PIM in Mobile Broadband Networks—Measurement and Management
by Nigel Chapman and Fiona Wilson, AceAxis
Passive Intermodulation (PIM) is an unpleasant side effect of the successful deployments of mobile networks, and is a problem that is growing in impact as complexity increases with the deployment of 4G and—in the very near future—5G networks. PIM has the potential to degrade the efficiency of a cell site, and this network degradation directly impacts the edge of cell performance and/or the throughput of the cell site.
What is PIM?
PIM is a form of intermodulation distortion that occurs in components that are normally thought of as linear, such as cables, connectors and antennas. However, when subjected to the high RF power levels found in cellular systems, these devices can generate intermodulation signals at -80dBm or higher.
Passive intermodulation signals are generated late in the signal path, so they cannot be filtered out and they may cause more harm than the stronger, but filtered, IM products from active components. A PIM test is a comprehensive measure of linearity and construction quality. PIM shows up as a set of unwanted signals created by the mixing of two or more strong RF signals in a nonlinear device, such as a loose or corroded connector, or nearby rust. Other names for PIM include the “diode effect” and the “rusty bolt effect.”
The following expressions can be used to predict PIM frequencies for two carriers with frequencies F1 and F2:
nF1 – mF2 [1]
nF2 – mF1 [2]
where the constants n and m are integers
When referring to PIM products, the sum of n + m is called the product order, so if m is 2 and n is 1, the result is referred to as a third-order product (Figure 1). Typically, the third-order product is the strongest, causing the most harm, followed by the fifth- and seventh-order products.
These PIM products in turn can mix with signals from other sources, producing intermodulation across a wide bandwidth that has the effect of a raised noise floor, some of which will be likely to fall into one of the cellular receive bands.
What Causes PIM?
PIM is created via three primary mechanisms:
Poor installation of the cell site—where dirty, loose, or poor PIM quality components have been used at the cell site—or simply poor configuration of the cell site, for example the way in which the antennas are positioned relative to other antennas or cell sites.
Physical effects that may be created when the antennas radiate into a PIM reflective material, for example a rusty roof or rusty chains. With densification efforts ongoing, it is increasingly difficult to find “clean” cell sites that are PIM–free. Even tower-mounted antennas commonly suffer from PIM due to the equipment mountings themselves.
Adjacent RF bands: carrier aggregation is a key requirement of 4.5 (LTE-A) and 5G networks, yet aggregating carriers carries with it the risk that the multiple carriers that are aggregated will create PIM.
As the density of cellular solutions increases, PIM effects will also increase. It is important to note that whilse PIM is mostly observed with high-power cell sites, it is present even at low power levels, and the effects of PIM will continue to grow across all types of RF systems. This is particularly significant in 5G with the growth of seamless integration of multiple base station technologies to service user needs.
Often PIM has been viewed as an installation problem, and while it is absolutely true that good site installation will minimize PIM, by its very nature it is an ongoing and evolving problem. The industry has worked hard to address PIM at the cell site during installation. However, this does not mean that just because a site is PIM-free today that tomorrow PIM will not occur. Today PIM is more likely to occur due to adjacent bands and/or physical effects in the vicinity of the cell site. For example, if a new RF band is added to an existing cell site, a new physical structure is added within the range of a cell site, or over time the cell site connectors corrode or work loose, this is when PIM will reoccur. The cellular industry is constantly updating and growing the network to meet bandwidth demands, and hence PIM cannot be considered “just” an installation problem.
How Does PIM Impact the System Performance?
PIM raises the noise floor, and hence desensitizes the receiver. As a simple example, consider the case where 2 x 20W, 5MHz carriers impinge on a PIM-causing defect. The receiver has a 5MHz channel bandwidth and a noise figure of 2dB, giving an inherent noise floor of -105dBm. The PIM measured from a CW test has been shown to be ~2.1dB lower than that when the defect is excited by a UTRA signal1, so we add this conversion factor. The noise floor of the receiver raised by the impact of PIM as measured by a 2 x 20W CW test is plotted in Figure 3.
Antennas and antenna-line components are typically specified to meet -150dBc (using the 2 x 20W CW test). This is a challenging specification to meet in factory conditions, and even more so in the field. It can be seen that this would have negligible effect on the noise floor. However, any degradation in this figure quickly begins to impact the noise floor and hence, sensitivity of the receiver. For example, should a component degrade to -145dBc, the uplink budget would be reduced by 3dB.
PIM only impacts the uplink and this can result in an imbalance in the uplink and downlink budgets. As well as reducing the available data rates and throughput on the uplink, this has the undesirable effect of disrupting the handover process. Since the UE bases its handover decisions on the downlink signal, it will remain attached to a base station suffering from PIM, even when the uplink performance to a neighboring cell would be much better.
Every increase of the noise floor and interference level leads to system degradations like cell range reduction or reduced system capacity, and subsequently means a financial loss for the operator.
