以核磁共振方式實作量子電腦與其所需之脈衝程式產生器

Abstract

The physical implementation of quantum computers using a nucleus magnetic resonance (NMR) method is mature nowadays. However, not many people can practically exercise the theory of the implemental techniques. Therefore, people couldn’t easily apply the algorithms on quantum computing for mass production. Based on the accumulated experience on implementing quantum computers, we are able to perform computation operations by commanding a NMR machine to give certain pulses on atoms to make changes in spinning state that implicitly contains some quantum information. To achieve hierarchical programming, a NMR pulse program generator is designed and implemented to acts as a transformer for high-level and low-level languages used in quantum computers. This device generates pulse programs for the needed quantum circuits and carries out algorithms easily without the implementation technology. By simply applying quantum gate sequence (e.g. CN-gate) as input, pulse program containing instructions and parameters for the NMR spectrometer could be generated without too much overhead and efforts. In the past, NMR experiments often involve lengthy and sophisticated pulse program that a need for an automatic generator is highly anticipated. In other words, the proposed NMR pulse program generator is somehow like a friendly communicator that lies between human and the machine, it allows efficient pulse program generation and avoid artificial errors. The proposed generator is based on the liquid-state NMR quantum computer, and mainly generates executing instructions for those using carbon and hydrogen as the core material. Having the material’s states and parameters from different experiments as input, it expands to a customized tool that is more general with less restrictions. So far through the experiments for two-bit and seven-bit quantum computers with different materials, it makes the fundamental logics on quantum computing. When working with a high-level language for quantum computer, programming might be as convenient as in classical computing.