Modeling Oscillators for Simulation
Thursday | 9:00 - 9:20 am
- Kishan Shenoi - CTO, Qulsar
With the advent of 5G, not only is the need for more base-stations/access-points increasing, the density and location of deployments is getting more complicated. 5G is rapidly becoming the mechanism for supporting not just traditional Telecom applications but also such diverse applications such as Industrial Internet of Things (IIoT), sensor networks, autonomous vehicles, and many more. The synchronization requirements are getting stricter and the number of indoor installations, where GNSS signals are weak or unavailable, is increasing. The solution is likely to be based on hybrid clocks using packet-based methods with physical-layer assist.
In this presentation we discuss methods for simulating clocks in order to evaluate the quality of synchronization that is traditionally quantified by metrics such as TDEV, MTIE, and max|TE|. Specifically, we cover models for oscillators that can be used in simulation studies either in the time-domain or in the frequency-domain as appropriate. The models developed include those suitable for simulating the random noise inherent in the oscillator as well as those for the modeling of the effect of temperature suitable for incorporation in simulation studies. The study focuses on TCXO oscillators.
Temperature effects are best addressed in a time-domain simulation environment where the ambient temperature is modified in accordance with an agreed-upon pattern as a function of time. The random noise effects can be addressed in frequency-domain simulations as well as time-domain methods.