Quantum Metrology Against Environenment Niose

Quantum metrology promises us with better measurement precision with limited resources,such the numberof probes and the interference time.When there is no quantum effect such quantum entanglement were utilized,probes interfering the parameter channel for timewould give a parameter estimation with imprecision of the order 1/?at the best circumstances.This is the so-called standard quantum limit.Quantum effect such quantum entanglement can bring the parameter imprecision down to 1/with same resources,which is called the Heisenberg limit.Environment noise can detriment the Heisenberg limit and render it to standard quantum limit.Moreover the requirement of great amount of experiment data and unrealistic initial state and final measurement schemes would also limit the realization and application of the quantum metrology.Quantum error correction has been developed for suppressing Markovian noise and restoring Heisenberg limit.We investigate how can one build quantum error correction code fighting uncorrelated Markovian noise for parallel quantum metrology without ancilla.We have shown that parallel quantum metrology schemes are superior to se-quential quantum metrology schemes in terms of utilizing quantum error correction to eradicate Markovian noise without ancilla.Adaptive-feedback quantum metrology schemes designed by machine learning has been shown to be able to surpass the standard quantum limit and has great resilience against noise such qubit loss and phase fluctuation.They also provide single-shot es-timation requiring limited amount of experiment data and accommodate for different input states and final measurement schemes.We propose an feedback ansatz to simplify the designing complexity of the feedbacks and thus bring the computation complexity of designing an adaptive feedback quantum metrology scheme forprobes from~7to~4.We can design adaptive-feedback quantum metrology schemes for much big-ger systems in a much shorter time.We designed adaptive-feedback quantum metrol-ogy schemes for up to 207 probes compared to previous schemes for 100 probes.The schemes we designed offers the almost the same parameter estimation precision and good resilience against noise.Further,they also have great flexibility.For instance the207-probe feedback scheme designed can be applied to up to 220 probes and produce parameter estimation with almost the same precision scaling.We designed adaptive-feedback schemes for both the sine input states and spin-squeezed states.When the input state is a spin-squeezed state with its squeezing time integrated to the machine learning procedure,the adaptive-feedback quantum metrology would give much better parameter estimation precision,better noise resilience and better flexibility.