Maritime Magnetic Anomaly Mapping with a Diamond Nitrogen Vacancy Sensor
Tyler Fleig and Paul Frontera, US Naval Academy
Location: Big Sur
Accurate position information in the absence of a GPS signal is desirable for many civilian and military applications. One source of navigation information is the detection and measurement of Earth’s magnetic field anomalies, which may act as an input to a navigation filter in order to provide accurate position information. These systems estimate position by comparing magnetometer measurements to that of mapped or modeled magnetic fields. The goal of this paper is to compare data sources for use in magnetic anomaly mapping. Comparisons will be made between the currently available maps as well as new data from a diamond nitrogen vacancy (DNV) magnetometer and a fluxgate magnetometer. The comparisons will include spatial resolution, absolute precision, proper geo-registration, and any potential limitations for use in a magnetic navigation filter. There are several current magnetic anomaly models including the World Magnetic Model (WMM) and the Enhanced Magnetic Model (EMM). There are also databases that include data gathered by airplane and other survey platforms. These include the North American Magnetic Anomaly Database (NAMAD) and the World Digital Magnetic Anomaly Map (WDMAM) .
Diamond Nitrogen Vacancy Sensor
Magnetometry using nitrogen-vacancy defects within diamonds allows for accuracy in the nanoTesla range and has other benefits over methods of measuring weak magnetic fields . DNV sensors are comprised of a diamond with nitrogen vacancies purposefully place along 4 axes in the lattice structure. A green light is directed through this diamond while it is simultaneously subject to RF stimulation. A red light results varying in intensity proportional to the magnetic field in which the sensor resides.
Scalar maps and sensors
Scalar magnetic field maps are easy to generate and represent visually, but navigation filters which use only scalar measurements are inherently less accurate than those which use a vector representation of the local magnetic field . These scalar maps can be created with a scalar magnetometer, which can be accurate to fractions of nT . It is straightforward to create these maps, as data collected during survey only has to be tagged with a location and a magnitude. This is in contrast to vector maps that require measurement of the magnetic field in the local tangent plane. Therefore, the sensor’s attitude must be determined precisely to determine the magnetic field orientation.
Vector maps and sensors
Vector magnetic field maps are produced using either a vector magnetometers or a by a unit consisting of tri-axially mounted scalar magnetometers. These maps can be accurate to about 100 nT using a fluxgate magnetometer . This reduction in accuracy in comparison to a scalar sensor is due to the noise of all three individual fluxgate sensors in the vector array, as well as a lack of perfect orthogonality between the three sensors. The accuracy of maps generated using a diamond nitrogen vacancy magnetometer is not yet known and will be discussed in this paper once the experimental data has been analyzed. The DNV magnetometer only has one channel where noise can be introduced, whereas the fluxgate sensor has independent noise on all three output channels (X, Y and Z axes). The paper will discuss whether the extra resolution and noise reduction using a DNV magnetometer increases map quality by a significant amount over a fluxgate magnetometer.
There are several added challenges in creating a vector map versus a scalar map. Vector readings from the sensor must be rotated by the attitude of the sensor itself. This way the readings are in the East, North, Down frame rather than the sensor’s frame. This means that an accurate inertial measuring unit is needed in addition to the GPS data necessary to properly geo-register the data collected during survey. Also, visually representing a vector map in order to determine quality and suitability for use in a filter is much more difficult, as a vector value must be plotted at each point, rather than a scalar value which can be easily represented by color. This paper will also discuss how a vector map will be more suitable for use in a navigation filter due to the ability to find three independent gradients as one moves through an area, allowing for much more accurate and precise position determination.
Expected data and results
Data has been collected using a DNV sensor as well as a 3-axis fluxgate sensor in two separate regions with modeled magnetic anomalies in the vicinity of Annapolis, Maryland. The data was collected using a 108 foot US Navy Yard Patrol Craft over a period of several days. This paper will discuss magnetic anomaly maps generated using the data from both of these sensors as well as several other currently available maps and models.
 Aaron Canciani, “Absolute positioning using the earth’s magnetic anomaly field,” p. 242, Sep. 2016.
 L. Rondin, J.-P. Tetienne, T. Hingant, J.-F. Roch, P. Maletinsky, and V. Jacques, “Magnetometry with nitrogen-vacancy defects in diamond,” Rep. Prog. Phys., vol. 77, no. 5, p. 056503, May 2014.K.
 V. Y. Shifrin, V. N. Khorev, V. N. Kalabin, and P. G. Park, “Experimental estimation of the accuracy of modern scalar quantum magnetometers in measurements of the Earth’s magnetic field,” Phys. Earth Planet. Inter., vol. 166, no. 3, pp. 147–152, Feb. 2008.