To content
Fakultät für Elektrotechnik und Informationstechnik

P. Jörke, T. Gebauer, and C. Wietfeld. 2022. From LENA to LENA-NB: Implementation and Performance Evaluation of NB-IoT and Early Data Transmission in ns-3. In Proceedings of the Workshop on ns-3 (WNS3 2022). ACM, New York, NY, USA.

  • Best Paper Awards
Best paper award certificate © CNI

With the growing Internet of Things, more small devices require internet connections. To address the scaling number of Internet of Things devices, the Cellular-IoT solution Narrowband Internet of Things has been extended with new features for more efficient transmission of small data. With Early Data Transmission, a promising new feature has been introduced in 3GPP Release 15, which aims to reduce the overall overhead of NB-IoT data transmissions by transmitting data without an active Radio Resource Control connection. To evaluate the impact of Early Data Transmission on the efficiency and scalability from a user and network provider perspective, we have implemented NB-IoT in ns-3, called LENA-NB framework. It features NB-IoT performance improvements from 3GPP Release 13 to 15, such as Cellular-IoT C-Plane Optimization, Early Data Transmission, RRC Resume procedure, as well as a new cross-subframe scheduler, adaptive modulation and coding, and a detailed energy state machine. Using LENA-NB, a performance comparison of different NB-IoT transmission modes is performed, which results in a clear recommendation to use Early Data transmission per default, since it improves latency, battery lifetime, and spectral efficiency, especially in scenarios with a large number of devices. In terms of spectrum usage, Early Data Transmission uses 3.7x less spectrum for the same transmissions compared to the default NB-IoT transmission mode, thus making the network much more scalable. Besides the impact in the spectrum usage, our work determined the currenty used Random Access windows of public NB-IoT networks as a potential bottleneck, which indicates the necessity of a Random Access optimization in the future. In addition, our implementation of LENA-NB is provided as an open-source project.