Radio spectrum analysis

Radio spectrum analysis

In an example radio spectrum analysis, we study anomalies. The unexpected signals in the radio spectrum are caused by intentional and unintentional sources. Intentional interference can be resulted from illegally imported radio transmitters or jamming devices. Unintentional signal sources are commonly mal-functioning transmitters, which transmit in a wider frequency range than they should, or they can be license-exempt devices placed too close to each other. Independent of intentional or unintentional sources of anomalies, they should be detected, located and processed to maintain reliable communications.

Artificial intelligence has developed fast in recent years, and its application areas have reached most areas of society where electronic data is available. Anomaly detection is one of the mainstream applications of artificial intelligence. An example of that area is fault prevention by detecting changes in machinery before a critical fault occurs. Artificial Intelligence (AI) system is an effective method to detect anomalies in radio signals when the radio spectrum is static or when there are few discrete static states in the radio spectrum. The method can accurately detect frequency hopping anomalies, but when the normal radio spectrum environment has wanted frequency hopping signals, the directly applied method may not work well anymore. A combination of classification capability and careful configuration of the spectrum scanning extends the applicability of AI anomaly detection in the radio spectrum environment with wanted signals.

Radio spectrum statistics

Fairspectrum spectrum reservation statistics tool is implemented as a web service, and it can be accessed with a web browser. The Fairspectrum LSA Statistics tool helps the customer to analyze e.g. the following issues: Which frequency bands are used? Where spectrum is used? In which areas spectrum is used often? In which areas spectrum is used but not often? How often all spectrum resources are reserved? How often there is more demand than availability? How long the frequency resources are used at a time? What is the long term trend of the use? What kind of seasonal variation there is yearly, seasonally, monthly, weekly, or daily?