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

Analyzing complex electromagnetic environments, especially in EMSO applications, is extremely challenging with traditional test equipment. Spectrum analyzers and high-speed oscilloscopes have limited memory depth, making it difficult to effectively identify emitters. While vector network analyzers offer better sensitivity, they typically lack the necessary IQ bandwidth to thoroughly evaluate electronic warfare (EW) system performance. A further complication is that operational receivers often function at a higher level of abstraction, potentially missing critical signals and nuanced interactions.

In environments with numerous potential transmitters and complex emitter interactions, how can EW engineers reliably assess system performance?

This webinar will explore the capabilities of R&S®IRAPS™ and the FPGA Enhanced Development System (FEDS) for advanced signal classification using Mathworks’™ Deep Learning Processor (DLP) and demonstrate how the DLP, integrated into the FEDS architecture, leverages the potential of FPGA technology. We will demonstrate the process of capturing IQ data streams, training a 28-layer convolutional neural network (CNN) offline and deploying the trained model on the FPGA for real-time classification. Finally, we’ll compare the DLP’s live classification results with expected outcomes to evaluate performance.

Expect to learn more about:

• The motivation behind RF recording and its key application areas
• The R&S IRAPS™ and R&D FEDS systems
• Machine learning and signal classification using Mathlab™