Brushing Frequency Calculation

Transcript

Brushing frequency calculation In order to determine if the sampling rate of the BX-chip was sufficient we plotted the sampling values from a brushing session with our prototype toothbrush in a graph. The graph turned out pretty well and we could start working on a frequency analysis algorithm in Java. The Java method looks for top peaks (turning points) among the sampling values and calculates the elapsed time between two peaks in order to determine the brushing tempo. Acceleration Toothbrushing frequency analysis 750 700 650 600 X Y 550 500 450 400 350 300 1 26 51 76 101 126 151 176 201 226 251 Time The output from the brushing speed calculation turned out to give exponential values. This affected the music’s speed so that only small tempo changes occurred when brushing slowly, and big differences in the music’s tempo occurred when brushing faster. Beats Per Minute 600 500 400 y = 39,268e0,1044x BPM 300 Expon. (BPM) 200 100 0 1 3 5 7 9 11 13 15 17 19 21 By calculating the natural logarithm values of the brushing speed we got a much smoother change of tempo during brushing. Log BMP 7 6 y = 0,1044x + 3,6704 5 Log BMP 4 3 Linear (Log BMP) 2 1 0 1 3 5 7 9 11 13 15 17 19 21