Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned communication. One of the methods includes: receiving an RF signal at a signal processing system for training a machine-learning network; providing the RF signal through the machine-learning network; producing an output from the machine-learning network; measuring a distance metric between the signal processing model output and a reference model output; determining modifications to the machine-learning network to reduce the distance metric between the output and the reference model output; and in response to reducing the distance metric to a value that is less than or equal to a threshold value, determining a score of the trained machine-learning network using one or more other RF signals and one or more other corresponding reference model outputs, the score indicating an a performance metric of the trained machine-learning network to perform the desired RF function.

Abstract: One or more processors control processing of radio frequency (RF) signals using a machine-learning network. The one or more processors receive as input, to a radio communications apparatus, a first representation of an RF signal, which is processed using one or more radio stages, providing a second representation of the RF signal. Observations about, and metrics of, the second representation of the RF signal are obtained. Past observations and metrics are accessed from storage. Using the observations, metrics and past observations and metrics, parameters of a machine-learning network, which implements policies to process RF signals, are adjusted by controlling the radio stages. In response to the adjustments, actions performed by one or more controllers of the radio stages are updated. A representation of a subsequent input RF signal is processed using the radio stages that are controlled based on actions including the updated one or more actions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned communication over multi-input-multi-output (MIMO) channels.

Abstract: First information is obtained from a sensing device at a first time. The first information corresponds to a radio signal received at the device from a candidate location. The device is at a first location at the first time. Second information is obtained from the device at a second time. The second information corresponds to a radio signal received at the device from the candidate location. The device is at a second location at the second time. A system determines that a pattern is in each of the first and second information and determines relationships between the candidate location and the device at each first and second location. The system obtains inverses of the relationships and determines estimates of the received radio signals based on the information and inverses. The system measures or estimates energy emitted from the candidate location based on the estimates.

Abstract: One or more processors control processing of radio frequency (RF) signals using a machine-learning network. The one or more processors receive as input, to a radio communications apparatus, a first representation of an RF signal, which is processed using one or more radio stages, providing a second representation of the RF signal. Observations about, and metrics of, the second representation of the RF signal are obtained. Past observations and metrics are accessed from storage. Using the observations, metrics and past observations and metrics, parameters of a machine-learning network, which implements policies to process RF signals, are adjusted by controlling the radio stages. In response to the adjustments, actions performed by one or more controllers of the radio stages are updated. A representation of a subsequent input RF signal is processed using the radio stages that are controlled based on actions including the updated one or more actions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for correcting distortion of radio signals A transmit radio signal corresponding to an output of a transmitting radio signal processing system is obtained. A pre-distorted radio signal is then generated by processing the transmit radio signal using a nonlinear pre-distortion machine learning model. The nonlinear pre-distortion machine learning model includes model parameters and at least one nonlinear function to correct radio signal distortion or interference. A transmit output radio signal is obtained by processing the pre-distorted radio signal through the transmitting radio signal processing system. The transmit output radio signal is then transmitted to one or more radio receivers.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for placement and scheduling of radio signal processing dataflow operations. An example method provides a primitive radio signal processing computational dataflow graph that comprises nodes representing operations and directed edges representing data flow. The nodes and directed edges of the primitive radio signal processing computational dataflow graph are partitioned to produce a set of software kernels that, when executed on processing units of a target hardware platform, achieve a specific optimization objective. Runtime resource scheduling, including data placement for individual software kernels in the set of software kernels to efficiently execute operations on the processing units of the target hardware platform. The resources of the processing units in the target hardware platform are then allocated according to the defined runtime resource scheduling.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for processing communications signals using a machine-learning network are disclosed. In some implementations, pilot and data information are generated for a data signal. The data signal is generated using a modulator for orthogonal frequency-division multiplexing (OFDM) systems. The data signal is transmitted through a communications channel to obtain modified pilot and data information. The modified pilot and data information are processed using a machine-learning network. A prediction corresponding to the data signal transmitted through the communications channel is obtained from the machine-learning network. The prediction is compared to a set of ground truths and updates, based on a corresponding error term, are applied to the machine-learning network.

Abstract: First information corresponding to a radio signal received at a first sensing device from a candidate location is obtained. Second information corresponding to a radio signal received at a second sensing device from the candidate location is obtained. A first relationship between the first sensing device and the candidate location and a second relationship between the second sensing device and the candidate location are determined. A first inverse and a second inverse of respectively the first and second relationships are obtained. A first estimate of the radio signal at the first sensing device is determined from the first information and the first inverse. A second estimate of the radio signal at the second sensing device is determined from the second information and the second inverse. Energy emitted from the candidate location is measured based on the first estimate and the second estimate.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned communication over multi-input-multi-output (MIMO) channels.

