Real-time Emulation of mmWave network

5G mmWave networks provide high capacity and low delay connectivity thanks to the bandwidth and high spatial reuse factor. Furthermore, the higher mmWave frequencies, e.g., 70-90 GHz enable compact beamforming antennas that make it possible to construct rugged vehicular access points and gateways. Versatile network deployments with static backhaul nodes and V2X nodes are supported with wireless mesh network (WMN) topology that makes it easy to extend and adapt network capacity without need for cabling. Low delay and high-reliability can be ensured by WMN specific routing and link scheduling algorithms that enable Time-Sensitive Networking (TSN) over multi-hop wireless connections.

As the hardware for such networks are still under development, VTT has developed a real-time emulation environment for mmWave WMNs. In this system, the wireless links are emulated with wired Ethernet connections and a special radio link software. The WMN nodes are emulated with 1U network processor units (NPU) that can run the same protocol implementation as with real radio links. The connectivity is modelled by simulation results or by measurement data in real-life environments. The simulations can utilize open source GIS data for buildings, roads and other objects. The advantages of the real-time emulation are the possibility to use real applications for testing, support for up to 10 Gbit/s link speeds, and testing protocols and network control at true line speeds.

To demonstrate the TSN capabilities of mmWave WMN and the emulation environment, we have added gPTP time synchronization support by utilizing the internal WMN link synchronization mechanisms. The accuracy and precision of our gPTP implementation was tested against VTT Mikes atomic clock source that was transported to our premises with WhiteRabbit. It was shown that better than ±100 ns accuracy and precision can be achieved over wireless multi-hop backhaul connections, which shows the potential of mmWave WMN in supporting industrial TSN applications.


Kari Seppänen, [email protected]


Related Posts