PRIMARY SIGNAL DETECTION USING DISTRIBUTED MACHINE LEARNING VIA USER EQUIPMENT DEVICES
Methods and systems for primary signal detection using distributed machine learning are disclosed. In an example method, each user equipment (UE) device of a plurality of UE devices is caused to determine a machine learning model configured to detect an anomaly in the UE device's RF environment. The method further includes receiving, by a controller and from each UE device of the plurality of UE devices, anomaly data indicative of an anomaly detected by the UE device via its machine learning model. The method further includes determining, by the controller and based on the anomaly data from the plurality of UE devices, that a primary signal is present in one or more of the RF environments associated with the plurality of UE devices.
1 . A method comprising:
causing each user equipment (UE) device of a plurality of UE devices to determine a machine learning model configured to detect an anomaly in the UE device's RF environment;
receiving, by a controller and from each UE device of the plurality of UE devices, anomaly data indicative of an anomaly detected by the UE device via its machine learning model; and
determining, by the controller and based on the anomaly data from the plurality of UE devices, that a primary signal is present in one or more of the RF environments associated with the plurality of UE devices.
2 . The method of claim 1 , further comprising:
causing the plurality of UE devices to switch their respective operating frequency spectrums to a frequency spectrum that will not cause interference with the primary signal.
3 . The method of claim 1 , wherein the respective machine learning models determined by the plurality of UE devices are unsupervised machine learning models.
4 . The method of claim 1 , wherein the anomaly data received from the plurality of UE devices is associated with a predefined area, and the primary signal is determined to be associated with that predefined area.
5 . The method of claim 1 , wherein the anomaly data received from each UE device of the plurality of UE devices indicates a probability metric associated with the detected anomaly, and the determining that the primary signal is present is further based on the indicated probability metric.
6 . The method of claim 1 , wherein the primary signal is associated with a radar station.
7 . The method of claim 1 , wherein the plurality of UE devices operate in one or more of a 5G/NR cellular network or a 4G/LTE cellular network.
8 . The method of claim 1 , wherein the controller is integrated with a base station configured to wirelessly communicate with the plurality of UE devices.
9 . The method of claim 1 , wherein the controller is configured to determine a machine learning model based on anomaly data reported by UE devices, and the machine learning model is used by the controller to determine that the primary signal is present in the one or more of the RF environments associated with the plurality of UE devices.
10 . A system comprising:
a base station configured to communicate with a plurality of user equipment (UE) devices and further receive, from each UE device of the plurality of UE devices, anomaly data indicative of an anomaly detected by the UE device via a machine learning model determined by the UE device for anomaly detection in the UE device's RF environment; and
a controller configured to communicate with the base station, wherein the controller is further configured to:
receive, from the base station, the anomaly data indicative of the respective anomalies detected by the plurality of UE devices; and
determine, based on the anomaly data, that a primary signal is present in one or more of the RF environments associated with the plurality of UE devices.
11 . The system of claim 10 , wherein at least one of the base station or the controller is further configured to cause the plurality of UE devices to switch their respective operating frequency spectrum to a frequency spectrum that will not cause interference with the primary signal.
12 . The system of claim 10 , wherein the base station is configured for operation in one or more of a 5G/NR cellular network or a 4G/LTE cellular network.
13 . The system of claim 10 , wherein the anomaly data is associated with a predefined area, and the primary signal is determined to be associated with that predefined area.
14 . The system of claim 10 , wherein the anomaly data indicates a probability metric associated with each of the detected anomalies, and the determination by the controller that the primary signal is present is further based on the indicated probability metrics.
15 . A non-transitory computer-readable medium storing instructions that, when executed, effectuate operations comprising:
causing each user equipment (UE) device of a plurality of UE devices to determine a machine learning model configured to detect an anomaly in the UE device's RF environment;
receiving, by a controller and from each UE device of the plurality of UE devices, anomaly data indicative of an anomaly detected by the UE device via its machine learning model; and
determining, by the controller and based on the anomaly data from the plurality of UE devices, that a primary signal is present in one or more of the RF environments associated with the plurality of UE devices.
16 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:
causing the plurality of UE devices to switch their respective operating frequency spectrums to a frequency spectrum that will not cause interference with the primary signal.
17 . The non-transitory computer-readable medium of claim 15 , wherein the anomaly data received from the plurality of UE devices is associated with a predefined area, and the primary signal is determined to be associated with that predefined area.
18 . The non-transitory computer-readable medium of claim 15 , wherein the anomaly data received from each UE device of the plurality of UE devices indicates a probability metric associated with the detected anomaly, and the determining that the primary signal is present is further based on the indicated probability metric.
19 . The non-transitory computer-readable medium of claim 15 , wherein the controller is integrated into a base station configured to wirelessly communicate with the plurality of UE devices.
20 . The non-transitory computer-readable medium of claim 15 , wherein the controller is configured to determine a machine learning model based on anomaly data reported by UE devices, and the machine learning model is used by the controller to determine that the primary signal is present in the one or more of the RF environments associated with the plurality of UE devices.