Font Size: a A A

Key Techniques For The Slow Light Fiber Optic Gyroscope

Posted on:2010-02-14Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y LiFull Text:PDF
GTID:1102360278496140Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
Fiber optic gyroscope (FOG) is an inertial measurement unit. The advantage andapplication prospect of FOG is received much concern. It has become the core compo-nent in the new generation of inertial guidance system. In order to improve the perfor-mance, a series of improvement measures are put forward with further study on FOG.Based on requirements of high precision of FOG, principle of FOG consisting of cou-pled micro-ring resonators using slow light technology in fiber, analysis and modeling ofFOG's random noise and error, and time difference measurement of output signal of slowlight FOG are studied deeply in this dissertation, which is funded by the Pre-ResearchFund for Explore No.1 *** Item. The main contents of this dissertation are as the follows:Principle for producing slow light and slowing velocity of light in fiber are analyzed.Two structures of fiber optic gyroscope with slow light is analyzed, one of which isdesigned configuration of slow light FOG with slow light that utilizes phenomena ofDoppler effect, it may greatly boost the scale factor of slow light FOG by n2g times (ng as large as 105~108 magnitude). However, the premise of the calculation is thatthere is a relative motion between the interferometer and the medium. This constraintindicates that the gyroscope with slow light is not suitable for the navigation gyroscope.Another structure is based on coupled-resonator fiber optic gyroscope with slow lightthat the group velocity slows down in high dispersive medium. In this dissertation, wedemonstrate the principle of using this process to enhance the sensitivity of fiber opticgyroscope, which can improve the scale factor by ng/n0 times. Meanwhile, the relationis obtained between output signal of new structure FOG and the numbers of coupledresonator rings N, the coupling coefficientμ, the resonator rings'radius R. At the sametime, the single axis table system for test slow light gyroscope is designed. In order tostudy new high sensitivity of FOG, a new way is opened up.Based on Allan variance principle, the analysis is proposed using dynamic Allanvariance (DAVar) for random error of slow light FOG by the applying of window func-tion to Allan variance. The types, source and characteristics of random noise whichinduce drift error of fiber optic gyroscope is analyzed systematically. Specially, the SBSeffect and stability of light velocity deteriorated noise figure and reduced signal-noise-ratio. The effect of classical Allan variance is to be compared with the DAVar in analysisrandom error of slow light FOG. Dynamic Allan variance has advantages in that not only the method can determine the coefficient of various random drift, but also can track anddescribe the variation in time of the clock stability. The results of simulation data showthat the dynamic Allan variance is the effective method for study the random error offiber optic gyroscope. Meanwhile, this method can reveal and identify in detail the sta-tistical characteristics of main random error which in?uence the precision of slow lightFOG.The random drift is main factor which in?uence the precision of slow light FOG.The mathematical model of random drift of slow light FOG is the basis to analyze therule of motion of FOG, to improve the dynamic performance. In this dissertation randomdrift model is built by time series ARMA model. Because of the problem that the identifi-cation precision is reduced when noise corrupts the observations. Thus the ARMA modelparameters are obtained directly form the estimates of the damped sinusoidal model pa-rameters with guaranteed stability and using a correlation matching technique. The sim-ulation results show that the proposed method has a good convergence and accuracy,in particular for low signal-to-noise ratio. The determinate ARMA model is applied topredict random signal model of slow light FOG. The model can describe well and trulyrandom process of slow light FOG by analyzing diagram and can predict accurately thechanges of random signal.Because of the magnitude is about ns of the time difference between two counter-propagating light in the new fiber optic gyroscope by slow light technology, so themethod which using the time difference measurement to derive the angle of slow lightFOG is proposed. In theory we do much research of the time difference measurementmethod and apply and simulate generalized cross correlation algorithm to estimate timedifference, but this method is poor when signal-to-noise radio low in time differenceestimation. In this dissertation the time difference estimation based on fourth-order cu-mulant is proposed. Better effect has been obtained by the method and the results underthe conditions of correlated and uncorrelated noise are compared with generalized crosscorrelation algorithm. Then some time difference measurement is studied and the time-to-digital converter chip TDC-GP2 is used to time difference measurement for slow lightFOG. This method combines coarse-count with fine-counts measurement, and the resolu-tion achieves 65ps. Finally hardware block diagram and software ?ow of time differencemeasurement system is given.
Keywords/Search Tags:fiber optic gyroscope, slow light, dynamic Allan variance, ARMA parameter estimation, time difference measurement
PDF Full Text Request
Related items