Design and Implementation of Digital Signal Detection System for RFOG
|Course||Microelectronics and Solid State Electronics|
|Keywords||Resonator fiber optic gyro digital PI controller CORDIC algorithm FPGA|
Fiber optic gyro is an inertial angular velocity sensor with high precision based on the Sagnac effect. It plays an important role in a variety of navigation and guidance systems. Compared with the well-developed interference fiber optic gyro (IFOG), resonator fiber optic gyro (RFOG) has the advantage of system miniaturization. The digital signal processor has better stability, quicker processing speed, smaller size and stronger anti-interference ability when compared to the analog circuit. Using FPGA as the core hardware, the digital signal processor for the RFOG based on the sinusoidal phase modulating technology is designed and implemented. The lock-in accuracy of the resonance frequency servo loop has been improved. A miniaturized digital RFOG system is carried out. The main research works of this article are as follows.A digital proportional-integral (PI) controller is incorporated into the resonance frequency servo loop. A simulation model is setup to optimize the loop parameters considering of the loop noise reduction and dynamic performance of the RFOG system. The digital PI controller implemented on FPGA significantly reduces the loop delay, which greatly improves the noise suppression in the low frequency range. Through a reset mechanism, the long-term lock-in performance of the PI controller is improved. The lock-in frequency stability of the resonance servo loop is about 0.044/h (1σ) with the integration time of 0.16 seconds (equivalent bandwidth of 1 Hz). It is close to the theoretical sensitivity of the RFOG.A digital signal processor is designed and implemented on FPGA for an RFOG based on the sinusoidal phase modulation technology. Both the modulation signal generators and the synchronous demodulation are all digitalized based on the CORDIC algorithm. To overcome the backscattering induced noise, two sinusoidal voltage waveforms with different modulation frequencies are applied to the clockwise (CW) and counterclockwise (CCW) lightwaves of the resonator, respectively. Their amplitudes are also optimized to reduce the backscattering induced noise. The digital lock-in amplifier implemented on FPGA can detect a small signal of only to 10nV, corresponding to a rotation of 10-4°/s for the RFOG used in our experiment. To enhance the usability and flexibility of the RFOG system in debugging and data processing, a serial communication and control system between the computer and FPGA are designed and implemented.An RFOG based on the aforementioned digital signal processor implemented on FPGA is setup. Rotation rates have been measured from 0.05°/s to 5°/s. A bias stability of about 68.31°/h in an hour is successfully demonstrated in the digital RFOG with the integration time of 0.16 seconds.