摘要
By using ZCU 104 Evaluation Board with the Vitis AI Integrated Development Environment, we have successfully deployed machine learning-based quantum state tomography onto a FPGA device, with a slight reduction in the output average fidelity, i.e., from 0.99 to 0.98. But the signal processing speed when reconstructing quantum state from tomography is demonstrated to be 10 times higher, taking 2.94 ms, instead 38 ms previously. With the flexibility and parallel processing abilities, our FPGA-based QST offers a highly efficient and precise tool toward real-time diagnostics of quantum states. In addition to application to Gaussian states, as illustrated here, this technology paves the way to dealing with more general quantum states, including non-Gaussian states and multi-partite quantum states at high throughput speeds. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.