The principles of GROT 12 creation
Ground penetrating radars, which are produced according to the traditional scheme and use quasi-monochromatic signals, can provide a probing depth of only a few meters, even under condition that soils absorb radio waves very weakly (the parameters of such soils are to be close to dry sand’s ones), and that makes these devices extremely inefficient for usage in central Russia or other places, where clay soils prevail. The main goal we set ourselves, while creating a series of ground penetrating radars of new generation, was to achieve the highest possible real potential of the device.
This objective can be achieved in two main ways: by using a powerful transmitter and by registration of the signal in its own frequencies range without its stroboscopic conversion to low frequencies spectrum. In other words, we have not just eliminated most of the operations on the signal, but also operations that could lead to the appearance of undesirable side "ringing". To get the required parameters of object detection, we had to develop compact spark and solid-state transmitters with pulse power of the order of several and up to dozens of megawatts. They are functionally modern versions of spark transmitters, used by Alexander Popov and Guglielmo Marconi in the first experiments on the radio communication.
The applied by us method of signal registration is based on the use of high-speed comparators, which compare the incoming signal with a certain predetermined threshold. The ability to change the threshold value and the receiver gain allows detecting the signal in a wider range of its values. Usage of the stroboscopic method allows (with the transmitter emitting one pulse) registering the value of the signal amplitude only at a certain moment of time, while our method within a single pulse registers all the moments, when signal amplitude exceeds the threshold value, throughout the whole time coordinate, describing the pulse propagation in the medium.
It was proved, that the most suitable receiving and transmitting antennas are resistively loaded dipoles, in which a portion of the pulse energy is absorbed by the resistors, distributed along its arms. By choosing the values of the resistors almost complete damping of by-oscillations of the pulse in the antennas can be achieved.
The optimal construction of the antenna, used under active or passive external interference conditions, is a resistive-loaded dipole covered with a dielectric box filled with carbon radio waves absorbent that weakens the air wave, while probing takes place from the Earth’s surface, and reflections, while working in tight quarters, for example, in the mines.
An important characteristic of ground penetrating radar is its real potential. Last years, the low real dynamic range, which (depending on the manufacturer) is 40-60-80 dB, has been preventing the increase of probing depth of ground penetrating radars, which are built according to the traditional scheme. The real potential is the maximum value of signal attenuation in the medium, despite which the radar is able to detect underground objects. This parameter is critically important to evaluate the limiting depth of GPR scanning. Unfortunately, quite often the real potential value, given in the device’s specifications, is calculated as the ratio of capacity of the transmitter to the sensitivity of the receiver, while that valuation doesn’t represent the practical capabilities of the device.
The work principle of most georadars, produced at the moment, is based on the stroboscopic method of conversion of the signal to the low-frequencies spectrum, in which it is registered. The main problem of this conventional solution deals with providing constant characteristics of amplitude-frequency and linear characteristics of phase-frequency, while processing stroboscopic transformation of the signal in the receiving track, and leads to significant side signal oscillations (forced "ringing") and concealment of weak signals by stronger ones. This is the main reason for the comparatively small real potential of this type of georadars.
The following instruction describes one of the ways to measure the comparative real potential of georadars, produced by different manufacturers. The ground penetrating radar, which is under study, is placed on floats and moved from the coast to the centre of a deep pond. While the device is being moved, the depth, at which the pulse reflection from the bottom of the pond disappears (as greater amounts of water absorb radio waves more efficiently), is registered. The values of the amplitude function allow determining the attenuation per meter as a ratio of the amplitude change to the depth change. The real potential is calculated as the multiplication of the attenuation and the depth, at which the signal was lost. The real potential, measured via procedures described above, of our GROT family GPRs is no less than 120-160 dB, depending on the type of the device.
Methods of GPR data processing
Radargrams, produced by our devices, nearly don’t have side interfering oscillations ("ringing" of the equipment) that are typical for many other ground penetrating radars. For this reason, we do not use standard GPR data processing programs, whose main task is to reduce the magnitude of "ringing" and to highlight the signal using various filtrations.
The method, which we use at the moment to restore the geological shears using radargrams, is based on the use of a procedure, which is known in seismology as the "common source point" (CSP).
At first, the radiolocation image is taken with the device, the distance between the receiving and transmitting antennas of which is predetermined. Then, the points are located, at which scanning is to be done, and scanning means, according to the method of the CSP, getting the hodographs from the layers and objects. The hodograph is a function of the delay of the signal reflection from layers (objects) that depends on the distance between the symmetrically separated receiving and transmitting antennas.
The hodograph allows determining both the real depth of the layer and the wave propagation speed in it. To convert the radargram into the geological cross-section, it is necessary to exclude the aliquot reflections from the layers and to transform the time axis into the dimensional one, while setting the speed of the wave in the layer. All the information, necessary for this procedure, can be obtained from the hodograph.
The usage of the quasi-seismic approach, while producing geologic cross-sections, can’t be considered satisfactory, as it doesn’t use the information that can be obtained by studying the amplitude of the signal, its temporal shape and polarization. This is why in many cases the measurement of the signal waveform is the only way to locate the studied object correctly. Algorithms and programs that are used to process signal waveforms were created, and the method is based on the use of a well-known algorithm of the Wavelet transform.
The presented principles of georadar construction and procedures of signal processing let GROT family ground penetrating radars solve all the basic tasks of modern georadiolocation.