Network Technologies, ASA(ALT), Phase I
Adaptive Filtering Techniques for Low-Cost RF Emitters
Objective
Develop an emitter system-agnostic solution for threat representative waveform generation and validation.
Description
Deployed T&E threat representative systems are currently not capable of validating that the emitted signal is representative of the source threat waveform. Low-cost open-air RF emitters using commercially available components creates perturbations as the signal moves through the various RF components causing a delta between the theoretical and the open-air signal. Adaptive filtering will demonstrate that any low-cost RF system (RF band agnostic) can be trained through an iterative process to produce the theoretical or known waveform.
Phase I
This topic is only accepting Phase I proposals for a cost up to $250,000 for a 6-month period of performance. Government and Industry will work collaboratively to refine topic objectives to reach a feasible commercial product.
The work will entail detailed studies on M&S simulation environment integration, adaptive filtering, AI/ML signal conditioning, and system integration requirements. By the end of Phase I, the deliverable will be a well-documented research report that outlines the steps needed to move forward with prototype development, including risk assessments, technical challenges, and a proposed plan for Phase II.
Phase II
A prototype design will be completed for production. A prototype will be delivered for demonstration. The prototype will be tested to validate the design against the thresholds identified.
Phase III
In the private sector, companies are increasingly applying artificial intelligence (AI) and machine learning (ML) to solve complex signal processing challenges, particularly in telecommunications [e.g., cognitive 5G/6G networks, distributed microcell technology, swarm drone cognitive communications systems, etc.].
AI/ML is used to optimize signal integrity, detect anomalies, and adapt in real-time to fluctuating environments. These approaches could be applied to the problem of validating RF emissions by using AI/ML algorithms to dynamically filter and correct waveform distortions, ensuring that live signals accurately reflect digital models.
Submission Information
For more information, and to submit your full proposal package, visit the DSIP Portal.
SBIR|STTR Help Desk: usarmy.sbirsttr@army.mil
References:
- Phartiyal, Deepmala, and Meenakshi Rawat. “LSTM-deep neural networks based predistortion linearizer for high power amplifiers.” 2019 National Conference on Communications (NCC). IEEE, 2019. LSTM-Deep Neural Networks based Predistortion Linearizer for High Power Amplifiers | IEEE Conference Publication | IEEE Xplore
- López Bueno, David. “Machine learning techniques for adaptive polynomial and neural network digital predistorters.” (2023). Machine learning techniques for adaptive polynomial and neural network digital predistorters (upc.edu)
- KEYWORDS: Machine learning; adaptive filtering; Artificial Intelligence; RF emitters; signal conditioning
Objective
Develop an emitter system-agnostic solution for threat representative waveform generation and validation.
Description
Deployed T&E threat representative systems are currently not capable of validating that the emitted signal is representative of the source threat waveform. Low-cost open-air RF emitters using commercially available components creates perturbations as the signal moves through the various RF components causing a delta between the theoretical and the open-air signal. Adaptive filtering will demonstrate that any low-cost RF system (RF band agnostic) can be trained through an iterative process to produce the theoretical or known waveform.
Phase I
This topic is only accepting Phase I proposals for a cost up to $250,000 for a 6-month period of performance. Government and Industry will work collaboratively to refine topic objectives to reach a feasible commercial product.
The work will entail detailed studies on M&S simulation environment integration, adaptive filtering, AI/ML signal conditioning, and system integration requirements. By the end of Phase I, the deliverable will be a well-documented research report that outlines the steps needed to move forward with prototype development, including risk assessments, technical challenges, and a proposed plan for Phase II.
Phase II
A prototype design will be completed for production. A prototype will be delivered for demonstration. The prototype will be tested to validate the design against the thresholds identified.
Phase III
In the private sector, companies are increasingly applying artificial intelligence (AI) and machine learning (ML) to solve complex signal processing challenges, particularly in telecommunications [e.g., cognitive 5G/6G networks, distributed microcell technology, swarm drone cognitive communications systems, etc.].
AI/ML is used to optimize signal integrity, detect anomalies, and adapt in real-time to fluctuating environments. These approaches could be applied to the problem of validating RF emissions by using AI/ML algorithms to dynamically filter and correct waveform distortions, ensuring that live signals accurately reflect digital models.
Submission Information
For more information, and to submit your full proposal package, visit the DSIP Portal.
SBIR|STTR Help Desk: usarmy.sbirsttr@army.mil
References:
- Phartiyal, Deepmala, and Meenakshi Rawat. “LSTM-deep neural networks based predistortion linearizer for high power amplifiers.” 2019 National Conference on Communications (NCC). IEEE, 2019. LSTM-Deep Neural Networks based Predistortion Linearizer for High Power Amplifiers | IEEE Conference Publication | IEEE Xplore
- López Bueno, David. “Machine learning techniques for adaptive polynomial and neural network digital predistorters.” (2023). Machine learning techniques for adaptive polynomial and neural network digital predistorters (upc.edu)
- KEYWORDS: Machine learning; adaptive filtering; Artificial Intelligence; RF emitters; signal conditioning