NeTS: Small: LayBack: Layered SDN-Based Backhaul Architecture and Optimization Framework for Small Cells and Beyond

PIs: Martin Reisslein and Anna Scaglione

Supported by National Science Foundation (NSF), Directorate for Computer & Information Science & Engineering (CISE), Division of Computer and Network Systems (CNS)Award Number 1716121, 10/01/2017 – 11/30/2021.


Wireless networks typically connect users with a technology-specific chain of components to the Internet at large. These isolated networking chains limit the resource sharing and optimization across different wireless technologies. The resulting bottlenecks in wireless Internet access stifle a myriad of wireless network applications in our smart-phone based society. For high-speed wireless Internet access it is increasingly important to efficiently utilize and share wireless networking resources across different technologies, e.g., between the cellular wireless and WiFi technologies. This project develops LayBack, a novel Software Defined Networking (SDN)-based layered backhaul access network architecture and novel multi-stage optimization framework. LayBack will judiciously share the wireless resources of multiple technologies, thus fundamentally speeding up wireless Internet access. LayBack can be overlaid in an evolutionary manner on top of existing wireless networks and can thus effectively work with the installed network infrastructure. The project will support the engineering talent pipeline through innovative outreach activities to middle school students in grades six through eight, who will learn network engineering principles through English Language Arts and Physical Education activities.

This project seeks to significantly improve the flexibility and efficiency of radio access networks through the LayBack architecture and optimization framework. The intellectual merit contributions of this project include the development of the LayBack wireless access network architecture, which unifies heterogeneous radio access networks and their gateways through a Software Defined Networking (SDN)-based backhaul. The project also develops the LayBack multi-stage optimization framework, which is tightly integrated with the LayBack architecture. The LayBack optimization framework combines a Lyapunov drift-plus-penalty approach with multi-layer decomposition. The proposed optimization framework will be developed and evaluated to account for the different operational time-scales of the LayBack architecture layers. The project will examine specific case studies that exploit the LayBack architecture and optimization framework, e.g., signaling, coexistence of multiple heterogeneous wireless technologies (LTE and WiFi), and optimized content distribution. The broader impacts of this project include the public distribution of a LayBack software package. The project will also include outreach to middle schools (grades 6-8). The outreach activities will be thoroughly evaluated, refined, and distributed.


Code Repository