@misc{oai:repo.qst.go.jp:00079553,
 author = {Minh Tuan, Hoang and Masuyama, Yuta and Yamazaki, Yuichi and Sato, Shinichiro and Ohshima, Takeshi and Iwasaki, Takayuki and Hisamoto, Digh and Hatano, Mutsuko and Masuyama, Yuta and Yamazaki, Yuichi and Sato, Shinichiro and Ohshima, Takeshi and Hatano, Mutsuko},
 month = {Dec},
 note = {Introduction
Nitrogen-vacancy (NV) centers in diamond are promising solid-state quantum sensors. The sensor can potentially monitor the real-time magnetic field at room-temperature toward the brain-machine interface. One of the biggest challenges is to implement a highly sensitive sensor in a compact system. There are two major ways to achieve it. An analog signal processing method with a lock-in amplifier is fast to obtain the signal, but it increases the size of the sensor system. The digital signal processing method with the Fourier transform is simple, but it needs more computational resources. Reducing computational resources with the simple system is an important task for monitoring a real-time magnetic field. In this study, we demonstrate an alternative method of digital signal processing with less computational resources than the Fourier transform. The computational time of the method to compute the signal from n points of data is O(n), whereas the time with Fourier transform is O(n log n).

Method and Result
We confirm our method with a large detection volume of the ensemble NV centers. The microwave and optical systems used in our experiment are shown in Fig. 1. The diamond sample containing NV centers was placed between the microwave resonator and a compound parabolic concentrator1. We used a coplanar waveguide resonator (CWR) with a wide center electrode for strong and uniform microwave irradiation2. The power of microwave with frequency modulation is 25 dBm before the CWR. The 300 mW laser beam to the diamond was focused with a beam diameter of about 30 μm. The fluorescence from the NV centers was collected through the compound parabolic concentrator and detected using a photodiode. The detected signal of the photodiode is analyzed by digital signal processing. The method obtains the result using a digital filter that effectively becomes a sinc filter using the orthogonality of trigonometric functions by multiplication of the acquired data by a trigonometric function. 
Figures 2 shows continuous-wave optical detected magnetic resonance (CW-ODMR) spectrum using the digital signal processing with a different modulation frequency of the microwave. The maximum slope and the noise of the signal become smaller with increasing the modulation frequency for different reasons. The low transition rate between quantum states of the NV center reduces the maximum slope for high modulation frequency. While the noise becomes smaller for high modulation frequency. We evaluate the maximum signal slope (Fig. 3) and noise level of the spectrum (Fig. 4). We found the reduction of the noise is greater than the reduction of the maximum slope.
This work was supported by MEXT Quantum Leap Flagship Program (MEXT Q-LEAP) Grant Number JPMXS0118067395., The 2nd International Forum on Quantum Metrology and Sensing への出席},
 title = {Temperature measurement and annealing behavior of silicon vacancies in 4H-SiC},
 year = {2019}
}