Method of correcting for errors in radar imaging

• US 4,771,287 A
• Filed: 06/23/1983
• Issued: 09/13/1988
• Est. Priority Date: 06/23/1983
• Status: Expired due to Fees

• Current Assignee: Northrop Grumman Corporation
• Original Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Inventors: Mims, James H.
• Primary Examiner(s): Tarcza, Thomas H.
• Assistant Examiner(s): Sotomayor, John B.
• Application Number:
  US06/507,260
• Time in Patent Office: 1,909 Days
• Field of Search: 342/25, 342/66, 342/179, 342/191, 342/161, 342/63, 342/65, 342/120, 342/123, 342/194, 342/195, 342/189
• US Class Current: 342/25.00A
• CPC Class Codes: G01S 13/90 using synthetic aperture techniques , e.g. synthetic aperture radar [SAR] techniques
• CPC Class Codes: G01S 13/9019 Auto-focussing of the SAR signals

Abstract

A method of correcting for focus errors or higher order errors, or various combinations thereof, in the final image effected by the processing of a radar. With regard to correcting focus errors, the method coherently combines the signal focus data from selected target time histories derived by a first stage processor of the radar to form a complex signal representative of the average focus error of the radar image. From the phase of the complex signal is derived a set of focus error compensation signals which are used to compensate for focus errors in the target time histories derived by the first stage processor. With regard to the higher order error correction, the method coherently combines the signal data from selected target time histories derived by the first stage processor by shifting the phase and frequency centroids of each complex signal thereof to a common value. Light filtering may be provided to remove wide band noise with a linear phase filter. Accordingly, the coherently combined and filtered error correction signals are point wise inverted and thereafter used to correct the complex signals of the time histories derived by the first stage processor. An additional embodiment is provided to permit both the focus error and higher order error correction methods to be performed at an intermediate stage in the radar processing.

Claims

1. In a radar system including at least first and second stages for processing radar echo signals over a plurality of time samples of a single radar look, said radar echo signals being conditioned at each time sampling period by deramping in doppler and compressing into a plurality of range cells, said first stage for deriving from said conditioned radar echo signals at each time sample a set of complex signals of point-like targets according to a coarse azimuth resolution for each range cell of said plurality to form a time history of complex signals corresponding to said plurality of time samples for each point-like target, said second stage for deriving from said time histories of point-like targets a final azimuth resolution thereof to form a final radar image with improved resolution from a single radar look, a method of enhancing the focus of said final radar image by correcting focus errors developed in said time signal history data derived by said first stage prior to processing in said second stage, said method comprising the steps of:

coherently combining the complex signals of time histories selected from the time histories derived by said first stage to generate a set of correction signals corresponding to said time samples of said first stage derived time histories; and

correcting the time histories derived by said first stage with respectively corresponding correction signals of said set prior to being processed in said second stage.

2. The method in accordance with claim 1 wherein the step of coherently combining includes the steps of:

(a) combining the complex signal of a selected time history to form a first set of complex signals which are a measure of the instantaneous doppler frequencies of the corresponding time history data samples;

(b) combining the complex signals of said first set to form a second set of complex signals which are a measure of the instantaneous doppler frequency slopes of the corresponding time history data samples; and

(c) averaging the complex signals of said second set to form a third complex signal;

repeating steps (a) through (c) to form a third complex signal for each selected time history of complex signals; and

averaging said third complex signals to form a fourth complex signal; and

wherein the step of correcting each complex signal of the selected target time history includes the steps of;

forming a set of weighted unit vectors from said fourth complex signal; and

correcting each complex signal of the time histories derived by said first stage with its respectively corresponding weighted unit vector from said formed set.

3. The method in accordance with claim 1 wherein the step of coherently combining includes the steps of:

combining the complex signals of the time history of a selected point-like target of the set to form a first set of complex signals in accordance with the following expression;

##EQU8## where A_n and A_n+1 denote consecutive complex signal samples of the time history of the selected target, N denotes the total number of time samples, and B_n denotes a complex signal of the first set;

combining the complex signal B_n of the first set to form a second set of complex signals in accordance with the following expression;

##EQU9## where B_n and B_n+1 denote consecutive complex signal samples of the first set, and C_n denotes a complex signal of the second set; and

averaging the complex signal C_n of said second set to form a complex signal C_m representative of the measured focus error corresponding to the selected target m in accordance with the following expression;

##EQU10##

4. The method in accordance with claim 1 wherein the step of coherently combining includes the step of forming a complex signal, C=|C| exp (jQ"), representative of the measured focus error;

and wherein the step of correcting each complex signal of the target time histories includes the steps of;

forming a set of weighted unit vectors C_n from the phase angle Q" of said formed complex signal C in accordance with the following expression;

##EQU11## for n=1 to N; and

correcting each complex signal A_n of the time histories of complex signals N with its corresponding weighted unit vector C_n from said formed set in accordance with the following expression;

space="preserve" listing-type="equation">A'"'"'.sub.n =A.sub.n ·

C.sub.n for n=1 to N,

where A'"'"'_n denotes the focus error corrected complex signal of a time history.

