Ranging, detection and resolving in a multislope frequency modulated waveform radar system

• US 5,191,337 A
• Filed: 02/25/1992
• Issued: 03/02/1993
• Est. Priority Date: 02/25/1992
• Status: Expired due to Term

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Hughes Aircraft Company (Rtx Corporation)
• Inventors: Nguyen, Trung T., Brovko, Oleg
• Primary Examiner(s): Sotomayor, John B.
• Application Number:
  US07/841,133
• Time in Patent Office: 371 Days
• Field of Search: 342/200, 342/163, 342/90, 342/91, 342/92, 342/159, 342/160, 342/161, 342/162, 342/98, 342/99, 342/101
• US Class Current: 342/200
• CPC Class Codes: G01S 13/34 using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal
• CPC Class Codes: G01S 13/584 adapted for simultaneous range and velocity measurements
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/9321 Velocity regulation, e.g. cruise control

Abstract

A maximum likelihood estimator and range-only-initialization target detection method employed to detect and resolve targets in a multislope linear frequency modulated waveform radar. The method resolves a large number of target returns without a large amount of signal processing and without creating a significant number of false alarms, or ghosts. The method simultaneously estimates range and doppler for each target. The method rejects undesired long-range targets that fold into target regions, and processes target regions of interest around a nearest target to reduce signal processing throughput requirements. Using a K out of N detection rule, the method detects targets that compete with mainlobe rain clutter, mainlobe ground clutter, and receiver leakage. The method simultaneously estimates target parameters and optimally resolves any number of targets. The method is limited only by the number of frequency modulation ranging slopes, the slope values, and the doppler filter resolution set by the radar waveform design. The method has the ability to process and detect extended targets.

Claims

1. In a multislope linear frequency modulated waveform radar system comprising means for transmitting and receiving radar signals and for processing received frequency modulated waveform radar signals to produce magnitude detected target signals, a method of processing the magnitude detected target signals to resolve targets corresponding to the magnitude detected target signals, said method comprising the steps of:

centroiding magnitude detected target signals to provide centroided target signals, wherein the magnitude detected target signals comprise a set of detections observed for each phase (slope segment) of the frequency modulated waveform;

associating the centroided target signals using a maximum likelihood estimation procedure to predict the filter position for the next phase, and wherein the associating step groups the detection from each phase by expecting hits to be within a gate size of a projected target doppler filter location, and wherein the projected doppler filter location and the gate size depend on all previous phase observations;

eliminating the number of targets to minimize the number of ghosts by reducing the number of candidate targets that correlate in a predetermined number of phases by grouping all paths with a selected number of common hits; and

computing velocity and range for the remaining detected targets.

2. The method of claim 1 wherein the associating step comprises:

associating the detected target filters from two phases with slopes having the same sign

establishing the gate size for the first association based upon desired minimum and maximum target ranges to minimize the number of potential associations in these two phases and also eliminate all targets at ranges longer than the maximum range.

3. The method of claim 1 which further comprises:

sorting the list of possible associated targets having two same sign slopes to provide the locations of the nearest targets ordered by their distance.

4. The method of claim 1 wherein the associating step comprises:

an N out of N detection rule.

5. The method of claim 1 wherein the associating step comprises:

a K out of N detection rule to provide target detection when the target return is predicted to fall in the region for which it has to compete with receiver leakage, mainlobe clutter, and mainlobe rain clutter in some of the FMR phases.

6. The method of claim 5 which further comprises:

skipping the phases in which the K out of N detection rule predicts the target return is masked by an interference signal.

7. The method of claim 1 wherein the centroiding step produces the target position within a filter of the FFT.

8. The method of claim 1 wherein the centroiding step groups target returns of adjacent filters for each phase to form discriminants that estimate target position and extent.

9. The method of claim 1 wherein the eliminating step minimizes the number of ghosts by selecting a target out of each group that is the one with the minimum estimation error.

10. The method of claim 9 wherein the eliminating step groups individual targets into a single target by grouping targets with the same doppler and a range difference of less than a specified length.

11. The method of claim 1 wherein the step or computing velocity and range for the detected targets provides target range (R_est) and velocity (R_est) estimates using the following formulas:

##EQU5## where c is the speed of light, λ

is the wavelength, N is equal to the number of associated slopes, ##EQU6## where ##EQU7## and where f_est.sup.(N+1) is the position of the phase return, ω

is the average of the target return, ω

_i is the slope value of target returns for slope i, ρ

is is the average of the slope value, N is the number of slopes used, s_i is the value of the slope i, s_i² is the square of the slope value, E_N (s²) is the second moment of the slope value, E_N (s) is the mean of the slope value, σ

_s² (N) is is the variance of the slope value, and s_N+1 is the slope value of the next phase.