Wireless signal monitoring and analysis, and related methods, systems, and devices
Wireless signal classifiers and systems that incorporate the same may include an energy-based detector configured to analyze an entire set of measurements and generate a first signal classification result, a cyclostationary-based detector configured to analyze less than the entire set of measurements and generate a second signal classification result; and a classification merger configured to merge the first signal classification result and the second signal classification result. Ensemble wireless signal classification and systems and devices the incorporate the same are disclosed. Some ensemble wireless signal classification may include energy-based classification processes and machine learning-based classification processes. In some embodiments, incremental machine learning techniques may be incorporated to add new machine learning-based classifiers to a system or update existing machine learning-based classifiers.
1. A system, comprising:
a radio; and
a signal classifier, the signal classifier configured to detect and classify a signal of a wireless environment monitored by the radio, the signal classifier comprising:
an energy-based classifier configured to analyze an entire set of measurements and generate a first signal classification result;
a cyclostationary-based classifier configured to analyze a subset of the entire set of measurements and generate a second signal classification result; and
a classification merger module configured to merge the first signal classification result and the second signal classification result.
2. The system of claim 1 , wherein the cyclostationary-based classifier comprises:
a data reducer configured to define a subset of the entire set of measurements responsive to a data reduction algorithm;
a data reducer configured to define a subset of the entire set of measurements responsive to a data reduction algorithm; and
a feature calculator configured to identify cyclostationary features of measurements received from the data reducer.
3. The system of claim 1 , wherein the signal classifier comprises a data store having stored thereon signal class models, and further wherein the signal class models are indicative of cyclostationary features of one or more predefined wireless signal protocols.
4. The system of claim 3 wherein at least one of the signal class models is a one-class support vector machine model.
5. The system of claim 1 , wherein the signal classifier comprises a signal/noise classifier.
6. The system of claim 5 , wherein the signal/noise classifier comprises a learned noise/signal classifier, wherein the learned noise/signal classifier was trained using supervised learning techniques to distinguish between noise and signals.
7. The system of claim 1 , wherein the signal classifier comprises a known signal classifier, the known signal classifier configured to cluster signals of the wireless environment for a known signal class, wherein at least some of the clustered signals were present in parts of a frequency spectrum other than a first part of the frequency spectrum used to train a known signal classifier trained to classify the known signal class.
8. The system of claim 1 , wherein the signal classifier is configured to receive radio-frequency measurements for the wireless environment at which the radio is deployed.