Internet Measurement Research: Methodologies, Tools, and Infrastructure (IMR)

With this solicitation, the National Science Foundation's (NSF) Directorate for Computer and Information Science and Engineering (CISE), in partnership with the Directorate of Mathematical and Physical Sciences (MPS), is launching a new, focused program to support methodologies, tools, and research infrastructure for Internet measurement spanning access (both wireless and fixed broadband) and core Internet. Currently, Internet measurement is conducted in a piecemeal and uncoordinated manner, and the infrastructure to collect, share, and process the data does not include data on all aspects of the network covering both wireless and wired Internet. The scope, complexity, and means of accessing the Internet have changed dramatically throughout its existence. Internet measurement work has mostly focused on the wired core networks for which existing Internet measurement repositories/infrastructure provides yeomen service. Methods, data collection, and data sharing have not kept up with the importance and proliferation of wireless and fixed broadband access networks. This leaves important aspects of the access network (both wireless and fixed broadband) in many geographic regions unmeasured or undermeasured.  With citizens now using cellular phones for accessing the Internet, more accurate and/or additional ways to measure and assess performance, connectivity, network topology, and service gap have also become necessary. The goal of the IMR program is to encourage, coordinate, and connect research in Internet measurement in a comprehensive manner. Such research is essential and timely to assess the health of the Internet more comprehensively, improve network technologies and systems, and develop new methods of networking, especially with the development of new methodologies and tools.  

The IMR program will support three award tracks: 

• Track 1: Methodologies and Methods (MM) track awards support the creation of new methods for collecting, anonymizing, modeling, and analyzing Internet measurement data. The award track will support three subtracks. The first subtrack is statistical methodologies, with awards supporting the creation of new stochastic models and statistical methodologies for Internet measurement research, such as methodologies that adjust for nonrepresentative data (e.g., data imputation), provide accurate results despite limited or sparse data, or methodologies that support new ways to analyze Internet measurement data. The second subtrack is privacy-preserving methodologies, with awards supporting innovative techniques or methodologies to ensure privacy protection during collection, sharing and analysis of Internet measurement data. The third subtrack is other methodologies, with awards supporting the creation of new Internet measurement methodologies, analyses, or post-processing not covered by the first two subtracks, such as improving the footprint of current data collection methods (e.g., to access networks), methods for measuring IPv6 address space, or integrating the measurement of access (both wireless and fixed broadband) and core networks. 

• Track 2: Measurement Tool Development and Demonstration (MT) track awards support the creation and deployment of new tools to collect Internet measurement data. These tools may be based on methodologies such as those supported by Track 1 or other prior work. These tools should ultimately be publicly available and include both active and passive measurement tools.   

• Track 3: Internet Measurement Related Infrastructure-Planning (RI-P) track awards will support the creation of infrastructure for hosting measurement tools and data. The funded infrastructure will make data available to the research community, including curating the data, ensuring an appropriate level of privacy protection, and developing necessary exchange formats, tools, and mechanisms. This infrastructure will eventually host tools and data, integrating the outcomes from Track 2. Currently in this track, planning proposals are sought towards such an infrastructure, while a separate future solicitation will likely be issued later with further details for a full infrastructure proposal submission, which is expected in FY 24 for Track 3, subject to the availability of funds. 

The methodologies developed in Track 1 will help facilitate sharing or analysis of the data and will eventually be stored in the infrastructure sought through Track 3 in a privacy-preserving manner and will lead to innovative tools for data collections developed in Track 2. Tools developed in Track 2 may lead to curated data being added to the infrastructure identified through Track 3. Thus, the infrastructure sought through Track 3 will have multiple purposes. The integration of outcomes of the three tasks should help synthesize a more holistic or comprehensive approach to Internet measurement across different components of the Internet including core, access, wired, and wireless networks, considering security and privacy implications.