Technical article: Brief analysis of the principle of spectrum analyzer

The architecture of the spectrum analyzer is like an oscilloscope for time domain use. The appearance is shown in Figure 1.2. There are many function control buttons on the panel to adjust and control the system functions. The main function of the system is to display the spectral characteristics of the input signal in the frequency domain. . According to different signal processing methods, there are generally two types of spectrum analyzers: Real-Time Spectrum Analyzer (Real-Time Spectrum Analyzer) and Sweep-Tuned Spectrum Analyzer (Sweep-Tuned Spectrum Analyzer). The function of the real-time frequency analyzer is to display the signal amplitude in the frequency domain at the same instant. Its working principle is to have corresponding filters and detectors for different frequency signals, and then through a synchronized multi-task scanner The signal is transmitted to the CRT screen. The advantage is that it can display the instantaneous response of Periodic Random Waves. The disadvantage is that it is expensive and the performance is limited by the bandwidth, the number of filters, and the largest multitasking exchange. Switching Time.

The most commonly used spectrum analyzer is a scanning tuned spectrum analyzer. Its basic structure is similar to a superheterodyne receiver. The working principle is that the input signal is directly applied to the mixer through the attenuator. The variable local oscillator is connected to the CRT. The synchronous scan generator generates an oscillation frequency that changes linearly with time. The intermediate frequency signal (IF) mixed and down-converted by the mixer and the input signal is then amplified, filtered, and detected to the vertical direction plate of the CRT. The vertical axis of the CRT shows the correspondence between signal amplitude and frequency. The signal flow architecture is shown in Figure 1.3.

The important part that affects the signal response is the filter bandwidth. The characteristic of the filter is a Gaussian-Shaped Filter. The function of the filter is the resolution bandwidth (RBW, Resolution Bandwidth) commonly seen during measurement. RBW represents the minimum bandwidth difference between two signals of different frequencies that can be clearly distinguished. If the bandwidth of two signals of different frequencies is lower than the RBW of the spectrum analyzer, the two signals will overlap, making it difficult to distinguish. The low RBW certainly helps the resolution and measurement of signals of different frequencies. The low RBW will filter out the signal components of higher frequencies, resulting in distortion when the signal is displayed. The distortion value is closely related to the set RBW. The higher RBW Although it is helpful for the detection of broadband signals, it will increase the noise floor value and reduce the measurement sensitivity, which is easy to hinder the detection of low-intensity signals. Therefore, an appropriate RBW width is important for the correct use of a spectrum analyzer. concept.

Figure 1.2: Appearance of the spectrum analyzer

The other video bandwidth (VBW, Video Bandwidth) represents the minimum bandwidth required for a single signal to be displayed on the screen. As explained before, when measuring the signal, the video bandwidth is too high or not suitable, which will cause troubles in the measurement. How to adjust it must be understood. Generally, the RBW bandwidth is greater than or equal to VBW. Adjust the RBW without significant change in signal amplitude. At this time, the RBW bandwidth can be used. When measuring the RF video carrier, the signal is amplified, filtered (determined by RBW) and detected after the frequency is reduced by the mixer inside the device. If the scan is too fast, the RBW filter will not be fully charged to the peak amplitude of the signal Therefore, it is necessary to maintain sufficient scan time, and the width of RBW is interactive with the scan time. The larger the RBW, the faster the scan time, and vice versa, the choice of the appropriate width of RBW therefore shows its importance. The wider RBW can better reflect the waveform and amplitude of the input signal, but the lower RBW will be able to distinguish signals of different frequencies. For example, it is used for the measurement of 6MHz bandwidth video channels. Experience has shown that when the RBW is 300kHz and 3MHz, the peak amplitude of the carrier does not change significantly. When measuring 6MHz video signals, the 300kHz RBW is usually used to reduce noise. When measuring antenna signals, the Span of the spectrum analyzer uses 100MHz to obtain a wider signal spectrum requirement, and the RBW uses 3MHz. These measurement parameters are not static, and will be adjusted according to the site conditions and past measurement experience.

1. Analyze the information processing process of the spectrum analyzer

When measuring high-frequency signals, the intermediate frequency after the mixing of the heterodyne spectrum analyzer can be amplified to obtain higher sensitivity, and the frequency bandwidth of the intermediate frequency filter can be changed, and the frequency resolution can be easily changed. However, because the superheterodyne type spectrum analyzer scans within the frequency band, unless the scanning time is approached to zero, the real-time (Real Time) response of the input signal cannot be obtained. The super-heterodyne spectrum analyzer with the same performance has a very fast scanning speed. If the scanning time is smaller than the time constant of the IF filter, the correct amplitude of the signal cannot be obtained. To improve the frequency resolution of the spectrum analyzer, and to get an accurate response, there must be an appropriate scanning speed. From the above description, it can be seen that the superheterodyne spectrum analyzer cannot analyze the spectrum of the transient signal (TransientSignal) or pulse signal (Impulse Signal), and its main application is to test periodic signals and other spurious signals (Random Signal) spectrum. The characteristics of the spectrum analyzer system and the display on the panel are detailed in the description of Appendix I. Understanding this content will help the operation of the spectrum analyzer. Generally, the frequency of the output signal of the local oscillator is higher than the frequency of the intermediate frequency signal. The frequency of the output signal of the local oscillator can be adjusted to the frequency of the harmonics, that is, ƒIN = n⋅ƒLO ± ƒI F n = 1, 2, 3. ......(2)

