Week 22 (04/11/2021)

This week the team worked on correcting the spikes being collected every time the multiplexer switches. It has been understood that these spikes begin appearing at the bandpass filter, specifically the low pass filter in the bandpass filter, and from there it the amplitude of these spikes gets amplified with the rest of the collected signal. The team spoke with Dr. Jie Xu, and he recommended that the team symbolically set up the transfer function for the low pass filter, finding the breakpoints, and then gradually widen the bandwidth. Dr. Xu suggests that by changing the alpha, the smoothing constant, the team should notice the spikes disappear.

Besides working on the multiplexer, the team collected more "yes" and "no" datasets. Two "yes" and two "no" datasets were collected from each team member, totaling 8 datasets there was collected this week. Each of the 8 datasets was used as an input for the machine learning k-means clustering algorithm on MATLAB. The machine learning algorithm correctly labeled three of the recently collected datasets without a doubt, while it labeled the other five datasets correctly but also suggested for the machine learning algorithm to be adjusted to verify in the future. In summary, the machine learning algorithm is correctly able to determine whether the wearer of the EEG helmet was thinking "yes" or "no".

The main parts of the helmet has finally been printed by the Engineering Design Center at Loyola Marymount University. The team has begun assembling the components to the helmet.

THIS WEEK

Tyler

• Continued to debug the multiplexer

• Began assembling the EEG helmet

• Collected further data using the EEG helmet for further analysis

Mark

• Collected further data using the EEG helmet for further analysis

• Performed different methods of k-means clustering on the data collected for comparison against previously collected data

NEXT WEEK

Tyler

• Continue to debug the multiplexer

• Finish assembling the EEG helmet

• Collect further data using the EEG helmet for further analysis

Mark

• Collect further data using the EEG helmet for further analysis

• Assist Tyler in debugging the multiplexer

• Assist Tyler in assembling the EEG helmet