To cope with the increasing scarcity of spectrum resources, a promising approach is needed to extend the new-generation of LTE/LTE-A cellular systems to unlicensed bands, which leads to so-called unlicensed LTE (ULTE). However, such extension is by no means straightforward primarily because of the lack of compatibility between the radio resource management schemes used by LTE and systems already deployed in the unlicensed band (e.g., Wi-Fi), which may lead to unfairness, low throughput, and poor scalability, among others. 

Technology Overview

Northeastern University researchers propose a novel idea of a cognitive coexistence scheme to enable spectrum sharing between U-LTE and Wi-Fi networks, referred to as CU-LTE. The proposed scheme is designed to jointly determine dynamic channel selection, carrier aggregation and fractional spectrum access for U-LTE networks, while guaranteeing fair spectrum access for Wi-Fi based on a newly designed cross-technology fairness criterion. The first derive a mathematical model of the spectrum sharing problem for the coexisting networks; then design a solution algorithm to solve the resulting fairness constrained mixed-integer nonlinear optimization problem. The algorithm, based on a combination of a branch and bound and convex relaxation techniques, maximizes the network utility with guaranteed optimality precision that can be set arbitrarily to 1 at the expense of computational complexity. Performance evaluation indicates that near-optimal spectrum access can be achieved with guaranteed fairness between U-LTE and Wi-Fi. 


  • This technology allows U-LTE to access the entire unlicensed spectrum band, and hence can utilize the spectrum more efficiently. 
  • U-LTE base stations can send ABSFs in a dynamically and optimally fashion. Moreover, the technology relies on centralized scheduling for U-LTE networks, which is more spectrally efficient than listen-before-talk. 
  • It can enable harmonious coexistence between LTE and Wi-Fi in a diverse set of network settings, with nearly optimal spectrum assignments for LTE networks and with always guaranteed spectrum access for Wi-Fi. 
  • The technology defines mathematically for the first time a cross-system fairness criterion, based on which the performance degradation caused by an unlicensed CU-LTE Pico cell to the Wi-Fi network should be no more than another Wi-Fi network.


  • The technology has the potential to be implemented in next-generation 5G cellular networks to increase the spectral efficiency of wireless networks
  • With this technology, service providers can serve more mobile users with higher quality-of-service (QoS) without extra monetary investment to obtain new spectrum resources.


  • License
  • Partnering
  • Research collaboration
Patent Information:
For Information, Contact:
Dormant Computer
Northeastern University
Tommaso Melodia
Zhangyu Guan