Research On Techniques Of Weak Magnetic Sensors Based On Giant Magnetoimpedance Effect

The use of geomagnetic field matching techniques assists navigation of underwater vehicles. Currently the navigation methods for underwater vehicles such as inertial navigation, acoustic navigation, gravity navigation, INS/GPS navigation and others have some limitations. The geomagnetic assisted navigation, however, due to several advantages such as passive, autonomous, non-cumulative error and covert, has stimulated great research interest of domestic and foreign scholars in recent years. One of the key technologies of geomagnetic assisted navigation is how to achieve accurate measurements of the magnetic field. Compared with magnetic sensors based on giant magnetoresistance, Hall effect, fluxgate and superconducting quantum interference, magnetic sensors based on giant magnetoimpedance(GMI) effect possess many advantages such as high resolution, fast response, small size, low power consumption. Therefore, study of GMI magnetic sensors based on the amorphous wire and the weak magnetic signal detection technologies has important theoretical and practical value. The dissertation researches on the key technologies of weak magnetic sensors based on GMI effect and the application in the weak magnetic anomaly signal detection. The main research work and results are summarized as follows:(1) The influencing factors and material types of GMI effect are studied and the results show that the GMI effect is different with different material types and processing methods. For example, after suitable annealing the cobalt-based amorphous wires and ribbons have better performance. Currently, various theories of the mechanism of GMI effect can only give approximate interpretation, or even qualitative description. By testing the GMI effect of the amorphous wire sample, it was noted that the maximum change of the impedance can reach 160%.(2) Due to the defect of the traditional peak detection circuit of GMI magnetic sensors, particularly difficult to detect the noisy signal of SNR less than 0d B, two novel weak magnetic signal detection systems are proposed. The first GMI magnetic sensor detection circuit is based on orthogonal-vector lock-in amplification technique and the second is based on a differential-amplification approach. For the first approach, due to the defect of traditional voltage-mode amplifiers in gain-bandwidth product and slew rate, high-frequency quadrature oscillation circuits based on two current-mode amplifiers were designed and the performance was verified by simulation experiments. For the second one, the output signal from the GMI probe is directly sampled after conditioning and then sent to the high-speed digital signal processing system based on FPGA and DSP in which the weak signal can be extracted by software algorithms.(3) The hybrid detection method based on wavelet transform and correlation analysis is proposed to achieve detection and estimation of GMI magnetic signals from amorphous-wire probe with low SNR. The principle of the hybrid method is that the true signal is extracted by wavelet decomposition from noisy sampled signal, then processed by correlation with reference signal and at last calibrated with the standard magnetic field. It is stated that the more the decomposition layer is, the more sparse the data is. Thus, the porous algorithm is adopted in the fast wavelet decomposition. The multiplication amount will increase when the data is great in the correlation computation. Thus, the fast algorithm based on FFT for the correlation computation is selected. Moreover, the relevant simulation experiments were carried out according to the two signal types from GMI magnetic probe. The results indicated that this detection method can detect weak magnetic signals with SNR as low as-15 d B.(4) The application of GMI magnetic sensors in the weak magnetic anomaly signal detection is studied, including orthonormal basis functions(OBF) detection method based on signal characteristics and entropy detection method based on noise characteristics. In order to avoid the defect of OBF detection method, the pre-whitening OBF detection method based on an adaptive AR model is proposed. In contrast to the OBF method, the entropy detection method requires less prior knowledge. Simulation results showed that the two algorithms can be used to detect weak magnetic anomaly signal.