It is known from formula (2) that the signal measurement range of the spectrum analyzer has been broadened invisibly, and input signals lower or higher than the local oscillator or other harmonic frequencies can be mixed to generate intermediate frequencies. The mixing principle of extending the frequency of the input signal is shown in Figure 1.4, where the vertical axis represents the input signal (Æ’IN), and the horizontal axis represents the local oscillation frequency (Æ’LO), and the positive and negative integers in the figure represent the formula (2) the corresponding positive negative.

Figure 1.3: Signal flow of the spectrum analyzer

From Figure 1.4, we can understand the working principle of the spectrum analyzer using the local oscillation harmonic signal to extend the frequency of the input signal. However, Figure 1.4 may correspond to multiple input signal frequencies. In order to eliminate this phenomenon, a frequency preselector (Preselector) is added in front of the attenuator to improve the dynamic range of the spectrum analyzer, and at the same time, the output results can remove other unnecessary Frequency really reflects the frequency of the input signal.

Figure 1.4: The principle of expanding the signal frequency using the harmonic signal of local oscillation

It is known from the above that the superheterodyne or spectrum analyzer cannot analyze the spectrum of transient signals or impulse signals, and its main application is to test the spectrum of periodic signals and other random signals.

2. Noise characteristics

Due to the thermal effect of the resistor, any device has noise, and the spectrum analyzer is no exception. The noise of the spectrum analyzer is essentially thermal noise, which is random. It can be amplified and attenuated. Because of the randomness The signal and the combination of the two noises can only be added and cannot produce a subtraction effect. It is also quite flat in the frequency band, and its bandwidth is much larger than the bandwidth of the internal circuit of the device. The noise value detected by the detector is related to the set resolution bandwidth (RBW). Since the noise is randomly added to the signal power, the displayed noise level and the resolution bandwidth are in a logarithmic relationship. When the resolution bandwidth is changed, the noise changes accordingly. The mathematical formula related to the amount of noise change is shown below :

For example: if the bandwidth is adjusted from 100kHz (BW1) to 10kHz (BW2), then the amount of noise change is:

That is, reduce the noise amount by 10dB (1/10 of the original), and relatively increase the signal-to-noise ratio by 10dB. It can be seen that to reduce the amount of noise purely, using the narrowest bandwidth will achieve the goal. Regardless of whether the noise is generated externally or internally, the accuracy of the signal amplitude will be affected during the measurement, especially in the case of low-level signals. When the noise is too large, the signal is even masked so that the size of the signal cannot be judged correctly The two types of noise that affect the measurement quality can be summarized as the following three major items:

(1). Impulse noise generated by the digital circuit of the exchange function, ignition system and DC motor, such noise is common in the discussion field of EMI (Electromagnetic Interference).

(2). Random noise comes from the electronic movement of nature or circuits, also known as KTBW (or thermal) noise, Johnson noise, broadband noise or White noise (White) noise, etc. Focus on noise, the mathematical formula is:

Pn = kTBW, (5)

Where: Pn = noise power = 3.98 * 10−21 W / Hz or -174dB / Hz

k = Boltzman constant, 1.38 * 10−23 joule / oK

T = Normal temperature indicated by absolute temperature = 290 oK

BW = system noise power bandwidth (Hz).

The noise power at 4MHz, 75 Ω, 290 oK is -59.1dBm. Knowing from the noise power, when the signal bandwidth is reduced, the system noise power is reduced accordingly, and the signal quality is expressed by the signal-to-noise ratio

(SNR; Signal-to-Noise Ratio), the subtraction value of signal strength (unit: dBm) and system noise power (unit: dBm) is the signal-to-noise ratio, the mathematical formula is:

3. Matching factors

The input impedance of the measurement equipment sometimes cannot match the characteristic impedance of the connection line of the device under test. According to the electromagnetic theory, when the impedance is matched, the output power is the largest and there are no other adverse side effects. The impedance mismatch will cause signal reflection and affect the stability of the system frequency And cause loss of signal power. The signal will generate standing wave and noise during transmission on and off the transmission line, which will affect the signal quality of the receiving end and the accuracy of the measured value. The shortcomings of the mismatch between the input impedance of the measuring equipment and the impedance of the DUT can be summarized as:

A. Signal reflection, the transmission cable generates standing waves online.

B. The noise increases.

C. Reduce the signal output power.

D. affect the stability of the system frequency.

E. Affect the accuracy of the measured value.