NeTS: Small: Collaborative Research: Identifying Design and Operational Loopholes in Cellular Network Protocols

Ohio State University, The, Columbus OH

Investigators

Abstract

The cellular network is the largest wireless infrastructure deployed today. It plays an important role to provide users mobile Internet access and cellular voice service. A key design component of the cellular network is its signaling protocols, which operate on the control plane. Compared with their Internet counterparts, these signaling protocols are more complex to ensure vital control functions, such as mobility support, radio resource control, session management for data and voice, to name a few. Consequently, their proper design and operation are critical to both the cellular operators and the users. The objective of this project is to devise tools and verify the correctness of signaling protocols in current cellular networks by: (i)identifying design loopholes in the protocols defined by the 3GPP standards, as well as operational slips made by operators and vendors, (ii) analyzing their root causes and performance penalties, and (iii) devising techniques that fix such problems. The proposed activities can enhance the reliability of cellular networks in both design and practice. They will help us to not only better understand how to avoid design and implementation mistakes in a cellular system, but also to renovate the current protocol design and even affect the upcoming 5G cellular technology in its standardization. If successful, the results will facilitate the design and practice of a more reliable mobile Internet infrastructure for our society. The project will also recruit and train a new generation of students and engineers, including those from minority groups, who are technically ready for the mobile Internet era. The PIs will interact closely with the related industry for possible technology transfer. The research in this project has two main thrusts. One is to devise tools and identify design loopholes and operational slips in cellular networks. The project will develop a cellular-specific verifier that adapts generic model-checking techniques with domain-specific cellular knowledge, and uses phone-based empirical validation. The project will apply model checking to identify a superset of potential candidate loopholes, and then validate the real ones through experiments over operational 3G/4G carriers, by focusing on design slips in two categories: (a) protocols interaction with shared certain context (e.g., connection status, commands, IP address, virtual connection), which is popular in cross-layer, cross-domain, and inter-system protocol communications; (b) Out-of-sequenced signaling delivery. When the designated sequence of control messages is disrupted, its cost is way beyond lost or delayed messages. Missequenced signals are reacted based on defined semantics and may trigger wrong protocol operations. This project will also assess the real-world impact of both categories of issues through user studies over operational carriers. The other thrust of this project is to propose fixes to the identified loopholes. The proposed solutions incorporate new heuristics into cellular signaling protocols and call for concerted effort on the mobile device and the network infrastructure.