| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | September 4, 2018 |
| Latest Amendment Date: | July 26, 2021 |
| Award Number: | 1815676 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Ann Von Lehmen
CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2018 |
| End Date: | September 30, 2022 (Estimated) |
| Total Intended Award Amount: | $487,739.00 |
| Total Awarded Amount to Date: | $537,739.00 |
| Funds Obligated to Date: |
FY 2021 = $50,000.00 |
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
150 MUNSON ST NEW HAVEN CT US 06511-3572 (203)785-4689 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
17 Hillhouse Ave New Haven CT US 06511-8965 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): |
Information Technology Researc, Networking Technology and Syst |
| Primary Program Source: |
01002122DB NSF RESEARCH & RELATED ACTIVIT |
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
Part 1: The Internet of Things (IoT) is expected to have a multifaceted impact on economy, industry, society, and everyday life. Scaling, performance and complexity of management are among the key challenges in IoT and will largely determine its success. This project aims to leverage Software Defined Networking (SDN) and Network Function Virtualization (NFV) in IoT to address those challenges. SDN and NFV facilitate innovation in network control, lowering the barrier to entry for new and advanced functionality in the network. However, so far, the research on these technologies has been primarily focused on centralized systems such as cloud-based telco and enterprise networks. There are major challenges in applying them to IoT networks because of the large number of heterogeneous and resource-constrained devices and services in IoT (e.g. for health, home, environmental, and transportation monitoring). This project seeks to address that challenge by developing new approaches for the design of IoT network control, specifically in the SDN control plane and the deployment and management of NFV functionality in data plane, in order to support the large and expanding number of IoT applications that are becoming more critical in everyday life.
Part 2: In this project, a hierarchical decentralization of the SDN control plane is proposed along with an optimization framework for the dynamic balancing of data transmissions and virtual function processing allocations. First, this project will explore a decentralized SDN control plane optimized specifically to meet IoT needs. Second, it will provide a systematic methodology for optimizing the deployment and management of NFV functionality in the data plane based on a solid conceptual foundation. Cutting-edge theoretical methodologies will be pursued for addressing the above issues, in parallel with network prototyping and experimentation, leveraging ongoing network experimentation efforts at Yale. This will verify the efficacy of the proposed solutions, and will facilitate further development the SDN/NFV-enabled IoT architectures under consideration.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
In this project, we designed scalable and high-performance Software Defined Network (SDN) control plane architectures for the Internet of Things, developed optimal Network Function Virtualization (NFV) service placement and chaining policies, proposed and analyzed hybrid centralized and distributed algorithms for balancing of transmission and processing of IoT traffic. Finally, we prototyped implementations and tested the proposed SDN/NFV solutions. In more detail, in the first year of the project, we studied the optimal deployment of Software Defined Network (SDN) functionality in a network and proposed novel algorithms for optimizing the deployment decisions. Based on the novel deployment algorithm, we studied the placement of data-intensive services at the edge of the wireless network and developed algorithms with approximation guarantees to reduce energy and operating costs. We proposed such an approximation algorithm and rigorously proved its worst-case performance compared to the optimal solution. Then we generalized the algorithm to applications such as the IoT networks and the VR/AR services. Next, we considered the online version of the service placement problem where the performance of a service function is revealed only after its execution while the service requests and network costs might also vary with time.
We also explored the application of novel programmable network architectures to facilitate the implementation of the services. We leveraged the P4 language, which makes the data plane of the SDN architecture programmable, in order to enable a packet classification service using deep learning method. Besides, we explored the benefits of deep learning methods for improving network control using graph-based Deep Neural Networks (DNNs).
We proposed a new softwarized architecture for slicing resources of the radio access network (RAN) of possibly different technologies and across multiple providers. We designed a double auction mechanism for negotiating resource allocations in a way that guarantees convergence to optimal social welfare in finite time. We demonstrated the feasibility of our proposed system by using open source softwarized-RAN systems such as EmPOWER (WiFi) and FlexRAN (LTE). We also studied the pricing of wireless services which is very crucial for their economic sustainability. We proposed the joint optimization of wireless infrastructure deployment and pricing policies with a particular use case of a cellular network overlaid with WiFi access points (APs).
Last Modified: 01/29/2023
Modified by: Leandros Tassiulas
Please report errors in award information by writing to: awardsearch@nsf.gov.
