Real-Time Compensation for SLD Light-Power Fluctuation in an Interferometric Fiber-Optic Gyroscope

Abstract

An interferometric fiber-optic gyroscope (IFOG) demodulates a rotation signal via interferometric light intensity. However, the working environments of IFOGs typically involve great uncertainty. Fluctuations in temperature, air pressure, electromagnetic field, and the power system all cause the power of the superluminescent diode (SLD) light source to fluctuate as well. In this invited paper, we studied the effects of SLD power fluctuation on the dynamic and static performance characteristics of a gyro system through the use of a light-power feedback loop. Fluctuations of 0.5 mA, 1 mA, and 5 mA in the SLD source entering the IFOG caused zero-bias stability to be 69, 135, and 679 times worse. We established an effective method to monitor power fluctuations of SLD light sources and to compensate for their effects without increasing hardware complexity or system cost. In brief, we established a real-time power-sensing and -compensating system. Experimental results showed that for every 0.1 mA increase in the fluctuation amplitude of the driving current, the zero-bias stability became 4 to 7 times worse, which could be reduced about 95% through the use of SLD power compensation.

A Novel Closed-Loop Control to Solve Light Source Power Fluctuations in the Fiber-Optic Gyroscope

“…The two groups of experiments show that when the condition ( k f → k max ) is satisfied, the output signal of the gyroscope crosstalk the differential signal of the LSP. By subtracting a certain proportion of the differential signal of the LSP from the gyroscope’s output signal, it is able to effectively reduce the impact of LSP fluctuations [ 22 ]. However, this condition ( k f → k max ) is not satisfied in most practical applications, because the value of the coefficient k f is different in different applications.…”

“…When k f → k max , the feedback phase φ f generated by the step wave can almost completely cancel the light intensity difference of the previous period and, as a result, condition ( φ f [ n ] = − φ s [ n − 1]) is satisfied. According to our previous research work [ 22 ], if the differential signal of the light intensity, I 0 ( t ), is β ( t ), the output of the gyroscope can be expressed as follows [ 22 ]: where τ is the time required for light to pass through the fiber coil, and I 0 [ n ] is the average of light intensity, I 0 ( t ), in one period of the gyroscope. β [ n ] is the discrete signal of the differential signal of the LSP, which can be expressed as follows: …”

“…The smaller the value of coefficient k f , the lower the system noise, and the higher the value of coefficient k f , the better the dynamic performance of the IFOG. That is to say, in practical applications, the coefficient k f is uncertain, and the prerequisite, ( k f → k max ), of the compensation method used in the previous research may not be satisfied [ 22 ].…”

“…Therefore, the impact of LSP fluctuations varies with coefficient k f . We have given the solution to the case of ( k f → k max ), in our previous research work [ 22 ]. In this paper, we present a compensation scheme for any k f , a significant advance on previous work.…”

“…) is satisfied. According to our previous research work [22], if the differential signal of the light intensity, I 0 (t), is β(t), the output of the gyroscope can be expressed as follows [22]:…”