June 5, 2020

WiFi Interference on Microwave: Laboratory testing observations and conclusions

By Sergio Licardie, VP of Advanced Technologies and Innovation at Aviat Networks

The FAS expert system is designed to use several rules to determine if interference is happening in any of the radio links registered for monitoring in the application. As part of the overall design and implementation efforts, Aviat has conducted many laboratory tests that were intended to determine the effects of different types of interferer signals in point to point microwave links. These tests allowed not only for a better understanding of how the performance parameters of the radio link were being affected in the presence of interference; but also to the creation, validation, and refinement of the rules that were implemented in the FAS expert system. As we understand more about the impact interferers can cause to the radio links, the expert system rules will be refined, and new rules will be added. This process will be part of the overall expert system learning and evolution.

The built-in rules of the expert system use several performance parameters to be able to identify interference uniquely from other radio link impairments. In essence, the main signature of interference consists of a strong received signal level (RSL) accompanied by one or more degraded performance indicators, like SNR, BER, ES/SES, or Alarms in the link. These considerations not only allow the expert system to detect interference but to classify it in up to 5 levels of severity:

1. Link disrupting interference – when the radio link is lost due to interference.
2. Error generating interference – when the radio link is experiencing errors due to interference.
3. Performance affecting interference A – when the radio link is stressed due to interference, this causes modulation changes, power adjustments and/or link availability reduction, but not to the point of causing errors. This type of interference usually goes unnoticed.
4. Performance affecting interference B – In this case, the radio link degrades due to the interference, but such interference may be too low to trigger any compensation action, or the compensation mechanisms are not enabled. To determine if and how much degradation exists in this case, we will use a more sensitive set of parameters based on the U-BER. The U-BER will tell us how hard the Forward Error Correction decoder is working to clean the received signal. This type of interference usually goes unnoticed.
5. No interference or negligible interference level.

The expert system will apply several rules to determine which type of interference is being detected or if this detected instance may be considered a false positive (unconfirmed interference):

• Persistence, false positives (unconfirmed), and bursty/steady interferers – validates duration of interference, thus enabling determination of false positives, short single interference bursts, burst interferers that are constantly affecting the radio link, or steady interferers.
• Multipath – validates if the type of interference should be considered multipath or confirm that it is due to an external interferer.
• Symmetry – validates if the same degradation pattern is happening in both directions of the radio link. If this is the case, interference is unlikely as it is usually not symmetric in nature.
• Historical reference and regular/irregular behavior – validates against the historical database reference for the link under analysis if the interference pattern is happening at regular intervals or if it is showing an irregular behavior.