Structural principles and frequency band detection technology of spectrum analyzer

A spectrum analyzer is a special instrument used to analyze the frequency components often contained in signals. With the rapid development of wireless communication and automation technology, the technical performance and testing functions of spectrum analyzers are increasingly improving. At present, some novel and high-end spectrum analyzers have a large frequency measurement range, strong accuracy, sensitivity and stability, and can be used to measure many important parameters of the signal. Such as output power measurement, frequency measurement, deployment measurement, frame loss measurement, noise measurement, EMC/EMI measurement, etc.

Tektronix spectrum analyzers can be divided into two categories according to their structural principles, namely analog spectrum analyzers and digital spectrum analyzers. The early spectrum analyzers were of the analog type, and currently analog spectrum analyzers are still widely used. Digital spectrum analyzers are basically composed of high-pass filters or FFTs. Since digital spectrum analyzers are limited by the operating efficiency of digital systems, most of these spectrum analyzers are used in the low frequency range. In addition, some modern high-end spectrum analyzers can be used to measure both low-frequency signals and high-frequency signals. Their structure is a mixture of the above two categories, often called "simulation". Analog ¾ Data” Hybrid Spectrum Analyzer.

According to the implementation method of spectrum analyzer and the establishment technology of frequency band detection, spectrum analyzer can generally be divided into: bandpass filter detector, fast Fourier transform (FFT) detector, frequency offset spectrum analyzer and real-time spectrum analyzer. .

1. Bandpass filter detector

The initial way to implement a spectrum analyzer was to introduce the signal to be measured into a series of bandpass filters with the same bandwidth, but with the core frequency arithmetic increasing with the bandwidth of the stepper motor, and then detect the signals through each frequency detector. , obtain the output power of each frequency point, and finally display it on the monitor. This type of spectrum analyzer is called a bandpass filter spectrum analyzer.

Figure 1 shows a block diagram of the working principle of the bandpass filter detector. A detector that works according to this diagram is called a bandpass filter spectrum analyzer. Band Pass Filter The minimum frequency discriminating network bandwidth of the spectrum analyzer is determined by the network bandwidth of the band pass filter. Assuming that the network bandwidth of the bandpass filter is 100kHz, the frequency accuracy of the bandpass filter Tektronix spectrum analyzer is only 100kHz. The main reason is that if the power spectral density lines of several frequencies occur within the 100kHz frequency range of the same bandpass filter, then the test results of the spectrum analyzer of the bandpass filter within this 100kHz range will only indicate one power spectral density line. Line, the bandpass filter will measure the kinetic energy in its frequency range, regardless of how many frequency band components contribute to this total energy. Therefore, for closely adjacent frequency band components, the minimum frequency discrimination network bandwidth is limited by the wideband network of band-pass filters.

Figure 1 Block diagram of the working principle of the bandpass filter detector

The biggest advantage of a bandpass filter spectrum analyzer is that it can quickly track changes in signal frequency bands over time, but its biggest disadvantage is that in order to ensure the minimum frequency screen resolution network bandwidth, narrowband filters must be applied, and the total number of narrowband filters is required This improvement is accompanied by the expansion of the measurement frequency range of the bandpass filter spectrum analyzer and the reduction of the minimum frequency screen resolution. For this reason, bandpass spectrum analyzers are mainly used in situations where the bandwidth of the analyzed network can be very wide.

2. FFT detector

As we all know, the Fast Fourier Transform (FFT) can be used to determine the time domain representation (frequency band) of frequency domain signals. The signal must be mapped in the frequency domain and then the FFT algorithm is run to calculate the frequency band. Figure 1-2 shows a simplified basic principle program block diagram of an FFT spectrum analyzer.

As can be seen from Figure 2, the principle of the FFT spectrum analyzer is: the RF input signal passes through a variable optical attenuator to ensure different measurement ranges; then, the signal passes through a bandpass filter to filter out the frequency range of the spectrum analyzer. Unwanted high-frequency components; use the sampler to sample the signal waveform, then change the joint function of the sampling circuit and the digital-to-analog converter into a data mode, use FFT to measure the frequency band of the waveform, and put the results on the display Information is displayed on the screen and the signal frequency band is measured.

