Optical signal quality monitoring based on asynchronous sampling

碩士 === 國立臺灣科技大學 === 電子工程系 === 93 === We propose to use a 2×2 high speed optical switch as the sampling device to monitor the 2.5Gb/s NRZ optical signal in asynchronous sampling method. With this scheme, the monitoring of optical signal quality and the dispersion compensation module are combined by u...

Full description

Bibliographic Details
Main Authors: Chun-Hao Su, 蘇俊豪
Other Authors: San-Liang Lee
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/58234431883883047254
Description
Summary:碩士 === 國立臺灣科技大學 === 電子工程系 === 93 === We propose to use a 2×2 high speed optical switch as the sampling device to monitor the 2.5Gb/s NRZ optical signal in asynchronous sampling method. With this scheme, the monitoring of optical signal quality and the dispersion compensation module are combined by using a FP etalon. The switch can be regarded as a dynamic coupler by applying different DC voltage and is used to perform the optical sampling while it is biased with a DC voltage and electrical pulse. Here, the repetition rate of the electrical pulse is set as 625 MHz. The signal is optically sampled when the repetition rate is lower than the signal bit rate, so a low-speed O/E converter, of which the bandwidth is narrower than the signal bit rate can be used. With this technique, high frequency circuits are not necessary. The results of asynchronous amplitude histogram can be easily measured by optical sampling. According to the sampled signal, we can estimate the value of the Q-factor and use the obtained Q-factor to adjust the temperature of the etalon. The etalon is tuned by the temperature, so the center wavelength of the filter can be shifted conveniently. The response of the filters can be utilized to compensate the dispersion of the optical fiber to enhance signal quality. For performance monitoring, we adjust the threshold level, coefficient C and analyze the relationship between Q and the threshold level. Finally, we measure the power sensitivity, sampling pulsewidth and coefficient C to optimize the system performance. The results indicated that the input power must be higher than 4dBm and the optimum value of C is 0.25. When the sampling pulsewidth is the same as the bit length of signal, we can measure the Q-factor accurately and reliably.