SYSTEM AND METHOD OF PROCESSING A RADIO FREQUENCY SIGNAL WITH A NEURAL NETWORK

• US 20190319658A1
• Filed: 04/11/2018
• Published: 10/17/2019
• Est. Priority Date: 04/11/2018
• Status: Active Grant

• Current Assignee: BOOZ Allen Hamilton Holding Corporation
• Original Assignee: BOOZ Allen Hamilton Holding Corporation
• Inventors: CALABRO, Michael, MORTAZAVI, Mehrnaz, LEONARD, Wade, ROHDE, Zachary, SAUNDERS, Alex
• Granted Patent: US 10,944,440 B2
• Time in Patent Office: Days
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: H04B 1/0003 Software-defined radio [SDR] systems, i.e. systems wherein components typically implemented in hardware, e.g. filters or modulators/demodulators, are implented using software, e.g. by involving an AD or DA conversion stage such that at least part of the signal processing is performed in the digital domain digital baseband systems H04L25/00; digital modulation/demodulation H04L27/00; CDMA H04B1/707; TDMA H04B7/2643; image transmission H04N5/00
• CPC Class Codes: H04B 1/04 Circuits
• CPC Class Codes: H04B 1/16 Circuits

Abstract

A method of processing a radio frequency signal includes: receiving the radio frequency signal at an antenna of a receiver device; processing, by a radio frequency front-end device, the radio frequency signal; converting, by an analog-to-digital converter, the analog signal to a digital signal; receiving, by a neural network, the digital signal; and processing, by the neural network, the digital signal to produce an output.

Claims

1. A method of processing a radio frequency signal, comprising:

receiving the radio frequency signal at an antenna of a receiver device;

processing, by a radio frequency front-end device, the radio frequency signal;

converting, by an analog-to-digital converter, the radio frequency signal to a digital signal;

receiving, by a neural network trained to operate on a streaming radio platform, the digital signal; and

processing, by the neural network, the digital signal to produce an output.

2. The method of claim 1, wherein the processing is the classifying of a modulation scheme of the digital signal, and the digital signal is processed based on the determined classification of the modulation scheme.

3. The method of claim 1, wherein the radio frequency signal received at the antenna was generated by a neural network based transmitter device.

4. The method of claim 2, wherein the received radio frequency signal is any one type among two or more types of radio frequency signals, and the classified modulation scheme of the digital signal is unique amongst the modulation schemes of the two or more radio frequency signals.

5. The method of claim 4, wherein processing the digital signal to produce an output comprises processing the digital signal based on the determined classification of a modulation scheme of the digital signal.

6. The method of claim 1, wherein the neural network is two or more neural networks connected in series or parallel.

7. The method of claim 1, wherein the neural network has been trained by training vectors.

8. The method of claim 1, wherein the neural network has been trained by a plurality of training vectors that include hardware impairments and distortions caused by a transmission channel between the receiver device and a transmitter device.

9. The method of claim 8, wherein the training vectors include a range of time, phase and frequency errors, and a range of additive white Gaussian noise levels corresponding to a range of signal-to-noise ratios.

10. The method of claim 1, comprising:

performing, by the neural network, synchronization, channel equalization, demodulation, decoding, or other functions on the digital signal as necessary to reproduce information bits that were transmitted.

11. The method of claim 1, comprising:

performing, by the neural network, synchronization, channel equalization, demodulation, decoding, and other functions on the digital signal as necessary to reproduce information bits that were transmitted.

12. The method of claim 1, wherein the neural network of the receiver device and a neural network of a transmitter device are jointly trained for optimal performance in a particular transmission environment.

23. The method of claim 1, wherein the neural network is trained subject to SWaP (size, weight, and power) constraints of a receiver device of the streaming platform.

13. A method of constructing and training a neural network that is configured to process a signal, the method comprising:

identifying a processing function to be performed on the signal by the neural network;

determining, an optimum format for training vectors to be applied to the neural network;

determining a number of layers of the neural network, number of nodes per layer, the node connectivity structure, and initial weights of the neural network; and

inputting, into the neural network, a plurality of formatted training vectors in order to train the neural network to perform the processing function on the signal.

14. The method of claim 13, comprising:

recording trained weights of the neural network after completion of training of the neural network;

evaluating performance of the processing function of the neural network by using the trained weights; and

iteratively adjusting the training vector format, the number of layers of the neural network, the number of nodes per layer of the neural network, and the node connectivity structure until desired performance of the processing function is achieved.

15. The method of claim 13, wherein the processing function is classification of a modulation scheme.

16. The method of claim 13, wherein the number of layers is two.

17. The method of claim 13, wherein the optimum format of the training vector includes two-dimensional constellation points with resolution based on an amount or type of distortion in a transmission channel through which the radio frequency signal can travel.

18. A receiver device configured to process a radio frequency signal, comprising:

an antenna configured to receive the radio frequency signal;

a radio frequency front-end device configured to process the radio frequency signal;

an analog-to-digital converter configured to convert the radio frequency signal to a digital signal; and

a neural network trained to operate on a streaming radio platform and configured to receive the digital signal and to process the digital signal to produce an output.

19. The receiver device of claim 18, wherein the processing performed by the neural network includes classifying a first modulation scheme of the digital signal, and

wherein the digital signal is processed based on the determined classification of the modulation scheme.

20. The receiver device of claim 18, wherein the radio frequency signal received at the antenna was generated by a neural network based transmitter device.

21. The receiver device of claim 19, wherein the received radio frequency signal is any one type among two or more types of radio frequency signals, and the classified modulation scheme of the digital signal is unique amongst the modulation schemes of the two or more radio frequency signals.

22. The receiver device of claim 18, wherein the neural network is configured to process the digital signal based on a determined classification of a modulation scheme of the digital signal.

24. The receiver device of claim 18, wherein the neural network is trained subject to SWaP contraints of the receiver device.