Application of RF streaming in signal monitoring

With the development of digital signal processing technology, the improvement of data bus bandwidth and storage medium reading speed, there are more and more technical means to record a section of spectrum information in real time, without distortion and continuously, which is usually called RF streaming. Monitoring personnel can use spectrum analysis, digital signal processing, statistical analysis, demodulation, decoding and other methods to post-analyze the recorded signal, or simulate the on-site electromagnetic environment to re-transmit the recorded signal with a signal source.

This article introduces the concept and system structure of RF streaming in detail, discusses some applications of RF streaming in the monitoring field, analyzes the factors that need to be considered when selecting a streaming system, and finally introduces the mature RF streaming and analysis solutions provided by Keysight Technologies and X-COM.

1. What is RF streaming?

Streaming is the real-time and continuous transmission of data collected or processed by instruments and equipment to storage devices for recording. The data collection front end is a test instrument such as a sensor, data collector or spectrum analyzer. The collected and recorded data is real-time sampling points or processed IQ data. The difference between the two types of recorded data is shown in Table 1:

Real-time sampling point	IQ Data
The sampled data contains carrier and information	The carrier information is removed, only the in-band information is included
Baseband sampling, the sampling rate is greater than twice the highest frequency component of the signal	Bandpass sampling, the sampling rate satisfies the bandpass sampling law

Table 1. Differences between real-time sampling points and IQ data

As can be seen from the table above, real-time sampling points are suitable for recording baseband and low-speed signals, and IQ sampling points are suitable for recording RF signals modulated by carriers. Removing the IQ data of the carrier is beneficial for both compressing the data volume and reducing the storage samples, and for later signal analysis and playback. RF streaming and playback systems usually record IQ data, and the system structure is shown in Figure 1:

Figure 1. RF streaming system structure

The RF conversion of the signal recording path includes pre-selection filtering, down-conversion, intermediate frequency filtering, signal conditioning, data acquisition and IQ demodulation. It is recommended to use a general signal analyzer, which can not only ensure the quality of the collected and stored signals, but also facilitate system maintenance. The IQ data output by the signal analyzer is usually the sampling result of a 12-bit or 14-bit ADC, which cannot be directly written to a disk array (RAID). It is necessary to use a control circuit composed of FPGA to achieve signal conversion, synchronization and routing, and then write it to the disk array through the RAID controller interface. The recorded spectrum data can be re-transmitted by the signal source through the reverse process, or copied to a workstation and played back and analyzed using dedicated software.

2. Discussion on the application of RF streaming in the monitoring field

RF streaming has expanded the methods of signal monitoring and analysis. Traditional sweep frequency and step FFT monitoring receivers can only monitor and record a small amount of processed results in real time, and there are very few data and methods that can be analyzed later. RF streaming can monitor and save all digitized information of intermediate frequency circuits, which can not only reproduce the monitoring results at the time, but also support demodulation, decoding, search, correlation, statistics and other methods to analyze long-term spectrum information in different dimensions. The applications of RF streaming in the field of signal monitoring include:

2.1 Signal Recording and Evidence

For signals measured by monitoring receivers and spectrum analyzers, monitoring personnel usually use experience to determine whether they are normal signals or interference signals. For interference signals, they can only record the spectrum envelope (trace) or spectrum screenshot (screen). However, there may be many transmission signals that meet similar spectrum envelopes and spectrum screenshots, and the recorded information is not sufficient to serve as evidence of interference signals.

The RF streaming disk records all the information of the digitized RF signal. It can not only reproduce and analyze the interference signal in the frequency domain, but also analyze the emission parameters and transmission content of the interference signal with the help of demodulation, decoding and other technical means, which is sufficient and complete evidence of the interference signal.

2.2 Signal demodulation and decoding analysis

At present, there are relatively complete technical means to intercept analog signals. However, due to the complexity and diversity of modulation methods, source and channel coding, encryption and other reasons, digital signals are difficult to intercept. RF streaming has opened up the application of post-analysis of monitoring signals. Radio monitoring personnel can use various signal analysis software to demodulate, decode and analyze streaming data, or they can transfer the recorded content to professional units for information analysis.

Taking the recorded unknown digital modulation signal as an example, the bit stream of the signal can be restored by using the Vector Signal Analyzer software to automatically or manually set the modulation mode, symbol rate, filter and other parameters. The transmission information can be restored from the bit stream by using a suitable decoder or protocol analyzer.

2.3 Digital Signal Processing Analysis

The IQ data recorded by RF streaming can be analyzed by digital signal processing software such as Matlab and VSA. The recorded signal can be re-analyzed in the frequency domain, such as changing the resolution bandwidth (RBW), reference level, detection mode and other settings during signal playback, and adding digital filters. The signal can also be analyzed using methods that can characterize time domain and frequency domain information, such as short-time Fourier transform, time-frequency analysis, and wavelet transform.