Abstract: First information is obtained from a sensing device at a first time. The first information corresponds to a radio signal received at the device from a candidate location. The device is at a first location at the first time. Second information is obtained from the device at a second time. The second information corresponds to a radio signal received at the device from the candidate location. The device is at a second location at the second time. A system determines that a pattern is in each of the first and second information and determines relationships between the candidate location and the device at each first and second location. The system obtains inverses of the relationships and determines estimates of the received radio signals based on the information and inverses. The system measures or estimates energy emitted from the candidate location based on the estimates.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for placement and scheduling of radio signal processing dataflow operations. An example method provides a primitive radio signal processing computational dataflow graph that comprises nodes representing operations and directed edges representing data flow. The nodes and directed edges of the primitive radio signal processing computational dataflow graph are partitioned to produce a set of software kernels that, when executed on processing units of a target hardware platform, achieve a specific optimization objective. Runtime resource scheduling, including data placement for individual software kernels in the set of software kernels are performed to efficiently execute operations on the processing units of the target hardware platform. The resources of the processing units in the target hardware platform are then allocated according to the defined runtime resource scheduling.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for processing communications signals using a machine-learning network are disclosed. In some implementations, pilot and data information are generated for a data signal. The data signal is generated using a modulator for orthogonal frequency-division multiplexing (OFDM) systems. The data signal is transmitted through a communications channel to obtain modified pilot and data information. The modified pilot and data information are processed using a machine-learning network. A prediction corresponding to the data signal transmitted through the communications channel is obtained from the machine-learning network. The prediction is compared to a set of ground truths and updates, based on a corresponding error term, are applied to the machine-learning network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for processing communications signals using a machine-learning network are disclosed. In some implementations, pilot and data information are generated for a data signal. The data signal is generated using a modulator for orthogonal frequency-division multiplexing (OFDM) systems. The data signal is transmitted through a communications channel to obtain modified pilot and data information. The modified pilot and data information are processed using a machine-learning network. A prediction corresponding to the data signal transmitted through the communications channel is obtained from the machine-learning network. The prediction is compared to a set of ground truths and updates, based on a corresponding error term, are applied to the machine-learning network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned identification of radio frequency (RF) signals.

Abstract: First information corresponding to a radio signal received at a first sensing device from a candidate location is obtained. Second information corresponding to a radio signal received at a second sensing device from the candidate location is obtained. A first relationship between the first sensing device and the candidate location and a second relationship between the second sensing device and the candidate location are determined. A first inverse and a second inverse of respectively the first and second relationships are obtained. A first estimate of the radio signal at the first sensing device is determined from the first information and the first inverse. A second estimate of the radio signal at the second sensing device is determined from the second information and the second inverse. Energy emitted from the candidate location is measured based on the first estimate and the second estimate.

Abstract: One or more processors control processing of radio frequency (RF) signals using a machine-learning network. The one or more processors receive as input, to a radio communications apparatus, a first representation of an RF signal, which is processed using one or more radio stages, providing a second representation of the RF signal. Observations about, and metrics of, the second representation of the RF signal are obtained. Past observations and metrics are accessed from storage. Using the observations, metrics and past observations and metrics, parameters of a machine-learning network, which implements policies to process RF signals, are adjusted by controlling the radio stages. In response to the adjustments, actions performed by one or more controllers of the radio stages are updated. A representation of a subsequent input RF signal is processed using the radio stages that are controlled based on actions including the updated one or more actions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned compact representations of radio frequency (RF) signals. One of the methods includes: determining a first RF signal to be compressed; using an encoder machine-learning network to process the first RF signal and generate a compressed signal; calculating a measure of compression in the compressed signal; using a decoder machine-learning network to process the compressed signal and generate a second RF signal that represents a reconstruction of the first RF signal; calculating a measure of distance between the second RF signal and the first RF signal; and updating at least one of the encoder machine-learning network or the decoder machine-learning network based on (i) the measure of distance between the second RF signal and the first RF signal, and (ii) the measure of compression in the compressed signal.

Abstract: Methods and systems including computer programs encoded on computer storage media, for training and deploying machine-learned communication over RF channels.

Abstract: Methods and systems including computer programs encoded on computer storage media, for training and deploying machine-learned communication over radio frequency (RF) channels. One of the methods includes: determining first information; generating a first RF signal by processing using an encoder machine-learning network; determining a second RF signal that represents the first RF signal altered by transmission through a communication channel; determining a first property of the first signal or the second RF signal; calculating a first measure of distance between a target value of the first property and an actual value of the first or second RF signal; generating second information as a reconstruction of the first information using a decoder machine-learning network; calculating a second measure of distance between the first information and the second information; and updating at least one of the encoder machine-learning network or the decoder machine-learning network based on the first and second measures.