Because of the thermal sensitivity effect of resistance, any equipment has noise, and spectrum analyzer is no exception. The noise of spectrum analyzer is essentially thermal noise and belongs to Random. It can be amplified and attenuated. Because it is a random signal, the combination of the two noises can only add up and can not produce subtraction effect. The frequency bandwidth of the detector is much larger than that of the internal circuit of the device. The noise detected by the detector is related to the set resolution bandwidth (RBW). Because noise is random superimposed on signal power, the relationship between the noise quasi-position displayed and the resolution bandwidth is logarithmic. When the resolution bandwidth is changed, the noise changes accordingly. The mathematical formulas related to the noise change are as follows:
For example, if the bandwidth is adjusted from 100 kHz (BW1) to 10 kHz (BW2), the noise change is as follows:
That is to say, the noise is reduced by 10 dB (1/10 of the original), and the signal-to-noise ratio is increased by 10 dB. It can be seen that the narrowest bandwidth can be used to reduce the noise. Whether noise comes from outside or inside, measurement will affect the accuracy of signal amplitude, especially in low-level signal. When noise is too large, it even conceals the signal so that the size of the signal can not be correctly judged. The two kinds of noise that affect the measurement quality can be summarized as the following three main items:
(1) Pulse noise generated by digital circuits, ignition systems and DC motors with switching functions, which is common in the field of EMI (Electromagnetic Interference).
(2) Random noise comes from the electronic movement of nature or circuit, also known as KTBW (or thermal) noise, Johnson noise, broadband noise or White noise, etc. This book mainly focuses on thermal noise. The mathematical formula is: Pn = kTBW, among which: Pn = noise power = 3.98*10?21W/Hz or -174dB/Hz k = B. Oltzman constant, 1.38 * 10? 23 Joule / oKT = absolute temperature at room temperature = 290 oKBW = system noise power bandwidth (Hz).