Research on a Quantum Machine Learning Approach to Mismatched Filter Design

In signal processing of radar systems, mismatched filter is required for suppression of sidelobes. In this work, we propose a quantum machine learning approach to generalized mismatched filter design, which is based on variational quantum algorithm architecture and is applicable to noisy intermediate-scale quantum (NISQ) devices. The algorithm requires sufficiently less storage resources than its classical counterparts, and can be combined with other efficient quantum algorithms to provide building blocks for quantum signal processing workflows. The capabilities of the algorithm is analyzed with random waveforms and its bottleneck is also discussed in detail, with potential enhancement strategies put forward correspondingly,