Attended: Mengyuan, Creighton, Scott

• Updates:
  • Met with Andy and Daisy
  • Implemented FIR LP third order filter with uniform weights with real time processing
    • Results did not come out as expected - signal not rectified
    • need to consider other FIR LPF designs (orders and coefficient weights)
• To Do:
  • Implement other filtering methods and compare their accuracy and responsiveness to abrupt changes
  • Implement different frequency domain analysis methods to extract any useful information between indoor conditions and outdoor ambient noise

• Updates:
  • Met with Andy and Daisy and updated them on our progress.
  • Observed with the oscilloscope that the received pulses are getting farther and farther away from each other (the time between two received pulses is increasing). To remove this variability, we're going to test with just sending a few pulses and using only the first zero-crossing for the propagation time.
  • Built an amplifier circuit to amplify the received wave. However, we used a differential amplifier configuration and it required two sets of voltage dividers (one to offset the signal, and the other one to compare the signal against so that the amplifier will amplify the difference between the two). This isn't ideal since if the DC voltages from the dividers aren't exactly equal, their difference will get amplified as well. Further research was done and we found a non-inverting amplifier configuration that we could bias to be centered at 1.65V (Vcc/2). Below is an excerpt from a MIT OCW guide on biasing amplifiers.
• To Do:
  • Build and test the new non-inverting amplifier circuit.
  • Test just sending one or a few pulses rather than ten pulses.
  • Add in calculation of wind speed into program so it can print out the calculated wind speed directly, and conduct tests with more wind speeds and the other fan.
  • See if we can reduce the ringing of the emitter by sending 180-degree phase shifted pulses after we send our actual pulses (destructive interference).
  • See if we can use the Schmitt trigger for signal qualification purposes.

Contributing authors:

Created by mwu on 2017/05/08 00:02.