The following PAWR research projects were funded by the National Science Foundation following a Dear Colleague Letter solicitation for proposals in February 2020.
- ICN-WEN: Collaborative Research: SPLICE: Secure Predictive Low-Latency…
PI Panganamala R Kumar, Srinivas Shakkottai, I-Hong Hou
- NeTS: JUNO2: Resilient Disaster Communications in the Social-Media Era
PI K. K. Ramakrishnan
ICN-WEN: Collaborative Research: SPLICE: Secure Predictive Low-Latency Information Centric Edge for Next Generation Wireless Networks
Proposal ID: 1719384
(supplement: 2027119)
PI: Panganamala R Kumar
Co-PI: Srinivas Shakkottai
Co-PI: I-Hong Hou
Platform to use: POWDER
This project investigates a wireless communication framework for information centric networking based on experiments conducted over an NSF Platform for Advanced Wireless Research. The value of these experiments is to enable the realistic evaluation of the information centric networking paradigm, and its compatibility with existing wireless communication architectures. The effort enables the generation of expertise in the area of cellular communication systems and helps in training a workforce critical to 5G systems and beyond. The project also addresses the incorporation of such experiment-driven studies into courses on network optimization and control, as well as outreach efforts.
The goal of this project is the conduct of experiment-driven design and analysis of the elements of SPLICE, a Secure Predictive Low-Latency Information Centric wireless Edge. The project is organized into three inter-dependent thrusts at the wireless edge, namely (i) Wireless-optimized Information Centric Networking, (ii) Predictive caching applications, and (iii) Medium Access Control (MAC) to support low information age. A core component is the evaluation of algorithms developed using an open-source programmable cellular data network for conducting realistic over-the-air experiments. Together, these thrusts provide for novel modalities and architecture for real-time scheduling, caching and information dissemination over wireless edge networks.
ICE-T:RC: Accelerating NFV Service Function Chain Processing at Scale
Project ID: 1836772
PI Name: Zhi-Li Zhang
University of Minnesota
Platform to use: POWDER
Network Function Virtualization (NFV), coupled with Software Defined Networking (SDN), promises to revolutionize networking by allowing network operators to dynamically manage networks. Operators can deploy and scale out/in network functions (NFs) on demand, construct a sequence of NFs to form a service function chain (SFC) to meet various service requirements. In the emerging 5G technologies – besides innovations in radio technologies such as 5G new radio, NFV will be a key enabling technology underpinning the envisioned 5G packet core networks for support of network slicing and diverse services. Despite all the hype, realizing many touted advantages of NFV is a daunting challenge in practice, especially when applied to emerging 5G networks, where high scalability, availability and performance will be critical.
As part of the NSF funded ICE-T project, the project team has developed a novel distributed parallelization framework, dubbed HydraNF, for accelerating NFV service function chain processing at scale. The team will conduct experiments using the NSF PAWR POWDER and COSMOS testbeds for evaluating NFV support for emerging 5G services. The plan is to use POWDER for basic performance studies, while COSMOS will be used for protocol and software validation within the ONAP framework. Utilizing the 4G/5G reference implementation and ONAP facilities provided by these testbeds, the project team will develop, deploy and test a number of key NFV-based 5G core network functions and evaluate their performance in supporting emerging 5G enhanced broadband (eMBB) services.
NeTS: JUNO2: Resilient Disaster Communications in the Social-Media Era
Original NSF Proposal 1818971
(Supplement Proposal: 2027018)
PI: K. K. Ramakrishnan
University of California, Riverside
Platform to use: POWDER
Effective and timely communication among citizens in need of help, first responders and others who are able to help during and in the aftermath of a disaster can affect outcomes dramatically. The integration of social media in daily lives has also dramatically changed how victims, volunteers, and first responders exchange information during this time. This project is developing a communication framework in disaster situations that enables critical information to be shared in a timely manner among several groups that use multiple modes of communication, including the use of social networks for hyper-local assistance from first responders and volunteers. This project seeks to have significant societal impact in the way public safety is managed.
The experimental work performed as part of this supplemental effort will be to demonstrate the effectiveness of communication framework to provide timely delivery of relevant information to first responders using the PAWR testbed. This supplementary project proposes to complete the implementation of a prototype, test, and obtain measurements with it, on the PAWR platform. The project will also show the use of a social media engine for online low-latency NLP on edge cloud platforms and disseminating relevant information to the appropriate first responders in real-time by mapping individual free-form social media messages to a namespace used for publishing, expressing interest and creating subscriptions.