Bandpass Filters - Small Order FIR and IIR Systems
Information sources：本站 | Release date：
2019-02-15
| Browse volume：119
Bandpass filter and filter data are designed using a small-order FIR averaging filter and an IIR Butterworth filter; many real signals can be modeled as an oscillating component, a combination of low frequency trends and additive noise. For example, economic data typically includes oscillations that represent cycles that are superimposed on a slowly changing upward or downward trend. In addition, there is an additive noise component, which is a combination of measurement error and random fluctuations inherent in the process.
Suppose that a process is sampled once a year for a process, assuming that the process oscillates on a scale of about one week and one month. In addition, there is a low-frequency rise in data and additive white Gaussian noise.
The signal is created as a superposition of two sine waves with a frequency of 1/7 and 1/30 weeks/day. Add low frequency increase trend term and white Gaussian noise. Reset the random number generator to get reproducible results. The data was sampled at 1 sample/day. The resulting signal and power spectral density (PSD) estimates are plotted.
