System identification and analysis of subsurface radar signals

• US 5,384,715 A
• Filed: 08/27/1993
• Issued: 01/24/1995
• Est. Priority Date: 08/27/1993
• Status: Expired due to Fees

• Current Assignee: Texas A&M University System
• Original Assignee: Texas A&M University System
• Inventors: Lytton, Robert L.
• Primary Examiner(s): Harvey, Jack B.
• Assistant Examiner(s): Tran, Alan
• Application Number:
  US08/113,435
• Time in Patent Office: 515 Days
• Field of Search: 364/421, 364/422, 364/550, 364/420, 324/334, 324/337, 342/5
• US Class Current: 702/12
• CPC Class Codes: G01V 3/12 operating with electromagnetic waves operating with millimetre waves G01V8/005

Abstract

A method in accordance with the invention is implemented via a computer program to determine the density and water (or other fluid) content of the various layers within a multilayer system. For convenience, the program is referred to as SIDARS. A conventional ground penetrating radar (GPR) system is used to obtain digitized images of a reflected radar signal from a multilayer pavement system. From these images standard mathematical techniques are applied to determine the number of layers, the thickness of each layer, and the dielectric constant for each layer within the multilayer system. SIDARS takes advantage of the fact that each layer is itself composed of three distinct types of material: solids, fluids, and gases. Thus, the dielectric constant obtained for a layer is in fact a composite value, namely a combination of the layer'"'"'s solid, fluid, and gas dielectric constants. SIDARS employs a wave propagation model of the pavement system to generate a synthetic reflected radar signal. Through an iterative process, initial concentration estimates of each material (solid, fluid, gas) in each layer are adjusted to minimize the mean-squared-error between the (measured) reflected and (calculated) synthetic radar signals. This process converges rapidly, is largely unaffected by initial values (over a wide range) for a layer'"'"'s solids, fluid, and gas concentrations, and yields accurate values for the density and water content of each layer. Advantageously, these results have been obtained from a vehicle mounted GPR system while cruising at speeds of 40 to 50 miles per hour.

Claims

1. A method of determining the respective volumetric proportions, in a target layer of a multilayer system, of one or more of a solid portion of the target layer, a fluid portion of the target layer, and a gaseous portion of the target layer, comprising the steps of:

(a) directing an energy pulse into the target layer of the system;

(b) receiving a combined reflection signal from the system;

(c) generating a measurement vector signal in which is encoded a characteristic of the combined reflection signal;

(d) generating an estimate vector signal in which is encoded respective estimated volumetric proportions of said one or more of the target layer'"'"'s solid, fluid, and gas portions;

(e) generating an error value signal representing a differential between the measurement vector signal and the estimate vector signal;

(f) testing whether a characteristic of the error value signal is greater than a predetermined criterion and (1) if so, then

(A) adjusting the estimate vector signal to encode adjusted respective estimated volumetric proportions, and

(B) repeating steps e and f;

else, (2) generating a layer-content signal in which is encoded the respective estimated volumetric proportions encoded in said estimate vector signal.

2. The method of claim 1 wherein the fluid portion of the target layer comprises water.

3. The method of claim 1 wherein the energy pulse is electromagnetic.

4. The method of claim 1 wherein the combined reflection signal comprises a plurality of reflection signals, each reflection signal being a reflection of the energy pulse from a different location in the system and wherein the measurement vector signal comprises a plurality of measurement elements, each measurement element representing an instantaneous value of a characteristic such as the magnitude, phase, or frequency of a respective reflection pulse.

5. The method of claim 1 wherein, the estimate vector is generated by a process using a forward propagating wave transmission model.

6. A method of modeling the density and water content of a target layer in a multilayer formation, comprising the steps of:

(a) transmitting a radar signal into the formation,

(b) receiving a reflected radar signal back from the formation,

(c) converting the reflected radar signal to a digitized reference signal,

(d) generating a modeled reflection synthetic signal that represents the reflected radar signal that would be expected to be received from a model formation having a model target layer with a specified density and specified water content;

(e) testing whether the synthetic signal matches the reference signal to within a predetermined degree of satisfaction;

(f) if the test of step (e) is satisfied then generating a success signal indicating that the density and water content of the target layer in a multilayer formation are satisfactorily approximated by the density and water content in the model target layer;

else altering the synthetic signal to represent an adjustment to the model target layer in accordance with a predetermined iterative procedure and repeating steps (d) through (f).