Detecting RFI, revealing the causes of interference and locating unauthorized transmitters are tasks performed by modern direction finding equipment. Spectrum analysis and time domain displays give much information about the type of signal but are limited to recording the amplitude versus frequency and/or versus time, and often lose information due to display compression. The full picture is only available when the measured values are recorded without compression and separated into their real and imaginary components, usually referred to as the in phase and quadrature components, or I/Q for short.
The IDA 2 Interference and Direction Analyzer provides the full picture as it does not simply record and save the I/Q data. It can also evaluate the data immediately on-site, as the results are needed for tracing impairments and interference straight away. Of course, results can also be checked when back in the office.
I/Q Analyzer Mode
With the IDA 2, the I/Q analyzer mode can be selected, just like spectrum or time domain (scope) modes. As in time domain mode, the IDA 2 runs in zero span mode as an I/Q analyzer, being tuned to a fixed frequency, i.e. one channel that is selectively captured. The ability to set unusually high channel bandwidths (CBW) of up to 32 MHz is a special feature of the instrument.
When the measurement process is started, the IDA 2 records the results continuously in real time as I/Q data pairs with a memory depth of 250,000 data pairs. The instrument can even perform some evaluations online, e.g. displaying the pure I/Q data or the magnitude versus time, computing a High Resolution Spectrogram or building up a Persistence Spectrum.
When the measurement is stopped, either manually or by automatic trigger, the last 250,000 I/Q data pairs are still stored, uncompressed, in the background. In this way, any evaluation and display can be produced subsequently from one and the same data set.
Although it is necessary for the IDA 2 to compress the spectrums to correspond to the available number of display pixels in High Resolution Spectrogram Full display mode, everything is shown in High Resolution Spectrogram Zoom display mode: Every line of pixels corresponds to exactly one spectrum, with the color indicating the particular level. The IDA 2 also writes a specified number of spectrums over each other in a Persistence Spectrum where the color indicates how often a particular level value occurred.
Figure 1 High Resolution Spectrogram Full of two LTE channels (resolution 8 μs) showing the frame structure with its resource grid and synchronization signals.
Examples
An application where the I/Q analyzer can be utilized is in the GSM field to establish whether there is interference or an illegal transmitter hidden under the ‘active’ spectrum. This can be particularly difficult to determine if the GSM modulation method uses frequency hopping, where the communications channel switches frequency about every 4.6 ms. If the illegal transmitter also hops, it cannot be detected in the normal spectrogram. However, it is visible in the High Resolution Spectrogram, obtained from the I/Q data, revealed by the different duration and correlation.
Another example is LTE, where interference due to intermodulation from (and with) GSM signals is not uncommon because the antennas are usually located together on the same roof. The rectification effect of a couple of rusty rivets on the mast is enough to generate intermodulation that is superimposed on the RF field. A first for a handheld device: the High Resolution Spectrogram of the IDA 2 makes the whole frame structure visible (see Figure 1).
In the case when an interference signal is hidden beneath a DAB channel, this can best be detected in the transmission ‘gaps’: DAB transmits a null character for synchronization at fixed intervals, during which only the carrier frequencies remain. Any interference cannot avoid detection in the High Resolution Spectrogram as well as in the Persistence Spectrum of the IDA 2.
Figure 2 A GSM downlink signal triggered on the rising edge (time resolution 2 μs) showing the 546 μs timeslots (frame duration is measured using markers A and B: Δt = 4.616 ms).
In the case of an illegal transmitter, deliberate jamming, an unknown defective device, or intermodulation from authorized communications signals, the signal versus time characteristics often tell much about the type of signal. The magnitude setting of the IDA 2 displays the magnitude of the I/Q data versus time, so that the timeslot structure of a GSM intermodulation can be clearly seen, as shown in Figure 2.
Trigger
The IDA 2 shifts the I/Q data continuously through its memory on a first in, first out basis during the measurement. Just as with an oscilloscope, the triggers can be set to capture the measurement results when specific events occur, e.g. when a specific level is first exceeded or whenever this level is exceeded. The Trigger Delay setting is important because it facilitates the capture of the measurement values before and after the event, illustrating both the cause and the effect. Also, the SAVE function stores the I/Q data permanently in the IDA 2 memory for later evaluation.
Evaluation of I/Q Data
There is a causal relationship between the channel bandwidth, resolution bandwidth, window overlap, time resolution, and possible recording time in any digital analyzer like the IDA 2. At the maximum CBW of 32 MHz, the IDA 2 captures an I/Q data set every 31.25 ns, corresponding to the inverse of the CBW. This gives a recording time of 250,000 × 31.25 ns= approx. 7.8 ms with a memory capacity of 250,000 data pairs. This is enough to completely capture cyclical sequences in modulated communications signals. The recording duration increases correspondingly for a narrower CBW, so that at the other extreme, a CBW of 100 Hz would give a recording time of 2,500 seconds.
Figure 3 Persistence Spectrum of a GSM downlink signal showing underlying interference at around 932.8 MHz.
The High Resolution Spectrogram and Persistence Spectrum are possible evaluations of the I/Q data that can be made online or subsequently, but also immediately on-site. Figure 3 shows the Persistence Spectrum of a GSM downlink signal. The IDA 2 uses FFT analysis for this feature. The signals that have already been captured selectively by means of the selected CBW are further separated into their spectral components. Regardless of the setting used for capturing the measured values, users can determine or change the FFT parameters: the number of FFT samples and hence the resolution bandwidth (RBW) within the channel bandwidth, as well as the window overlap, i.e. the overlap of the time segments from the data set that are to be used for a FFT.
The rule is: The fewer the FFT samples and the greater the overlap, the finer the time resolution, i.e. the succession of spectrums. For example, the FFT yields a usable bandwidth of 25.6 MHz for a CBW of 32 MHz. With 256 FFT samples, the IDA 2 computes a spectrum with a RBW of about 240 kHz. If a window overlap (FFT Overlap) of 87.5 percent is selected, spectrums with a time resolution of 1 µs (corresponding to one million spectrums per second) will be obtained. For this reason, other analyzers compress the data for resolutions below 20 ms, whereas the IDA 2 retains the data without compression.
Based on the I/Q data, the battery operated, handheld IDA 2, which weighs 3 kg, offers a depth of analysis that was previously only available using costly and heavy laboratory instruments. Weak or sporadic interference, which may be hidden beneath strong and possibly variable frequency useful signals can now be revealed on site. In doing so, IDA 2 makes a significant contribution to the security of modern communications.
Narda Safety Test Solutions
Pfullingen, Germany
+49 7121 97 32 0,
www.narda-sts.com