No abstract

No abstract

The prototype for a nitrogen-cooled high-T, SQUID gradiometer has been developed and is being evaluated for magnetic anomaly detection of underwater targets in mobile surveys. The prototype's design is based on the concept of the Three-Sensor Gradiometer (TSG). In the TSG approach, balance of two independent SQUID magnetometers is more difficult to attain than for conventional low-T, gradiometers in which signal subtraction occurs prior to a single SQUID stage. Experiments have been conducted using a platform-motion simulator to evaluate performance of this gradiometer for mobile operation. Sensor configuration, experimental procedures, approaches for improved performance, and empirical results are reported. Interesting results of predictions to estimate detection range obtained from matched-filter calculations are included. The paper concludes with a description of current preparations for a sea test of this sensor and a perspective of future developments.

The U. S. Navy has demonstrated that total-field magnetometers and gradiometers as well as tensor fluxgate and superconducting magnetic gradiometers can be used in towed underwater platform environments. In these applications, the principal platform noise issue has been that caused by the rotation of onboard magnetic materials in the large background earth's magnetic field. The associated induced magnetic polarization changes and eddy currents cause secondary magnetic fields to be generated at the onboard sensor, and these overwhelm the magnetic signature of objects to be detected and localized.In these applications it has been established that knowledge of the earth's field vector time history in the platform reference frame can be used in a relatively simple fixed-parameter filter model to estimate the "motion noise" contribution to the measured signal, and to effectively remove it. The associated model has been modified to contain up to three magnetic dipole sources, and measured, time-windowed data has been used to simultaneously remove platform motion noise and localize magnetic targets.To operate a magnetic sensor onboard an AUV, it is necessary to deal with the problems encountered on a towed platform, that is, motion noise in the earth's field, but now there are numerous additional magnetic sources that are independent of the external field. These include onboard current loops and magnetic materials that move relative to the AUV platform. Examples are control surface mechanisms, motors and controllers, onhoard processors, and switch magnets.To model the noise from all magnetic sources, it is still necessary to monitor the earth's magnetic field onhoard the platform. It also is necessary to add measurements of onboard currents and magnetic field measurements made very near any sources that move relative to the platform. The basic issue here is that all onboard magnetic sensors will respond to the large earth's magnetic field changes due to rotation of the AUV in the earth's field. The additional magnetic field sensors should be positioned such that the additional field changes due to the local sources are measurable relative to the rotational changes.We describe both frequency-domain and time-domain filter models that incorporate onhoard AUV reference sensor measurements to cancel platform noise, and apply the models to data collected as described in companion papers. We evaluate the relative performance of the time-domain and frequencydomain models, and show that it depends on the measurement sets used to perform the modeling. We conclude that windowed time-domain filters hold much promise for future AUV-resident magnetic sensor systems

No abstract