Detecting, tracking, and transmitting to identified objects using models in a modular system
A modular, radio frequency (“RF”) system includes one or more directional antennas and is configured with both hardware and software components to enable the RF system to monitor (e.g., detect or track signals or objects) and/or interact with (e.g., track signals or objects, or transmit signals) objects in particular directions. The RF system includes one or more machine learning models to determine, based on received signals, one or more signals to transmit.
1 . A computer-implemented method comprising, by one or more hardware processors executing program instructions:
accessing or receiving, by a first RF system, detection data associated with a first object, wherein the detection data is collected or generated by one or more: RF systems, sensors, and devices configured to detect RF signals or objects;
collecting, by using one or more antennas of the first RF system, RF signal data associated with the first object;
based at least in part on the detection data and the RF signal data, identifying a first set of RF signals associated with the first object;
identifying, by application of a machine learning model, a type of object associated with the first object; and
based at least in part on the type of object, generating, by the first RF system, and causing transmission, by using the one or more antennas of the first RF system, of a second set of RF signals different from the first set of RF signals, wherein the first RF system operates in coordination with at least one other RF system, each system covering a different sector surrounding a protected area.
2 . The computer-implemented method of claim 1 , wherein the detection data includes a portion of the first set of RF signals.
3 . The computer-implemented method of claim 1 , wherein the detection data indicates a physical location associated with the first object.
4 . The computer-implemented method of claim 1 , wherein the causing transmission of the second set of RF signals includes transmission of the second set of RF signals in a direction associated with the first object.
5 . The computer-implemented method of claim 1 , wherein the machine learning model comprises:
inputting, into the machine learning model, the first set of RF signals such that the machine learning model outputs the type of object associated with the first object.
6 . The computer-implemented method of claim 1 , wherein the protected area is omitted from coverage, and wherein the protected area includes one or more buildings, electromagnetically sensitive equipment, or both.
7 . The computer-implemented method of claim 1 , wherein the first RF system reduces power supplied to a transmitting antenna when the object is located near electromagnetically sensitive equipment or the protected area.
8 . The computer-implemented method of claim 1 , further comprising:
prior to causing transmission of the second set of RF signals, filtering out, by the first RF system, one or more frequencies from the second set of RF signals RF signals, wherein the one or more frequencies are identified to reduce or avoid interference with electromagnetically sensitive equipment or structures located in or associated with the protected area.
9 . A computer-implemented method comprising, by one or more hardware processors executing program instructions:
based on a first set of RF signals associated with a first object and an application of a machine learning model that has been trained to identify a type of object associated with the first set of RF signals and operating on an RF system, generating and causing transmission of, using a first antenna corresponding to the RF system, a second set of RF signals, wherein causing transmission includes supplying power to the first antenna, wherein the RF system operates in coordination with at least one other RF system, each system covering a different sector surrounding a protected area;
determining, by the RF system, that the first object is moving out of an area associated with the first antenna and into an area associated with a second antenna; and
based on the determination:
reduce the power supplied to the first antenna, wherein the power supplied to the first antenna is reduced at a first rate;
increase the power supplied to the second antenna, wherein the power supplied to the second antenna is increased at a second rate; and
cause transmission, using the second antenna, of the second set of RF signals.
10 . The computer-implemented method of claim 9 , wherein the transmission of the second set of RF signals by the first antenna or the second antenna includes transmission of the second set of RF signals in a direction associated with the first object.
11 . The computer-implemented method of claim 9 , wherein the determination that the first object is moving out of an associated with the first antenna and into an area associated with a second antenna is performed using at least a direction finder.
12 . The computer-implemented method of claim 9 , wherein the second antenna corresponds to a second RF system.
13 . The computer-implemented method of claim 9 , wherein adjusting the power supplied to the first antenna comprises ceasing transmission of the second set of RF signals.
14 . The computer-implemented method of claim 9 , wherein adjusting the power supplied to the second antenna comprises initiating transmission of the second set of RF signals.
15 . The computer-implemented method of claim 9 , wherein the second antenna corresponds to the first RF system, and wherein total power used at any moment in time by the first antenna and the second antenna remains constant.
16 . The computer-implemented method of claim 9 , wherein the second antenna corresponds to a second RF system, and wherein total power used at any moment in time by the first antenna and the second antenna remains constant.
17 . The computer-implemented method of claim 9 , wherein the first rate and the second rate are the same such that power is supplied to both the first antenna and second antenna at the same time for a period of time proportional to the first rate.
18 . The computer-implemented method of claim 9 , wherein identifying, by application of the machine learning model, the type of object associated with the first object further comprises identifying the type of object associated with the first object.
19 . The computer-implemented method of claim 9 , wherein the area associated with the second antenna includes the protected area, further comprising:
filtering out, by the RF system, one or more frequencies from transmission by the second antenna, wherein the one or more frequencies are identified based on a likelihood of interference with electromagnetically sensitive equipment or the protected area.
20 . An RF system comprising:
one or more antennas;
a first RF module coupled to the one or more antennas, wherein the first RF module comprises a power amplifier;
a processing module electrically connected to the first RF module, wherein the processing module comprises;
a first one or more graphical processing units (GPUs);
a computer readable storage medium comprising program instructions and a machine learning model, wherein the machine learning model has been trained to identify a type of object associated with a first set of RF signals; and
one or more first processors that are configured to execute the program instructions to cause the RF system to:
receive a data packet comprising software updates, wherein the software updates include one or more of: firmware updates corresponding to hardware components of the RF system, updates corresponding to software components utilized by the RF system, and machine learning model updates that include object detections identified by a second RF module associated with a different system, wherein the software updates include sector coverage updates that identify one or more protected areas, wherein the RF system operates in coordination with at least one other RF system, each system covering a different sector surrounding the protected area; and
install the software updates to update functionality of the RF system, wherein the RF system is configured to generate and cause transmission of one or more RF signals.
21 . The RF system of claim 20 , wherein the software updates further include:
frequency filtering rules so that the RF system applies the frequency filtering rules during transmission to minimize potential interference within the protected area or with electromagnetically sensitive equipment;
power management rules that reduce power supplied to transmitting antennas when an object associated with the first set of RF signals is detected nearby electromagnetically sensitive equipment or the protected area; or
data corresponding to other RF systems comprising coverage maps, install locations, coverage areas, and configuration information for controlling activation or deactivation of antennas of nearby RF systems.