October 22, 2013

“2013 is shaping up as a kick-off year for small cells, driven by 4G small cells deployed for capacity upgrades.”

Those were the words of Stéphane Téral, principal analyst for mobile infrastructure and carrier economics at Infonetics Research, in a report earlier this year. It was a thought echoed by Ericsson EVP and head of networks Johan Wibergh who back in January said the age of dense small networks is about to begin.

There is no doubt that an incredibly large volume of small cells are going to be deployed. Given the typical locations where these cells are installed, such as on telephone poles and street lights, there will certainly be lots of metal objects nearby. So, while there is much excitement and promise with small cells, there is one aspect that might temper the enthusiasm – Passive Intermodulation (PIM).

Even though the power levels are significantly lower than a typical macro cell, our experience indoors has shown us that power levels as low as -20 dBm have the potential to generate high PIM. Luckily, for indoor installations, the operator usually has the freedom to move the antenna within a 1 m diameter to reduce PIM generated by a nearby PIM source. With small cells, this degree of freedom is lost. Moving the telephone pole 1 m is not a practical option.

Case for Quasi-omni Antennas

A few months back I had the opportunity to work with a network operator that recognized this problem and sought a solution. They theorized that deploying a quasi-omni antenna, an antenna with a “clover leaf” azimuth pattern rather than a true omni directional pattern as shown in Figure 1, might give them the ability to “steer” one of the azimuth nulls in the direction of a nearby PIM source. This could be accomplished by simply rotating the antenna on the mount. If successful, it gives operators back a degree of freedom with small cell deployments for mitigating PIM.

Figure 1

We used a 700 MHz Anritsu PIM Master MW82119A to measure PIM results on several quasi-omni antennas, as well as traditional “stick” omni antennas. A 20-foot jumper cable was connected between the PIM Master to the antenna under test. The antenna was mounted on an azimuth positioner inside an anechoic chamber, allowing us to accurately rotate the antenna 360° while measuring PIM inside a low-PIM environment.

To create an external PIM source, we soldered a Germanium diode to a wideband printed circuit board dipole, and constructed a PVC structure inside the chamber to place the diode at various heights and distances away from the antenna under test. Here is what we found:

Figure 2

Quasi-omni antennas – The idea works! Measuring PIM vs. TIME while rotating the antenna shows a significant reduction in PIM when the null in the azimuth pattern is facing the PIM source. We achieved as much as 30 dB reduction in PIM by simply rotating the antenna (figure 2).

Indoor omni – As expected, there was virtually no ability to change the PIM level by rotating the antenna (figure 3).

Figure 3

Conclusion

Our conclusion? Quasi-omni antennas look like a good option for small cell deployments, especially for sites where PIM sources are unavoidable. The “clover leaf” azimuth pattern will have little impact on coverage in a small cell environment, but will provide an extra degree of freedom for resolving PIM issues.

Read the original article here

Acknowledgment: I would like to acknowledge Nicolas Cordaro at Verizon Wireless for proposing quasi-omni antennas as a solution for reducing PIM in small cell deployments and CSS Antennas for providing the antennas and the measurement range to enable evaluation of the concept.