June 12, 2015
If you’re like most field engineers and technicians responsible for cellular network quality, you view interference as the bane of your existence. An early indication that interference may be adversely affecting your wireless network is poor switch statistics. Once you’ve received this bad news, determining the cause becomes priority number one. Often, it requires you driving out to the base station to see if it is one of myriad of potential causes – bad feedlines, a broken TMA, poor transmitter power, passive intermodulation (PIM), or something else. In this blog we’ll focus on external interference—the only possibility left when you’ve ruled out those other sources.
Handheld spectrum analyzers have become a frequent passenger on the journey to optimum network performance. For example, when searching for external causes of interference, they are used to acquire a spectrum reference for future analysis, as well as to view the spectrum over time to monitor fluctuations. Our inaugural post is written to help you use a handheld spectrum analyzer to spot interference, characterize interference types, and identify signals to help you ensure cellular network operation meets design specifications.
Spotting Interference
The first step to determine if your cellular network is suffering from interference is to compare real-world performance to KPI targets. If the network is not up to snuff, you can turn on your handheld spectrum analyzer (figure1) and conduct an interference hunt to determine if the problem is internal and/or nearby interference.
Figure 1: The Spectrum Master MS2720T is an example of handheld spectrum analyzers well suited for interference hunting.
The first and best place to start hunting is at the receiver input. This might be from a test port or via a CPRI link, for example. Another approach is to turn off the base station and disconnect the antenna. A third alternative is to have a separate receive antenna that is as high as the base station antenna. It’s best to measure the signal after the pre-filter (if the receiver has one) to have a “visual ID” early on so you are sure you are on the same signal later. Once you have eliminated internal sources, such as bad transmit power or PIM, as potential causes you can concentrate on external factors.
Uplink, Downlink Concerns
Interference can occur at both uplink and downlink frequencies, however it is a bigger concern for the uplink because it can adversely affect the base station receiver and ruin an entire network sector. That is not to say you shouldn’t be concerned with interference in the downlink. In fact, our experiences have shown that while most carriers focused solely on uplink interference in the past, there are now initial indications they are addressing downlink interference, as well.
It is a bit more difficult to see an interfering signal in the downlink, because the large base station transmit signal tends to mask it. To help make the interfering signal visible in the presence of the BTS signal, you have some options:
1. Directional antenna – This can help by up to 10 dB or so for a small Yagi. A panel antenna, as used by the base station, can give you 20 dB or more of gain, at the expense of size.
2. Min-Hold – A feature in some spectrum analyzers that can reduce the modulation on the base station signal by as much as 30 dB. If the interferer is CW, it will stay at the same amplitude.
3. Filters – If you are somewhat fortunate, the interfering signal will be at a frequency near the BTS signal, rather than on top of it. While this will still affect UE receivers, it will let you use the spectrum analyzer (RBW) or other (narrow-band cavity) filter to see the interferer separate from the BTS transmitter.
4. Drive Test – Data acquired via this process will show the location of dropped calls. This is a good place to start looking for the interferer, as the BTS signal will be relatively low, and the interfering signal relatively high.
Characterizing Interference
An important step to remember is to characterize the type of interference before disconnecting from the receiver’s signal. Simply adjust the spectrum analyzer to best view the signal by using the pre-amp, reference level, span, and resolution bandwidth controls. This will allow you to observe the signals’ shape, bandwidth, and behavior. Frequency drift, amplitude changes, and frequency hopping are key attributes to analyze because they:
1. Provide a clue as to the emitter causing the interference
2. Indicate when the signal is on and off
3. Guide you on how to set up the spectrum analyzer when signal hunting
4. Let you capture a reference trace to compare what you see later in the field
The spectral shape of the signal is another clue; one that is perhaps more important than those listed above. If you’ve seen the shape before, you probably recognize it. For example, cable TV has a very distinct pattern, with repeating channels every 6-8 MHz, depending on the country.
Instrument Features You Need
We mentioned the advantages of Min-Hold when locating interference in the downlink; it’s also useful for the same reason when looking at interference in the uplink. Another helpful tool in field spectrum analyzers is Max-Hold, which allows you to create an envelope so it is easier to locate intermittent or “bursty” signals. In these scenarios, a spectrum analyzer with a Burst Detect feature is also advantageous, as it can capture emitters as narrow as 200 µs.
When searching for interfering signals with long intervals between appearances, a “Save on Event” capability comes in handy. This feature uses a mask automatically generated from the “normal” signal and only saves a trace when something unusual appears. Once saved, the traces can be examined for time-of-appearance, and signal characteristics, so you can predict when the interfering signal will appear, rather than waste time taking measurements when the interference is not present.
Figure 2: Spectrogram displays can show field technicians how signals change over time.
Another useful capability is a Spectrogram, which shows how signals change over time. The signal in Figure 2 is clearly shown to be unstable in frequency through Spectrogram. This sort of oscillation in frequency is characteristic of a repeater with insufficient isolation between its input and output antennas. This is one example how Spectrogram helps identify a variety of signals.
To locate signals that don’t belong on the input to a receiver, it’s important to know which signals are supposed to be there. If you know who might be putting out interfering signals, you can save a lot of time. Once again, the handheld spectrum analyzer comes in handy, as it can be used to help identify the signal source.
It may be possible to demodulate the signal and listen for a station ID call sign from a TV or radio station, or a Morse code station ID at the start of a page. For those of you who are experienced, audio demodulation can give significant clues on the type of signal being generated. A good example is power line arcs, which generate a very distinctive audio signal. Sometimes, a signal can be identified by its frequency and location using government databases. For instance, the United States Federal Communications Commission (FCC) maintains a data base of signals and locations, linking them to owners, with contact information.
You can learn much more about locating interference sources by downloading a free application note on interference hunting techniques.
Comments