5. In a radar system including at least first and second stages for processing radar echo signals over a plurality of time samples of a single radar look, said radar echo signals being conditioned at each time sampling period by deramping in doppler and compressing into a plurality of range cells, said first stage for deriving from said conditioned radar echo signals at each time sample a corresponding set of complex signals of point-like targets according to a coarse azimuth resolution for each range cell of said plurality to form a time history of complex signals, corresponding to said plurality of time samples, for each point-like target, said second stage for deriving from said time histories of point-like targets a final azimuth resolution thereof to form a final radar image with improved resolution from a single radar look, a method of improving the final radar image by correcting for higher order errors developed in said time signal history data derived by said first stage prior to processing in said second stage, said method comprising the steps of:

selecting a first subset of complex signal target time histories from said time histories derived by said first stage;

deriving first and second complex signals for each target time history of said first subset using the complex signals respectively corresponding thereto, said first and second complex signals being representative respectively of the average doppler frequency and average doppler phase of their corresponding target time histories;

adjusting the complex signals of said selected time histories as a function of their correspondingly derived first and second complex signals to form a second subset of frequency and phase adjusted complex signal time histories;

coherently combining the complex signals of said second subset in accordance with common time samples of the time histories thereof to form a single time history of combined complex signals;

generating a time history of error correction signals as a function of the corresponding time samples of said combined complex signal time history; and

correcting the complex signals of the time histories derived by said first stage with corresponding signals of said derived time history of error correction signals.

6. The method in accordance with claim 5 wherein the steps of deriving and adjusting include the steps of:

(a) combining the complex signals of the time history of a selected target to derive a first complex signal representative of the average doppler frequency corresponding to said selected target;

(b) adjusting the complex signals of the selected time history as a function of said derived first complex signal associated therewith to form a corresponding time history of frequency adjusted complex signals;

(c) averaging the frequency adjusted complex signals of the time history of said selected target to derive a second complex signal representative of the average doppler phase corresponding to said selected target;

(d) adjusting the frequency adjusted complex signals of the selected time history as a function of said derived second complex signal associated therewith to form a corresponding time history of frequency and phase adjusted complex signals; and

(e) repeating steps (a) through (d) for each target of said first subset to from a second subset of frequency and phase adjusted complex signal target time histories.

7. The method in accordance with claim 6 wherein step (a) includes the steps of:

combining the complex signals A_n of the time history of N complex signals of a selected target to form a second time history of complex signals B_n in accordance with the following expression;

##EQU12## where A_n and A_n+1 denote consecutive complex signal samples of the time history; and

averaging the complex signals B_n of the second time history to derive the first complex signal B in accordance with the following expression;

##EQU13## wherein step (b) includes the steps of;

forming a weighted time history of complex signals exp (-jndQ) for n=1 to N from the normalized conjugated first complex signal B; and

adjusting the complex signals A_n with their corresponding complex signal exp (-jndQ) for n=1 to N to form a time history of N frequency adjusted complex signals C_n ;

wherein step (c) includes the steps of;

averaging the complex signals C_n to derive the second complex signal C in accordance with the following expression;

##EQU14## wherein step (d) includes the steps of;

adjusting the complex signals C_n with the inverted (conjugate) of the second complex signal C to form a time history of N frequency and phase adjusted complex signals D_n.

8. The method in accordance with claim 5 wherein the step of coherently combining includes the step of coherently averaging the complex signals of the second subset corresponding to a common time sample for each time sample of the time histories thereof to form a single time history of averaged complex signals.

9. The method in accordance with claim 8 wherein the step of coherently combining includes the step of coherently averaging the complex signals D_n of the M selected targets of the second subset at a common time sample n for each time sample from 1 to N to form a single time history of averaged complex signals F_n in accordance with the following expression:

##EQU15##

10. The method in accordance with claim 5 wherein the steps of generating and correcting includes the steps of:

filtering the time history of averaged complex signals to form a corresponding time history of filtered complex signals;

inverting the filtered complex signals to form a corresponding time history of error correction signals; and

correcting the complex signals of the time histories derived by the first stage with the corresponding error correction signals of the inverted time history.

11. The method in accordance with claim 10 wherein the step of filtering includes the step of filtering the time history of averaged complex signals F_n with a K point filter to form a corresponding time history of N filtered complex signals G_n in accordance with the following expressions:

##EQU16## where H_k,j denotes the kth filter weight of the jth filter weight set, and [K/2] denotes the integer part of [K/2].

12. In a radar system including at least first and second stages for processing radar echo signals over a plurality of time samples of a single radar look, said radar echo signals being conditioned at each time sampling period by deramping in doppler and compressing into a plurality of range cells, said first stage for deriving from said conditional radar echo signals at each time sample a corresponding set of complex signals of point-like targets according to a coarse azimuth resolution for each range cell of said plurality to form a time history of complex signals, corresponding to said plurality of time samples, for each point-like target, said second stage for deriving from said time histories of point-like targets a final azimuth resolution thereof to form a final radar image with improved resolution from a single radar look, a method of improving the final radar image by measuring and correcting for focus and higher order errors developed in said time signal history data derived by said first stage prior to processing in said second stage, said method comprising the steps of:

selecting a first subset of complex signal target time histories from said time histories derived by said first stage;

coherently combining the complex signals of a time history of said first subset for each time history thereof to generate a set of focus correction signals corresponding to the time samples of said selected time histories;

correcting the complex signals of the time histories of said first subset with respectively corresponding focus correction signals of said set to form a second subset of focus corrected selected target time histories;

deriving first and second complex signals for each target time history of said second subset using the complex signals respectively corresponding thereto, said first and second signals being representative respectively of the average doppler frequency and average doppler phase of their corresponding target time histories;

adjusting the complex signals of the time histories of said second subset as a function of their correspondingly derived first and second complex signals to form a third subset of frequency and phase adjusted complex signal time histories;

coherently combining the complex signals of said third subset in accordance with common time samples of the time histories thereof to form a single time history of combined complex signals;

generating a time history of higher order error correction signals as a function of the corresponding time samples of said combined complex signal time history; and

correcting the complex signals of the time histories derived by said first stage with corresponding signals of both of said focus error and higher order error correction signal time histories.