See the future in our lab
Our lab is equipped with hardware and software for experiments and testing of new functionalities and features in 5G+/6G mobile networks, edge computing, and various use-cases, icluding areas of robotics, vehicles, and autonomous systems. The implementation of mobile network in our lab is based on the open source OpenAirInterface (OAI) and srsRAN. We exploit the software for development and validation of our theoretical solutions. The open source software enables us to modify arbitrary communication parameters and implemet our own solutions to go beyond the standardized communication protocols. To validate the cutting edge solutions, we also implement our ideas from a scratch.
Demos and Experiments developed in our lab
Prediction of the channel quality via Deep Neural Network
In this demo the quality of channel between two devices is predicted based on the idea proposed in M. Najla, Z. Becvar, P. Mach, D. Gesbert, “Predicting Device-to-Device Channels from Cellular Channel Measurements: A Learning Approach“, IEEE Transactions on Wireless Communications, vol. 19, no. 11, pp. 7124 – 7138, 2020. As D2D communication is not yet fully available in OAI, we demonstrate prediction between a user equipment and a base station with knowledge of channels between the user equipment and few neighboring base stations.
Multi-access Edge Computing
Our Multi-access Edge testbed is based on the OAI with Open vSwitch local breakpoint. For the evaluation we exploit an Android application for augmented reality Percipio developed within our lab that supports computation offloading . The MEC testbed with the application enables not only to demonstrate MEC capabilities, but we can also optimize and measure communication latency and/or energy consumption of the mobile devices.
Self-optimized RAN
The self-optimization of RAN is done via developed API providing communication interface to MEC, where optimization of RAN is performed. We demonstrate the self-optimization capabilities via control of UE transmission power. The power control reduces the energy consumed by the UE and prolongs its battery lifetime. The details are in bachelor thesis of Jakub Novy “Mobile network optimization exploiting Multi-Access Edge Computing”.
Radio resource scheduler
We demonstrate efficiency of a centralized version of scheduler based on game theory, as proposed in I. Bistritz and A. Leshem. “Game theoretic dynamic channel allocation for frequency-selective interference channels.“, IEEE Transactions on Information Theory 65, no. 1 (2018): 330-353. This scheduler maximizes the data rate of each user by achieving close to optimal pure Nash equilibria.
Communication for autonomous railways
We develop communication technologies to support communication of autonomous trains in collaboration with Ixperta (Czech company).
Our equipment
Hardware:
- 2x USRPs B205 mini (SDR on flying base station)
- 6x USRPs B210 (SDR for fixed base stations and user equipment)
- 8x USRPs N310 (SDR for fixed base stations and user equipment for 5G)
- 3x Autonomous platforms (robot, vehicle) with MEC offloading capability
- Flying base station (drone – hexa-copter and HW+SW for emulated mobile connectivity)
- 2x MEC servers (each with Intel Xeon Gold 6154 @3GHz, 18 cores, 256 GB RAM)
- GPU nVidia GeForce RTX 3090Ti
- PCs and laptops for emulation of radio and core network
- Smartphones, tablets, and USB dongles
Software:
- RAN and Core Network for 4G, 5G, and beyond 5G emulation based on OAI and srsRAN.
- GNU Radio for emulation of vehicular communication networks
- NS-3 for Cloud Radio Access Network development.
