Enhancing Radar Sensitivity with Deep Learning
A recent study published by FVD in collaboration with Thales showed that deep learning techniques can enhance radar sensitivity to such an extent that individual persons can be identified in a group of fifty based on their walk past a radar. This is a significant step forward in the development of a radar security system that can quickly recognize individuals it has seen before. However, there is still much ground to cover before such techniques can be introduced in Thales products.
Identifying Drones with Deep Learning
Another research project currently underway is the use of deep learning to identify drones flying on a wind farm. Traditional radar techniques struggle to distinguish between small drones and wind turbines due to the latter’s overpowering signal. However, deep learning can disentangle the signals and not only detect the drone but also determine its characteristics, such as the number of rotors, from the micro-Doppler signature. This could be very useful in operating radars around offshore wind farms.
Optimizing Radar Operation with Deep Learning
Dynamic adaptation of waveform, control, and processing using deep learning is a key focal point for Harmanny and his team. By anticipating measurements more quickly, radar operation can be optimized for maximum efficiency. Additionally, the power of phased-array radars is being leveraged to provide greater flexibility in beam steering through individual control of transmit and receive elements. The team is exploring how to enable and enhance this flexibility while making the best use of it.
Leveraging Quantum Computing for Radar Processing
Quantum computing has the potential to revolutionize many fields, and radar processing is no exception. Harmanny is interested in exploring how quantum computers can be used to run new algorithms in radar processing. The enormous computing power of quantum computers could potentially be leveraged to process radar data more quickly and accurately than ever before.
Conclusion
As radar technology continues to advance, the use of deep learning, phased-array radars, and quantum computing offers exciting opportunities to optimize radar operation and enhance its capabilities. Harmanny and his team are at the forefront of this research, and their work could lead to the development of innovative radar systems that could transform many industries.