While resolving RF interference problems in complex RF environments can be a tricky task, gapless recording offers a viable solution to the measurement challenges presented by the typical signal analyzer. The technique solves the problem of not knowing when or where an interference event will occur, or how long it will last, by enabling continuous acquisition of data over long durations. Because there is no gap in the data recorded, the signal-of-interest, such as an intermittent RF event, is easily captured.
Figure 2: In this block diagram, the signal analyzer in Figure 1 has been modified for gapless recording Click on image to enlarge
An example of a signal analyzer modified for gapless recording is illustrated in Figure 2. It is the same signal analyzer shown in Figure 1. However, it now includes a high-speed data link or bus that allows the engineer to move data from memory as it is acquired. By bypassing processing and display updates, and writing acquired data directly to final storage using a circular Random-Access Memory (RAM) buffer, it’s possible to create high-bandwidth recordings with no gaps in the data. With a circular RAM buffer, the engineer can simultaneously write to and read from it. When recording at wide bandwidths for long durations, a Redundant Array of Independent Disks (RAID) storage system is required.
One such wideband gapless recording solution is Agilent Technologies’ dual-channel M9392A PXI Vector Signal Analyzer, which can provide two independently tunable channels—each able to record data at 100-MHz bandwidth over many hours (Figure 3).
Click on image to enlarge
Figure 3: Agilent’s gapless recording system comes in predefined packages that rely on the wide bandwidth and fast throughput benefits of PCIe, and are used with either a regular PC hard disk drive or external mass storage. The system shown here offers 32 TB of storage.
While wideband recording in an RF environment has proven itself useful as a characterization tool for long duration RF interference studies, powerful search tools can reduce the burden of searching long recordings for culprit signals. The Agilent 89600 Vector Signal Analyzer software, for example, can be used with the M9392A to provide key insight into the characteristics of the interferer and its effects on the victim signal in select data obtained during gapless recording. Using such software simplifies and reduces the time to find target signals-of-interest and also speeds up the process of analyzing and fixing problems.
It may also be helpful if the gapless recording solution sports key functionality like time-stamping so that recorded data can be mapped to an absolute time, triggering and pre-triggering. Pre-triggered data provides engineers access to signal data leading up to a specific trigger event.
Another key capability is dual-channel recording. In a single-channel recording system it can be difficult to trigger on only the desired signal. As a result, more data is usually recorded than is actually required to ensure the interference event is captured. This additional data takes time and resources to process. A dual-channel recording system like the M9392A reduces the likelihood of false triggering and provides an innate ability to record just the data that’s needed. Signals can be acquired and triggered on one channel, while being recorded on the other. By enabling more efficient discovery of the signals in the RF environment, such discretionary triggering saves a great deal of time and also helps the engineer more effectively solve interference problems.
"" Tracking down interference in complex RF environments "" well yea, but:
Interference usually is in the receivers, hand-helds, base stations or auxiliary receivers.
If the hand-helds are moving about, then interference zones may be noticed by the users. , in a particular building, a geographic square block, or fraction of a larger area.
The engineer is in the know, who is a close frequency user. as adjacent channel user, or third intermodulation contributor. all of this comes into play "the math" and analysis before, the recording data comes into play.
The recorder itself, just shows, the place and time values. ' it is not really the solver of the fact " however an item of merit you may of wanted to touch on was.
Propagation measurements, of band by band comparison, to fulfill an existing radio footprint plan system. that is of merit, similar or better.
In my studies of this, equal merit or better, the recording data is "Missing" or has not been done at all.
A lot of new systems are poorly designed in coverage. getting in building coverage has been a failure, the recording analyzer will tell you what you have now, and when the new system goes in, what you have got to replace it with .
Now i am simplifying my thought here, as this is a very important topic. \
i will say, in my work experience as a principal, i ordered ten percent of project value to be in Test equipment, and bought many pieces of gear, by your wonderful company ""AGILENT""
in standing down, the users, will tell you the noise interference problems, and the locations, and when they have occurred , "you just have to ask" don't worry about ambiguity, and when you have solved, " good luck " ask again.
Dave, i so enjoy your company's expertise. i value the technical thought as first class. active voice and problem solvers, " i thank you "
from Ralph T. Gerwing yvr.ca Metro Vancouver Cheers.
pretty penny this analyzer your promoting, but, in real world problem solving " especially building new band services " it is very valuable indeed.
Thank you for your comment that it is actually the knowledgable Engineer that is critical to solving the interference problem and your keen observation [borne out of experience] that there is often a whole lot of detective work done before the measurement system even becomes necessary. I certainly agree with your assertion.
This was a short paper in which I attempted to show that a recording system can be very useful for helping to root-cause a specific interference problem caused by signals in the external environment that are both intermittent and probably non-compliant.
Having a permanent record of the signal environment, which is known to cause an interference problem with some RF system or device, affords us the luxury of processing this data "offline" back in the lab.
Even with a recording system, it can still be tricky to capture the actual culprit signal, as it occurs, because the analyzers triggering system may false trigger on other signals too.
Note that I'm suggesting it is generally not too efficient to record hours and hours of data.
Even a few seconds of recording at wide RF bandwidths can produce Giga Bytes of data.
After all, who has time to process that much data.
It is usually much better to selectively record.
In that case I posited that an additional independently tunable receiver could be tuned specifically to the frequency and bandwidth of the suspected intererence signal to provide more selective triggering capability, since the recording in the afflicted channel could be initiated by this trigger channel.
This is funcionally equivalent to a frequency selective trigger with the added benefit that we can trigger off of out-of-band signals.
Not covered in this paper but also very useful is the fact that these recording systems have significant amounts of pre-trigger capability that allow us to record many seconds of data prior to the actual trigger event.
This opens up the possibilty of applying some signal processing to the data in the triggering channel to make a sophiticated criteria based trigger.
It doesn't matter that this processing could take seconds; the pre-trigger capability of the recording channel provides the time.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.