System and method for energy efficient sensors with compression, artificial intelligence, and security
Devices and methods for processing detected signals at a detector using a processor are provided. The system involves (i) a data compressor that implements an algorithm for converting a set of data into a compressed set of data, (ii) a machine learning (ML) module coupled to the data compressor, the ML module transforming the compressed set of data into a vector and filtering the vector, (iii) a data encryptor coupled to the ML module that encrypts the filtered vector, and (iv) an integrity protection module coupled to the ML module, wherein the integrity protection module protects the integrity of the filtered vector.
1. A detector, comprising:
a data compressor, wherein the data compressor implements an algorithm for converting a set of data into a compressed set of data;
a machine learning module coupled to the data compressor, wherein the machine learning module transforms the compressed set of data into a vector and filters the vector;
a data encryptor coupled to the machine learning module, wherein the data encryptor encrypts the filtered vector;
an integrity protection module coupled to the machine learning module, wherein the integrity protection module protects integrity of the filtered vector, where the integrity protection module is used in parallel with the data encryptor; and
a transmitter configured to transmit the encrypted filtered vector to a remote location.
2. The detector of claim 1 , wherein the data compressor implements compressive sensing.
3. The detector of claim 1 , wherein the machine learning module comprises an inference module.
4. The detector of claim 1 , wherein the data encryptor encrypts the filtered vector using encompression.
5. The detector of claim 1 , further comprising an analog-to-digital converter coupled to the data compressor.
6. The detector of claim 1 , wherein the machine learning module comprises a feature extractor and a classifier.
7. The detector of claim 6 , wherein the classifier is linear.
8. The detector of claim 6 , wherein the classifier is non-linear.
9. The detector of claim 6 , wherein the classifier comprises a neural network.
10. The detector of claim 6 , wherein the classifier is a support vector machine.
11. The detector of claim 1 , wherein the integrity protection module comprises a hashing module.
12. The detector of claim 11 , wherein the hashing module comprises at least one of SHA-2, SHA-3, and SHA-4.
13. The detector of claim 11 , wherein the hashing module comprises a cryptographic hashing module.
14. A method for processing detected signals at a detector using one or more processors, the method comprising:
converting a set of data into a compressed set of data using a data compressor controlled via the processors;
transforming the compressed set of data into a vector and filtering the vector using a machine learning module controlled via the processors;
encrypting the filtered vector using an encryptor controlled via the processors; and
protecting integrity of the filtered vector using an integrity protection module controlled via the processors, where protecting integrity of the filtered vector is performed in parallel with encrypting the filtered vector.
15. The method of claim 14 , wherein the set of data is converted into a compressed set of data using compressive sensing.
16. The method of claim 14 , wherein the machine learning module comprises an inference module.
17. The detector of claim 14 , wherein the encrypting of the filtered vector is accomplished using encompression.
18. The method of claim 14 , further comprising converting a set of analog data into the set of data using an analog-to-digital converter.
19. The method of claim 14 , wherein the transforming of the compressed set of data into a vector is accomplished using a feature extractor, and filtering the vector is accomplished using a classifier.
20. The method of claim 19 , wherein the classifier is linear.
21. The method of claim 19 , wherein the classifier is non-linear.
22. The method of claim 19 , wherein the classifier comprises a neural network.
23. The method of claim 19 , wherein the classifier comprises a support vector machine.
24. The method of claim 14 , wherein the integrity is protection module comprises a hashing module.
25. The method of claim 24 , wherein the hashing module comprises a cryptographic hashing module.
26. A device, comprising:
a data compressor, wherein the data compressor converts a set of data into a compressed set of data;
a machine learning module coupled to the data compressor, wherein the machine learning module transforms the compressed set of data into a vector and filters the vector;
a data encryptor coupled to the machine learning module, wherein the data encryptor encrypts the filtered vector;
an integrity protection module coupled to the machine learning module, wherein the integrity protection module protects integrity of the filtered vector, where the integrity protection module is used in parallel with the data encryptor; and
a transmitter configured to transmit the encrypted filtered vector to a remote location.