The Process

When starting this project, we all had a general idea of how to approach the problem. We had broken down the project into three different parts that were simple on their own.


  • The Anti-Noise Signal
  • The Mixer
  • The Metric of Intelligibility

This was a more naive approach as having to build this with a single microphone offered a more difficult process. We needed to research any methods that were already taken to handle using only a single microphone. Eventually we concluded that the Fx_LMS algorithm was the best approach to completing our objective.






The Milestones

  • Developed our three subsystems.

  • Developed a working MATLAB simulation of Fx_LMS algorithm.

  • Presented to our client and company.

  • Presented to UGRADs audience and people interested.





The Tools

Software Used

  • MATLAB

Hardware Used

  • Arduino Nano
  • Sound Board
  • Speakers
  • DC converter

This hardware was used towards the very end when we attempted to port our simulation to something physical.





The Results

Our simulation model in MATLAB.

portfolio image

This model shows the flow of signals and each output.


The simulation outputs these two graphs to show the error and the filter weights adjusting in real time.


error
Filter Weights





UGRADs

Our team at UGRADs and our poster


team

Our Poster

team