Experimental Investigation Of Quantum Algorithms On Nitrogen-vacancy Center In Diamond

With the development of quantum information science,quantum mechanics provides a new approach to improve the computing power of computers.Quantum algorithms based on the principles of quantum mechanics can enable quantum computers to achieve exponential speed-up compared with classical computers when solving specific problems,such as large number factoring,which has aroused widespread interest.The design and implementation of quantum algorithms are expected to utilize the advantages of quantum computing and solve specific problems in machine learning,optimization and other fields.Among the physical systems suitable for implementing quantum algorithms,nitrogen-vacancy center(NV center)in diamond has attracted much attention due to the ultra-long coherence time at room temperature and the quantum control methods.This thesis focuses on the experimental research of quantum algorithms with the hybrid spin system in NV center.First,the theory of quantum algorithms,especially quantum machine learning algorithms,is presented.Then we introduce the basic properties of the electron spin-nuclear spin system in NV center,the methods to control the spins and the experiment platform.The main part of this thesis includes the following work:1.Utilizing the properties of the electron spin and nitrogen nuclear spin,we controlled the quantum correlation between the spins with flexible quantum control methods and realized a spin quantum heat engine.The importance of quantum correlation between two qubits in the quantum heat engine was experimentally investigated,and the experimental results verified the optimized work extraction inequality proposed in this work.Finally,this work experimentally demonstrated the relationship between the work extraction in quantum heat engine and quantum steering,a special kind of quantum correlation.This work provided a new idea for the intersection of quantum thermodynamics and quantum information.2.A new type of quantum principal component analysis algorithm was developed and the algorithm was experimentally demonstrated.Combining the idea of resonance excitation,we proposed a resonant quantum principal component analysis algorithm,and reduced the number of ancillary qubits to one,compared with the original algorithm.In the experiment,we used the electron spin and nuclear spins proximate to the NV center to demonstrate the analysis and extraction of quantum principal component.Finally,after utilizing dynamical decoupling to suppress the noise,The experiment achieved 90%extraction accuracy and 86%extraction efficiency.3.Using the NV center system,a quantum anomaly detection algorithm was implemented.We demonstrated the quantum anomaly detection of audio samples by using quantum processors composed of solid state spins in diamonds.By learning from normal samples,the quantum processor was utilized to calculate the overlap between the normal samples and new sample in parallel,and detected the anomaly with logarithmic resources.The minimum error rate to detect the anomalies was 15.4%.The experimental results demonstrated the potential of quantum anomaly detection in machine learning tasks.