FM/CW radar linearization network and method therefor

• US 4,539,565 A
• Filed: 08/16/1982
• Issued: 09/03/1985
• Est. Priority Date: 08/16/1982
• Status: Expired due to Fees

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Norsworthy, Keith H.
• Primary Examiner(s): Tubbesing, T. H.
• Application Number:
  US06/408,251
• Time in Patent Office: 1,114 Days
• Field of Search: 343/5 NQ, 343/14, 343/17.5, 343/17.7
• US Class Current: 342/128
• CPC Class Codes: G01S 13/343 using sawtooth modulation
• CPC Class Codes: G01S 7/4008 of transmitters
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

An FM/CW radar linearization network provides target identification data discriminating a target from background reflections and/or false targets by compensating for random variations in the linearity of the frequency sweep of the radar transmitter in the processing of the radar receiver signal, the latter being characteristic of the range and physical size of the target. Linearization is achieved by sampling the transmitter signal, generating an error signal proportional to and indicative of any deviations in the frequency sweep of the transmitter from that for linear operation, and applying that error signal to correct the target data signal, thereby frequency normalizing the same for processing and utilization. A number of alternatives in respect of the development of an appropriate error signal and further alternatives in its use as the basis for data correction are disclosed, including sampling of the transmitter in respect of the phase angle over a predetermined period or the time for a predetermined number of cycles, in order to develop an error signal proportional to and indicative of any deviations in the frequency sweep of the transmitter from that were its operation linear; while data correction may be made in one of a number of alternative ways, including phase rotation of the raw target data input signal, time-shifting of the sampling rate of such data, or frequency mixing the raw target data with an error control signal. Representative networks and suitable methodologies to achieve linearization are disclosed herein.

Claims

1. An FM/CW radar linearization network compensating for random variations in the linearity of the frequency sweep of the radar transmitter in the processing of the radar receiver signal characteristic of the range and physical size of a target to provide target identification data discriminating said target from background reflections, said network comprising:

a. raw target data generating means for developing a target data signal containing frequency un-normalized target identification information, wherein the instantaneous frequency of said target data signal is characteristic of the range of a target and the amplitude of said target data signal is characteristic of the size of said target at that range;

b. transmitter characteristic data generating means for developing a sampling signal containing transmitter frequency sweep information;

c. error data generating means receiving said sampling signal for developing an error signal characteristic of any nonlinearity in the sweep of the transmitter frequency; and

,

d. linearization means receiving said raw target data signal and said error signal for frequency normalizing target identification information by compensation for any random variation in the linearity of the sweep of transmitter frequency.

2. The linearization network of claim 1, wherein said raw target data generating means comprises a transmitter/receiver frequency comparator means for developing a target data frequency signal characteristic of the size and range of said target.

3. The linearization network of claim 2, wherein said transmitter/receiver frequency comparator means is comprised of a first transmitter/receiver frequency comparator means for developing a first target data frequency signal characteristic of the size and range of said target, said network further comprising second transmitter/receiver frequency comparator means for developing a second target data frequency signal characteristic of the size and range of said target rotated from said first target data frequency signal through a predetermined rotation angle.

4. The linearization network of claim 3, wherein said error data generating means comprises phase angle comparator means for developing a phase error signal characteristic of any phase angle deviation in the sweep of transmitter frequency resulting from random nonlinear variations thereof.

5. The linearization network of claim 4, wherein said linearization means comprises phase angle rotation means receiving said first and second target data frequency signals and said phase error signal, for rotating the phase angle between said target data frequency signals by a value proportional to said phase error signal to frequency-normalize said target data signal by compensation for random nonlinear variations in said transmitter frequency sweep and developing a normalized target signal.

6. The linearization network of claim 5, wherein said network further comprises target data processing means receiving said normalized target signal for frequency discrimination thereof into a plurality of frequency bands characteristic of the size and range of said target.

