Webinar: RF Record, playback, real-time processing and Machine Learning of Radar & EW Signals

Analyzing complex electromagnetic environments, especially in EMSO applications, are extremely challenging to analyze with traditional test equipment. Spectrum analyzers and high-speed oscilloscopes have limited memory depth, making it difficult to identify emitters effectively. Vector network analyzers offer better sensitivity, but they typically lack the necessary IQ bandwidth to thoroughy assess electronic warfare (EW) system performance.

In environments with many potential transmitters and emitter interactions, how can EW engineers effectively analyze the performance of their systems?

A further challenge is that operational receivers often operate at a higher level of abstraction, potentially overlooking critical signals and interactions that a more detailed analysis could reveal.

In this webinar, we explore the capabilities of IRAPS™ and the FPGA Enhanced Development System (FEDS) for advanced signal classification using Mathworks’™ Deep Learning Processor (DLP). We demonstrate how the DLP, intellgrated into the FEDS design, showcases the potential of FPGA technology. By capturing training frames from real-time IQ streams, we conduct offline training on a 28-layer Convolutional Neural Network. The trained model will then be deployed for real-time signal classification, allowing us to compare the expected results with the classifications generated by the DLP in real-time.

In this webinar, you will learn more about:

• Motivation for RF recording and key applications
• The R&S IRAPS™ and R&D FEDS systems
• Machine learning and signal classification using Mathlab™