2.4 Signal Statistical Analysis

Monitoring receivers and spectrum analyzers cannot directly measure statistical information that requires long-term observation and calculation, such as spectrum utilization, signal occurrence probability and frequency, signal correlation, and time-division signal period. There is also a large measurement error when analyzing the signal envelope recorded by frequency sweep. With the help of dedicated playback, search, and statistical analysis software for RF streaming and tools such as Matlab, signal time domain and statistical analysis problems can be accurately solved.

3. Factors to consider when choosing a RF streaming system

RF streaming is a system that includes RF converter, data collector, data recorder, data playback and analysis software, etc. The factors to consider when selecting a RF streaming system include:

3.1 Recording signal quality

In order to ensure accurate analysis, demodulation and decoding of the recorded signal, the quality of the RF streaming recorded signal is crucial. The quality of the recorded signal is mainly determined by the performance of the RF circuit and the intermediate frequency circuit. The in-band flatness and linear phase shift of the RF front end need to be calibrated to ensure that the signal is not distorted during the down-conversion process. The intermediate frequency circuit mainly includes signal conditioning, digital-to-analog converter (ADC) and IQ demodulator. At present, the ADC of the streaming system is usually 12bit or 14bit. The higher the sampling bit of the ADC, the smaller the quantization error and quantization noise, and the larger the dynamic range of the system. Spurious Free Dynamic Range (SFDR) is an indicator to measure the comprehensive performance of the two parts of the circuit. SFDR refers to the ratio of the RMS amplitude of the maximum signal in the band to the RMS value of the next largest noise component or harmonic distortion component, usually expressed in dBc. The SFDR of the RF streaming system with a 14bit ADC is usually greater than 75dBc within a 200MHz recording bandwidth.

3.2 Frequency range and bandwidth

The frequency range of the RF streaming system is determined by the frequency range of the front-end downconverter or signal analyzer, and currently covers shortwave, ultrashortwave, and microwave. The bandwidth of the RF streaming system is related to the performance of the ADC, the bandwidth of the data bus, and the read and write speed of the disk array. Currently, there are many solutions on the market with a maximum recording bandwidth of more than 100MHz.

3.3. Recording time

The RF streaming recording time is directly related to the recording bandwidth and storage capacity. Taking 100MHz recording bandwidth as an example, the IQ sampling rate is usually 2.5 times, each sample point is stored at 2Byte (12 or 14bit ADC), the recording information rate is 500MB/s, and a 1TB hard disk can record for 33 minutes.

RF streaming can be set to enable trigger conditions such as spectrum mask trigger, environmental level trigger, time-limited level trigger, data frame trigger, etc., so as to only record the signals of interest and reduce the amount of data.

3.4. Additional Information

In addition to recording frequency band IQ data, RF streaming can also record GPS location and time, markers, system configuration and other information. In subsequent analysis, these additional information can be used for map mapping, signal classification, statistical processing and other tasks.

3.5 Data playback and analysis software

The analysis of the recorded data on the disk provides more valuable information for monitoring. First of all, the recorded data format must be open and can be called by signal analysis software such as Matlab and VSA and demodulation and decoding software. Secondly, the RF streaming solution needs to provide software tools to play back, search, capture and save the recorded data.

4. Keysight Technologies RF Streaming Solution

The system composed of Keysight Technologies N9040B signal analyzer and X-COM's IQC5000B signal recorder is a typical representative of the current high-performance RF streaming system. The entire system not only has excellent hardware indicators, but also the supporting Spectro-X and 89601B signal playback and analysis software are powerful. The system's RF front end is the N9040B high-performance signal analyzer, and its analysis bandwidth (1GHz), real-time spectrum bandwidth (510MHz), spurious-free dynamic range (78dBc), phase noise and other indicators are fully leading in the industry. The IQC5000B signal recorder is widely used in foreign signal monitoring, defense and security fields. Some indicators are shown in Table 2:

Table 2. IQC5000B RF streaming recording equipment specifications

The system hardware is shown in Figure 2. The whole system is simple to connect and configure. Both the streaming control software and the signal playback analysis software can be installed on the N9040B signal analyzer, without the need for an external control computer. The streaming recorded data is transmitted to the N9040B or an external workstation via the PCIe Gen 2 bus for analysis.

Figure 2: N9040B + IQC5000B RF streaming system

Spectro-X is a software provided by X-COM for signal playback analysis. It supports playback analysis of spectrum, IQ envelope, waterfall chart, as well as signal search, interception and some digital signal processing functions. The data format of IQC5000B recording is .xdat file, and Matlab interface function class is provided. Figure 3 is the result of time-frequency analysis of IQC5000B recording spectrum using Matlab.

Figure 3. Time-frequency analysis of xdat data

5. Summary

RF streaming expands the technical means of signal monitoring and analysis. In conjunction with broadband real-time spectrum functions, RF streaming is very suitable for recording and analyzing transient and jump interference, and statistical signal period and duration information. For some interference signals, the signal monitoring site may not have the processing means or processing capabilities to analyze them. RF streaming can be used to perform post-analysis, demodulation and decoding of the signal. As a new technical tool, RF streaming has many applications in signal monitoring work that deserve in-depth exploration and research.