No abstract

To reduce the measurement cost and time, a time domain measurement method of quality factor for MEMS resonator is presented in this paper. The relation curve between quality factor and vacuum level is obtained. The results indicate that measurement error is larger under high vacuum level when the sampling frequency is lower. The reason causing the error is analyzed and measurement precision is improved by increasing the sampling frequency. The measurement results indicate time domain measurement is advisable and improvable for MEMS resonator measurement.

In this paper, a novel lapping method based on regulating the position of carrier centroid is proposed to modify interfacial normal pressure uniformity. Eight special points are selected to represent carrier weight. This lapping process can be divided into initial stage, regulated stage and stable stage. The purpose of initial stage is calculating the position of carrier centroid according to the equivalent mass of eight points. The regulated stage is to decrease total thickness variation (TTV) by regulating the position of weight. Finally, the stable stage will keep uniformity of material removal rate (MRR) uniform at each point. A 3-inch and 400 μm thickness silicon wafer is lapped to demonstrate the feasibility of this method. We can find that TTV of this wafer decreases from initial stage 20 μm to 3 μm and remain constant. Therefore, the uniformity of MRR has been greatly improved by this novel lapping method.

The basic features extracted by traditional methods for speech quality evaluation are not clear, which leads to the small correlation coefficient of subjective and objective evaluation value. Therefore, an automatic voice quality evaluation method for IVR service in call center based on stackable automatic encoder is proposed. All kinds of devices are used to simulate the real use of IVR service voice of call center and collect IVR service voice of call center. According to the process of sampling quantization frame pre emphasis window processing, the IVR service voice data of call center is pre processed. Based on the structure of stackable automatic encoder, the reconstruction process of coding and decoding is designed to extract the basic features of business speech. BP neural network is introduced to establish an automatic speech evaluation model to evaluate speech quality automatically. Experimental results: compared with the traditional method, the average correlation coefficients of subjective evaluation value and objective evaluation value are 0.023517 and 0.02258 respectively, and the average deviation of correlation coefficient is 0.048775 and 0.03485 respectively.