The Changing PIM Test Landscape
PIM is not a new problem, and the industry has worked hard at developing solutions to test for and eradicate PIM. Today there are three primary methods used to detect PIM:
RF based PIM testers: these are portable test boxes that allow the engineer to generate two 20W RF tones at a set RF frequency and to measure the effects of PIM
Spectral analysis over CPRI: a relatively new method, in which analysis of I/Q data on the fiber interface between the BBU and RRH enables the user to look for the characteristic profile of PIM
Noise rise monitoring: using the installed cellular equipment, set orthogonal channel noise simulator (OCNS) on all carriers under test and measure received signal strength indicator (RSSI)—per resource block where possible—in each potential victim uplink band.
Table 1 compares the pros and cons of each of these methods. Although they have been instrumental in helping to reduce the effects of PIM, they do have limitations, and the industry needs new solutions to support current and future 4G and 5G networks. There is a need for a strategy that incorporates the existing solutions, and which can build upon these so that operators have a comprehensive and cost-effective tool kit to deal with the PIM problems of today and tomorrow.
To define such a strategy, there are some key factors that need to be considered, as follows:
Cost: test equipment, training, support teams and network downtime
Cell site considerations: whether this is a new cell site in commissioning phase, and whether it is a tower cell site or rooftop cell site with easy access
PIM scenario: PIM is complex and there are number of ways it can occur, including:
Single-band PIM, created from a single RF frequency band
Multi-band PIM, created from multiple RF bands interacting with each other
On-site PIM, created on the physical cell site
Off-site PIM, created remotely to the cell site
For each PIM scenario, we also need to consider the bandwidths under investigation, the IM product that needs to be analyzed, and critically, the level of PIM that is present (for example, do you really need to tackle an IM11 problem if you have an IM3 elsewhere in the network?)
New Method
A new method—PIM over CPRI—has been developed to provide far greater insight into PIM than the techniques used to date. This allows the user to:
Monitor live traffic remotely without disturbing cell site RF equipment
Accurately identify and measure the location of PIM to better than ±2m off site and ±1m on site
Accurately measure PIM level near or below the RRH noise floor and/or in the presence of busy uplink
Measure PIM for complex multi-band scenarios
Provide a frequency-agnostic test method
Predict the practically realizable PIM mitigation level
PIM over CPRI works by analyzing the fiber optic link between BBU and RRH in the network. The downlink and uplink digital data streams to and from the RRH are analyzed by complex algorithms for PIM. With so many ways to tackle PIM, which one should be used? Table 2 provides some guidance on the best technique to use for different PIM test capabilities.
Of course this is just a guide, and the myriad of possible ways in which PIM can occur, and the real-world challenges that exist may dictate different strategies that would also be viable.
We should at this point also mention the option of PIM cancellation, or more precisely PIM mitigation. This is an extension of PIM detection over CPRI, since if we are able to detect PIM over CPRI then we are able to cancel the PIM as well. Today this method can be used to address PIM at the cell site—however, there are some technical and commercial issues that significantly limit widespread implementation of this method, including the following:
To cancel PIM for a reasonable number of likely PIM scenarios will involve significant digital processing resources, which in turn requires expensive hardware and software at the cell site. The ROI needs to be carefully considered. PIM cancellation will only be valid for cell sites where the levels of PIM are such that the benefit is clear.
Today the fronthaul interface is largely closed and the preserve of the OEM. Any widescale deployment of PIM cancellation requires either OEM support or an open interface.
Perhaps most significantly, PIM has a constantly evolving effect on networks. Making changes to the cell site, physical work near the cell site, ageing, and general wear and tear, all contribute to a continually changing dynamic PIM environment. PIM cancellation is able to address a specific use case at a point in time and cannot be guaranteed to adapt to all changes to and around the cell site. For this reason, it is likely that over time, PIM cancellation will only be able to cancel some specific scenarios, hence we refer to PIM not as cancellation but as mitigation.
For the reasons defined above, AceAxis believes that today it is better to develop PIM detection tools to help operators quantify and identify the nature of the problems they face at a cell site. Understanding the problem allows the operator to develop strategies to eliminate PIM which may include data to support an ROI for deployment of PIM mitigation.
Conclusion
PIM is a problem that is growing in significance. As we develop and deploy even more complex 5G networks, we have to ensure that tools and methods exist that enable networks to be optimized and kept free of PIM.
These tools and methods have to be cost effective, capable of dealing with ever more complex PIM problems and scenarios that are occurring, and above all easy to use, so that they can be implemented by field engineers without the need for detailed knowledge about the increasingly complex RF environments being deployed.
AceAxis has focused on developing PIM detection and location over CPRI using low cost portable network sensors and software applications that can be run on a PC or in the cloud—this offers the most comprehensive way to test for PIM at an affordable price to support complex network scenarios.
References
1. Roy Naddaf, Rune Johansson, Bo Franzon, “Passive Intermodulation in Real Networks,” MULCOPIM ‘08 conference 24-26 September 2008, Valencia, Spain.
(1788)
