Stats

  • Locked Locked
  • Replies 1
  • Subscribers 63
  • Views 14238
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Noise figure simulation of LNA in presence of transient signals

RP086

Hello,

I have been computing noise figure of a low noise amplifier by running s-parameter simulation. In this I mimic the DC operating point of my circuit by means of supplying external voltage sources to bias the device.

In my actual design, I have a LDO and a chargepump that has a variable oscillator frequency and hence lots of frequency content in it. I would like to now simulate my actual die where in the LDO and Chargepump are biasing the LNA so that I can access the contribution of the LDO, Chargepump, other analog blocks on the NF of the LNA if any.

What is the best way to do this ? Ideally if there is a "transient assisted spara" simualtion engine, then I could run transient on the design and then use the result as the operating conditions for spara and thereby compute the small signal parameters as well as NF.

I looked around and I couldn't find such a simulator within Cadence or ADS. 

Is there any other simulator I could use ? I tried using Harmonic Balance with an initial transient, but I couldn't compute the NF using this engine.

The pass band of the LNA is between 1.5-2.5GHz and the Charge pump has a variable frequency between 5-10MHz.

Please let me know if I am missing something. Any help would be greatly appreciated.

Thank you

  • For the noise figure of the LNA you don't need a large signal input to the LNA, so you could potentially simulate using pss/pnoise or hb/hbnoise with the oscillator mode enable and just that as the only frequency. Of course, if the oscillator frequency is changing over time (and I don't mean just settling to a fixed frequency), this won't work because the circuit is no longer periodic - it depends on how it adapts in frequency.

    If you are measuring the noise figure whilst driving the input close to compression, you might be able to use "semi-autonomous" hb analysis (this is where you have an oscillator and a driven signal), but again you need to have these as steady state frequencies.

    If the signals are truly time-varying, you don't have much choice other than using a transient noise simulation (this is transient with the noise sources enabled) - which can be somewhat slow (in any simulator) because you generally need lots of timepoints (for high frequency noise), a long simulation (for low frequency noise), and higher accuracy (because the noise signals are small and can get lost in the numerical accuracy of the simulator otherwise). That's not particularly conducive to having a fast simulation.

    You'd then have to calculate the power-spectral-density (using the psd function) to compute the noise figure somehow. Not that easy though, as there's nothing direct to compute noise figure in such a situation since the output of the simulation is the signal+noise.

    Regards,

    Andrew.

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.