When measuring in a geomagnetic field, the method of magnetic resonance sounding (MRS) is often limited because of the notably low signal-to-noise ratio (SNR). Most current studies focus on discarding spiky noise and power-line harmonic noise cancellation. However, the effects of random noise should not be underestimated. The common method for random noise attenuation is stacking, but collecting multiple recordings merely to suppress random noise is time-consuming. Moreover, stacking is insufficient to suppress high-level random noise. Here, we propose the use of time–frequency peak filtering for random noise attenuation, which is performed after the traditional de-spiking and power-line harmonic removal method. By encoding the noisy signal with frequency modulation and estimating the instantaneous frequency using the peak of the time–frequency representation of the encoded signal, the desired MRS signal can be acquired from only one stack. The performance of the proposed method is tested on synthetic envelope signals and field data from different surveys. Good estimations of the signal parameters are obtained at different SNRs. Moreover, an attempt to use the proposed method to handle a single recording provides better results compared to 16 stacks. Our results suggest that the number of stacks can be appropriately reduced to shorten the measurement time and improve the measurement efficiency.

The magnetic resonance sounding (MRS) technique is a non-invasive geophysical method that can provide unique insights into the hydrological properties of groundwater. The Cu coil sensor is the preferred choice for detecting the weak MRS signal because of its high sensitivity, low fabrication complexity and low cost. The tuned configuration was traditionally used for the MRS coil sensor design because of its high sensitivity and narrowband filtering. However, its narrow bandwidth may distort the MRS signals. To address this issue, a non-tuned design exhibiting a broad bandwidth has emerged recently, however, the sensitivity decreases as the bandwidth increases. Moreover, the effect of the MRS applications is often seriously influenced by power harmonic noises in the developed areas, especially low-frequency harmonics, resulting in saturation of the coil sensor, regardless of the tuned or non-tuned configuration. To solve the two aforementioned problems, we propose a matching network consisting of an LC broadband filter in parallel with a matching capacitor and provide a design for a coil sensor with a matching network (CSMN). The theoretical parameter calculations and the equivalent schematic of the CSMN with noise sources are investigated, and the sensitivity of the CSMN is evaluated by the Allan variance and the signal-to-noise ratio (SNR). Correspondingly, we constructed the CSMN with a 3 dB bandwidth, passband gain, normalized equivalent input noise and sensitivity (detection limit) of 1030 Hz, 4.6 dB, 1.78 nV/(Hz)^1/2 @ 2 kHz and 3 nV, respectively. Experimental tests in the laboratory show that the CSMN can not only improve the sensitivity, but also inhibit the signal distortion by suppressing power harmonic noises in the strong electromagnetic interference environment. Finally, a field experiment is performed with the CSMN to show a valid measurement of the signals of an MRS instrument system.

SUMMARY: Magnetic resonance sounding (MRS) can be used to implement groundwater detection non-invasively because of the direct relationship between the signal parameters and aquifer properties. Most current approaches to extracting MRS parameters need to use prior knowledge of the precise Larmor frequency to detect the signal envelope, and are vulnerable to the residual noise after de-noising. In this paper, we use a shaping-regularized Prony method to extract MRS signals from high-frequency oscillating data instead of from envelopes of data. Only an approximate Larmor frequency is required in advance. Besides, irrelevant noise components can be dropped during the extraction process. We investigate the selection of an appropriate decomposition quantity for the SRP method and present an intuitive regularization method to control the shape of the estimated coefficients. Impact of different types and magnitudes of noise on the proposed method are analysed. The extraction results obtained from numerical simulations and field surveys reveal that the SRP method yields accurate and reliable parameters estimation and preserves complete signal information. The tolerance to the interference in signal extraction of the proposed method is obviously better than that of the SD method.

UMRS can effectively explore the aquifer information in front of the tunnel face. However, the complex tunnel construction environment will reduce the reliability of the interpretation results. We propose and implement using joint inversion of UMRS-TEM data in tunnel detection using HMC for the first time to solve this problem. Joint inversion can update the UMRS kernel in real-time during the iterative process, obtain accurate aquifer information and resistivity structure, and improve the reliability of the interpretation of inversion. HMC is a MCMC method that uses Hamiltonian dynamics to propose future states in Markov chains. It can explore the target distribution more effectively, and it also has the advantage that MCMC can obtain a posteriori PDF of parameters, we believe that PDF information can effectively judge the accuracy of inversion results. We validated the effectiveness and practicality of the joint inversion using synthetic and observed data, and the experimental results showed its advantages in accuracy and noise resistance compared to a single UMRS inversion. In the field example, the water content PDF of the joint inversion is significantly increased by 31.8% compared with that of the single inversion. We analyzed the correlation between water content, resistivity, and layer interfaces and discovered some correlation laws. HMC improves the efficiency of computational and provides assistance for further research on the influence of parameters on inversion results. Our conclusions can improve the safety of tunnel construction and provide effective technical support for avoiding hydrogeological disasters in tunnels.