Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
wind_sensor:wind_sensor_documentation [2015/12/03 23:11] daisygreen [Implementing Code] |
wind_sensor:wind_sensor_documentation [2021/09/19 21:59] (current) |
||
|---|---|---|---|
| Line 15: | Line 15: | ||
| - squaring a function in time {x(t) -> x(t)<sup>2</sup>} is convolution in frequency domain {X(f) -> X(f)∗X(f)}. Also, sound is a real base band signal, so there for it is a low passed signal which is symmetrical around 0. (Squaring is similar to taking the absolute value. Either method could work) | - squaring a function in time {x(t) -> x(t)<sup>2</sup>} is convolution in frequency domain {X(f) -> X(f)∗X(f)}. Also, sound is a real base band signal, so there for it is a low passed signal which is symmetrical around 0. (Squaring is similar to taking the absolute value. Either method could work) | ||
| - A convolution between the same signal will be symmetrical and have it's peak in the middle. In this case it is at 0, and X(0) is the DC part which is the total energy in the signal. | - A convolution between the same signal will be symmetrical and have it's peak in the middle. In this case it is at 0, and X(0) is the DC part which is the total energy in the signal. | ||
| - | - You can find the frequency scaling depending on your sampling frequency. If sampling frequency is ω<sub>s</sub>, the frequency vector for the fft would go from -ω<sub>s</sub>/2 to ω<sub>s</sub>. | + | - You can find the frequency scaling depending on your sampling frequency. If sampling frequency is ω<sub>s</sub>, the frequency vector for the fft would go from -ω<sub>s</sub>/2 to ω<sub>s</sub>/2. |
| {{:wind_sensor:untitled2.png?nolink&400|}} | {{:wind_sensor:untitled2.png?nolink&400|}} | ||