Figure 2 Simplified basic principle program block diagram of FFT spectrum analyzer

An FFT spectrum analyzer can perform the same function as a multi-channel filter detector, but without using many bandpass filters. The difference is that the FFT detector uses signal processing to achieve the special functions of several filters. The basic theoretical basis of FFT spectrum analyzer is the uniform sampling theorem and Fourier transform.

Uniform sampling theorem: A signal in a relatively limited frequency band that does not have a component exceeding frequency fmax in the frequency band is accurately determined by sampling values ​​of the signal carried out at intervals not exceeding 1/2fmax. When this sampled signal is passed through an ideal bandpass filter with a cut-off frequency f, the original signal can be completely reconstructed. When the specific sampling frequency fs of the analog/digital converter of the FFT spectrum analyzer should reach:

The relationship between the cut-off frequency f, the sampling frequency fs and its fmax is as follows:

Fourier transform: According to Fourier transform, the signal can be expressed in detail using the frequency domain function f(t), or it can be expressed in detail using the time domain function formula F(jw), and there is a close relationship between the two. , only one of them needs to be clear, and the other one will be clear as well. Therefore, the transformation from frequency domain to time domain can be completed.

3. Frequency offset spectrum analyzer

Common spectrum analyzers currently use frequency offset amplification circuit solutions. Similar to wireless receivers, spectrum analyzers can automatically perform frequency offset in the entire frequency band of interest to indicate signal strength and frequency content. Frequency offset spectrum analyzers have been gradually replaced by FFT detectors in low-frequency stock bands, but within the frequency range of radio frequency, microwave radio frequency and millimeter wave communication, frequency offset spectrum analyzers have the upper hand.

There are generally two types of frequency offset spectrum analyzers. One is the automatic tuning filter spectrum analyzer. This type of spectrum analyzer works by moving the core frequency and bandwidth of a bandpass filter within the entire frequency range. middle. The core frequency automatically and continuously scans within the signal frequency band, thereby successively selecting the frequency band components of the signal to be measured, and adding them to the vertical offset circuit of the display after detection and amplification, while the input of the horizontal offset circuit The signal originates from the same scanner signal generator that automatically tunes the filter core frequency of the scanner signal. The horizontal axis indicates the frequency. The advantages of this type of spectrum analyzer are simple structure, low price, and no false alarm signals; the disadvantage is that the spectrum analyzer has low accuracy and poor screen resolution. The second is the frequency offset superheterodyne spectrum analyzer. The principle of this type of spectrum analyzer is widely used by contemporary detectors. For example, the Agilent8560EC series spectrum analyzers of foreign Agilent companies, the Agilent ESA-E series spectrum analyzers, the Agilent8590 series spectrum analyzers, and the MS2681A/MS2683A/MS2687B/MS2668C of Japan's Anritsu Corporation all use frequency offset superheterodyne. spectrum analyzer. The frequency offset superheterodyne spectrum analyzer uses a fixed, agile high-frequency amplifier as a frequency selection filter, and uses the local oscillator frequency offset component to derive a series of local oscillator signals from low to high frequencies, and then import Each frequency component in the signal to be measured is mixed one by one, turning it into the corresponding high-frequency frequency component, which is enlarged, detected and filtered, and finally the measurement conclusion is displayed on the CRT (display screen). Superheterodyne spectrum analyzers are widely used in contemporary wireless communication measurements. Therefore, the principles of superheterodyne spectrum analyzers will be explained in detail in the following chapters.

4. Real-time spectrum analyzer

With the continuous advancement of RF technology, current radio signals are burdened with complex modulation technologies. Compared with past radio signals, they are more intermittent and have stronger bursts. They change or skip frames at different points, quickly reach their peak value, and then fade away, making them unpredictable. The conclusion puts the method of measuring and studying this signal under an unprecedented test. The traditional superheterodyne spectrum analyzer cannot complete the ability to analyze instantaneous signals at different times in the time domain, frequency domain or deployment domain. How to correctly turn on, capture, accurately analyze and detect the current complex and time-varying RF signals? is becoming more and more important. The emergence of real-time spectrum analyzers has brought people a powerful special tool in the wireless communication detection industry.