December 7, 2021
As more complex wireless signals are introduced to the ever-crowded RF spectrum due to the rollout of 5G and expanded IoT use cases, the ability to effectively capture and analyze those signals is in greater demand. Interference — unintentional and nefarious — threatens the fidelity of communications systems in high frequency and millimeter-wave (mmWave) frequencies.
For these reasons, analysis of IQ data has become an important tool in the signal analysis toolbox used by field engineers and technicians. Using standard software applications, every detail of an RF signal can be analyzed and categorized. Leveraging IQ data capture and streaming capabilities allows network management to ensure key performance indicators (KPIs) are met.
IQ data is gathered using a signal analyzer, such as the Field Master™ Pro MS2090A (figure 1), that processes RF signals. To effectively use the acquired signals for additional insight requires that the test solution has three critical specifications to gather spectrum intelligence:
- Frequency Coverage: Full coverage up to 54 GHz ensures there are no gaps where signals can be missed.
- Bandwidth: 110 MHz of capture bandwidth means more information across more spectrum.
- Real-Time Performance: A minimum detectable signal of 5 ns means even the shortest duration signals cannot hide from detection.
IQ Streaming Explained
A signal analyzer captures portions of RF energy hundreds of thousands of times a second then processes that data into the frequency domain using a fast Fourier transform (FFT). The analyzer can translate the information into the graphic displays of spectrum that are typically associated with the instruments.
IQ streaming is the ability of the analyzer to take that data before it is processed, break it down into the I and Q parts, and store it to an external device for later processing. This enables greater insights into the signal, such as checking modulation quality, slowing the playback of the signal to see changes over fractions of time — down to nanoseconds of time resolution — or analyzing other signal characteristics that can help identify and track the signal.
Data Stream Options
Several options for configuring the data stream can be used, depending on application:
Capture Bandwidth and Sample Rate – The capture bandwidth and sample rate are tied together to determine:
- frequency span of the capture from the center frequency
- rate at which the analyzer samples the IQ data (see figure 2)
The Field Master Pro offers up to 110 MHz of capture bandwidth at 200 MS/s (option dependent), but more bandwidth is not always desirable. Additional bandwidth and faster sample rates mean more data must be stored into memory and analyzed in post-processing. Streams of 110 MHz fill up storage much faster than a 1 MHz capture bandwidth. So, the key is to select as narrow a bandwidth as possible for a particular application.
A regulator conducting spectrum monitoring may want to see spectrum activity on as wide a span as possible. Compare this to a radar transmitter that may only care about kilohertz worth of spectrum where their signal resides. Setting the right bandwidth will make it much easier to process at a later time.
Timestamp – Users can choose whether to capture a timestamp with each IQ sample. A timestamp enables precise correlation between the time and any spectral phenomena, which is beneficial. There is a tradeoff, though. A timestamp takes up bits within the capture. So, the user must give up some resolution in the capture, resulting in a rise in the noise floor of the playback, or capture a greater bit depth. While the latter option may help offset the loss in resolution of the signal, capturing more bits equates to more data and more memory used.
Sample Format (Bit Depth) – Given unlimited bit depth and sample rate, an IQ capture can perfectly recreate an RF signal from any period in time. More bit depth and additional samples come at a cost of more bits that can quickly consume gigabytes of storage and tens of gigabits per second of streaming bandwidth.
Every capture contains noise. At some point, more bits only serve to better quantify the noise and no longer provide additional insight into the signal. More details on selecting proper bit depth for a given application are provided in Anritsu’s Choosing the Best Bit Depth for IQ Captures or Streams white paper.
Selecting a Storage Location for the Stream
There are many possible applications that require IQ data analysis, different use cases will have different requirements. Some cases might call for large data amounts to be stored and analyzed at a time. For other applications, portability may be the most important factor. Choosing the right device will depend greatly on the application.
One important consideration is the possible trade-off between capture bandwidth and gaps in the IQ data. A good analogy is to think of the data like water flowing through pipes. Different storage devices have different size pipes, therefore, support the transfer of different data amounts. When more data is being acquired that can be transferred, gaps will occur in the final IQ file. Streaming the data without gaps requires fast enough transfer speeds, so the data is cleared from the spectrum analyzer RAM and into the target storage device.
IQ data can be streamed to various conventional storage devices. Examples include USB 3.0, PC via PCIe, PC via Ethernet, or X-COM’s IQC5000B RF Record and Playback System. Some options are more portable but are limited in the amount of data that can be streamed. Others offer very wide bandwidth streaming but require extra equipment and/or cost. Table 1 shows each option with special considerations and is followed by more detailed explanations in the next several sections.
You can learn more about the importance of IQ streaming by downloading the IQ Streaming with Anritsu’s Field Master™ Pro Real-Time Spectrum Analyzer application note.