While at the outset this may solely appear as a mine detector, this system is quite general and capable of detecting numerous buried anomalies as defined by a difference in the electrical properties of the object buried in the earth background.  GeoModel, Inc. is one company producing systems for buried detection of objects. This would include pipes, rocks and gradations in earth denoting possible old structures now buried (archaeology).  Each represents a possible change in the permittivity, permeability and conductivity versus the background and would affect the received signal strength. The question then becomes how to optimize and interpret the interactions of the electromagnetic fields.  Several examples of operation are shown here which begin to elucidate the operation of this device. An understanding of the detection mechanisms will help improve the design.
A 2D HOC buried anomaly detector design over generic empty background earth (control).
a_411x201

One of the first simulations should be the detector design over an unfilled background earth.  The earth is simulated with a relative permittivity of 2.5 which is roughly comparable to dry loamy soil.  A 790 MHz CW source was used for the transmitter. This shows the detector in operation without an anomaly and will be considered the control for further analysis.

A 2D HOC buried anomaly detector design over a rectangular block of homogeneous dielectric constant.
b_411x201

As seen in the control simulation, most of the energy is directed into the earth.  This is good because more energy will be available for detection of anomalies below the earth surface.  Note from the outset that some energy is escaping at the left side of the detector where the transmit cavity is.  Note also that significant portions of the energy radiating into the earth are not directed towards the detection cavity.  This becomes an issue when multiple anomalies are present.  In essence, the transmit cavity and transmitter are not directive enough to determine where the anomaly is only that it exists.  This is the extreme, of course as the detector does detect location as will be shown below.  The key here is that energy flowing to the left will couple with objects on the left and then possibly reradiate to the receive cavity which could muddle the received signal from the anomaly directly beneath the cavity.  This effect could yield detection flaws.

A 2D HOC buried anomaly detector slightly modified design over a rectangular block of homogeneous dielectric constant.
c_411x201

The next simulation is based off a rectangular block of dielectric buried beneath the earth surface.  This anomaly has a width of 11.811 inches, a height of 3.1496 inches and a relative permittivity of 8.0.  The anomaly was centered 5.118107 inches below the surface of the earth and centered relative to the detector center.  The numbers were chosen to match a particular physical experiment.

Clearly, the signal strength at the receive (right) cavity has increased which would indicate the presence of an object.  This detection will be clarified a bit later, but for now consider the animation.  Locate the lensing effect at the upper left corner of the anomaly and also locate the pulses which travel along the upper surface of the anomaly.  Follow the pulses along the top of the anomaly and watch as they reradiate at the far right top edge of the anomaly and then couple into the receive cavity.  In this scenario, the reradiating appears to be the mechanism for detection.

The next simulation shows a simple modification of the detector was chosen to illustrate detector design issues.  The outer walls of the transmit and receive cavities were extended to the earth surface.  This could be accomplished via a light metal screen or a set of non-ferrous metal wires.  Non-ferrous would avoid some metal detection devices in a mine trigger device.  This, of course, begins to raise the issue of overall detector design.  The emitted signal must be of low field strength in order to avoid detection by the anomaly device.  However, the lower emitted signal strength will diminish the capability to penetrate the earth for detection.

Interactions of a 790 MHz CW buried anomaly detector with a set of three pipes. Note the pipe on the right barely perturbs the incident fields.
d_411x201

Note that this simple design modification did not appear to effect a change in the received signal strength.  The dominant coupling mechanism remains the lensing and surface guided waves through the anomaly.  The surface guided waves then reradiate upon termination of the anomaly geometry.

A slightly different structure shows the detector design works for pipes as well as rectangular objects.  These results could be compared to the control.  The center PVC (relative permittivity 2.8) pipe is 10 inches in diameter, 10 inches underground, centered relative to the detector and filled with water (relative permittivity 78.0).  The left metal pipe is 8 inches in diameter, 7 inches underground, 7 inches left of detector center and filled with jet fuel (relative permittivity 2.08).  The right PVC pipe is 4 inches in diameter, 5 inches underground, 7 inches right of detector center and filled with jet fuel (relative permittivity 2.08).

The left and center pipes clearly increase the receive signal strength.  The right pipe is barely perturbing the fields and would likely be undetected in the current system.

Interactions of a 790 MHz CW buried anomaly detector with a set of three pipes. The detector height was moved from 3.1496 inches previously to 7.0 inches in this case.
e_411x201

The next simulation illustrates the effect of raising the detector too high above the earth surface.  Only the height has changed to 7 inches from 3.1496 inches in the previous cases.  Clearly the receive signal has increased yet this is not a beneficial increase.

The coupling is clearly traveling through the air gap between the earth surface and the metal septum of the detector.  This corroborates the results in hayes91:_trans_line_matrix_tlm_method hayes90:_trans_line_matrix_model where detector height sensitivity was studied.

Scattered field result from control and rectangular homogeneous dielectric block data exhibiting lensing and surface guided wave.
ab_411x201

TDLO methods also offer the capability to study scattered fields.  By comparing the actual fields in the control and rectangular homogeneous dielectric data before the images were made, a scattered field result may be calculated, that is, the fields which have changed due to the presence of the anomaly.

Scattered field result from control and simple detector modification data exhibiting little change in receive cavity signal strength.
bc_411x201

This analysis shows this scattered field result which is the effect of the anomaly on the default background.  The lensing effect in the upper left corner of the anomaly and the top surface guided wave are clearly shown.  A scattered field result may also be calculated for the simple modification of cavity side walls extending to the ground surface as shown below.

Ideally, the most prominent change should be in the receive cavity indicating an increased receive signal strength.  Thus, the simple geometry modification was not very effective.