobile devices coupled with the increasingly bandwidth demanding applications are posing a significant challenge to cellular network operators that have to increase OPEX/CAPEX and deal with higher complexity in their networks. The advances in the access technologies that brought higher speeds and lower latency also reduced the area of coverage of cellular base stations. To cope with the increase in traffic, cellular network operators have been deploying more base stations. In addition, cellular providers have adopted “all-you-can-use” price models, which led users to ramp-up their usage, further worsening congestion in the network. To address this issue, we propose a scheme that uses Device-to-Device (D2D) communication along with Information-Centric Networking (ICN) to offload traffic from cellular base stations. Then, we build on this scheme and propose a cross-layer assisted forwarding strategy to enhance communication in the MANET. In D2D communication, users can retrieve content directly from their nearby peers. However, this type of communication poses challenges to the current connection-oriented communication model, as devices can move in and out of the communication range at any time, constantly changing routing state, and nodes are subject to hidden and exposed terminal problems. ICN addresses some of these issues with inherent support for transparent caching and named content retrieval, making the network more resilient to disconnections. Our proposed scheme can offload up to 51.7% of the contents from the backhaul cellular infrastructure when requesting the content from nearby peers first. Finally, we combine the concepts of the marketplace, D2D communication, and ICN to propose a platform for decentralized and opportunistic communication that uses COTS radios to relay packets, extending the reach of the Internet to sparsely populated areas with low cost and without the lengthy contracts from commercial network providers. Our platform can potentially link the remaining part of the population that is not currently connected to the Internet.
dc.description.degreeDoctor of Philosophy (PhD)
dc.description.departmentElectrical and Computer Engineering
dc.identifier.doihttps://doi.org/10.7275/15202406
dc.identifier.orcidhttps://orcid.org/0000-0002-3597-0846
dc.identifier.urihttps://hdl.handle.net/20.500.14394/18025
dc.relation.urlhttps://scholarworks.umass.edu/cgi/viewcontent.cgi?article=2807&context=dissertations_2&unstamped=1
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.source.statuspublished
dc.subjectInternet
dc.subjectInternet access networks
dc.subjectInternet economics
dc.subjectWireless networks
dc.subjectInformation-centric networks
dc.subjectDevice-to-device communication
dc.subjectOpen-Access networks
dc.subjectDigital Communications and Networking
dc.titleService Competition and Data-Centric Protocols for Internet Access
dc.typeopenaccess
dc.typearticle
dc.typedissertation
digcom.contributor.authorisAuthorOfPublication|email:tteixeira@umass.edu|institution:University of Massachusetts Amherst|Teixeira, Thiago
digcom.identifierdissertations_2/1770
digcom.identifier.contextkey15202406
digcom.identifier.submissionpathdissertations_2/1770
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
tteixeira_phd_defense.pdf
Size:
4.88 MB
Format:
Adobe Portable Document Format
Name: tteixeira_phd_defense.pdf
Size: 4.88 MB
Kind: Adobe PDF
Collections
Dissertations