7. The linearization network of claim 6, wherein said target data processing means comprises filter means for discriminating said normalized target signal into a plurality of frequency bands, wherein each frequency band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

8. The linearization network of claim 6, wherein said phase angle rotation means receives a first raw target data signal I(t) and a second raw target data signal Q(t) identical to said first signal rotated through an angle of 90°

and further wherein the vector representation of said raw target data signals (I(t), Q(t)) rotates through a phase angle φ

;

said phase angle rotation means comprising a sin/cos ROM element receiving said phase error signal and means for correcting said raw target data and frequency normalizing the same in response thereto to yield a normalized target data signal (I'"'"'(t), Q'"'"'(t)) in accordance with the relationships;

space="preserve" listing-type="equation">I'"'"'(t)=cos φ

I(t)+sin φ

Q(t),

space="preserve" listing-type="equation">Q'"'"'(t)=sin φ

I(t)-cos φ

Q(t).

9. The linearization network of claim 1, wherein said error data generating means comprises phase angle comparator means for developing a phase error signal characteristic of any phase angle deviation in the sweep of transmitter frequency resulting from random nonlinear variations thereof.

10. The linearization network of claims 1 or 2, wherein said error data generating means comprises transmitter signal time comparator means for developing a time error signal characteristic of any time deviation of said transmitter signal over a predetermined cycle sample count resulting from random nonlinear variations in said transmitter frequency sweep as compared with the period for the same number of cycles were the frequency sweep linear.

11. The linearization network of claim 10, wherein said linearization means comprises raw target data signal time sampling means receiving said raw target data signal and said time error signal, wherein the sampling time of said signal time sampling means is advanced or retarded by a value proportional to said time error signal to normalize said target data signal by compensation for random nonlinear variations in said transmitter frequency sweep and develop a normalized target signal.

12. The linearization network of claim 9, wherein said network further comprises target data processing means receiving said normalized target signal for frequency discrimination thereof into a plurality of frequency bands characteristic of the size and range of said target.

13. The linearization network of claim 10, wherein said target data processing means comprises filter means for discriminating said normalized target signal into a plurality of frequency bands, wherein each frequency band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

14. The linearization network of claim 10, wherein said linearization means comprises an analog-to-digital converter.

15. The linearization network of claims 1 or 2, wherein said error data generating means comprises a voltage controlled oscillator for developing an error signal, the frequency of which varies proportionately to any nonlinearity in the sweep of said transmitter frequency.

16. The linearization network of claim 15, wherein:

a. said raw target data means includes a target data mixer means for developing said target data signal; and

,

b. said linearization means comprises a linearization mixer receiving said target data signal and said error signal from said oscillator.

17. The linearization network of claim 16, further comprising target data processing means receiving a normalized target data signal from said linearization mixer for frequency discrimination thereof into a plurality of frequency bands characteristic of the size and range of said target.

18. The linearization network of claim 17, wherein said target data processing means comprises filter means for discriminating said normalized target signal into a plurality of frequency bands, wherein the frequency of each band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

19. An FM/CW radar linearization network compensating for random variations in the linearity of the frequency sweep of the radar transmitter in the processing of the radar receiver signal characteristic of the range and physical size of a target to provide target identification data discriminating said target from background reflections, said network comprising:

a. first transmitter/receiver frequency comparator means for developing a first target data signal I(t), the instantaneous frequency of which is characteristic of the range of said target and the amplitude of which is characteristic of the size of said target at that range;

b. second transmitter/receiver frequency comparator means for developing a second target data signal Q(t) identical to said first target data signal rotated through a 90°

angle;

c. transmitter characteristic data generating means for developing a sampling signal characteristic of the transmitter sweep frequency;

d. phase counter means receiving said sampling signal for developing a measured phase signal representative of the phase of said transmitter frequency at preselected sampling times;

e. signal generating means for developing a comparison signal characteristic of the phase of a transmitter operating under linear frequency sweep conditions at said preselected times;

f. phase error comparator means receiving said measured phase signal and said comparison signal for developing a phase error signal characteristic of any phase error in said transmitter frequency sweep resulting from nonlinear operation thereof; and

,

g. phase angle rotation means receiving said first and second target data signals and said phase error signal for rotating said first and second target data signals, I(t) and Q(t), respectively, by an error angle represented by said phase error signal to yield frequency normalized target data signals I'"'"'(t) and Q'"'"'(t), respectively, in accordance with the relationships;

space="preserve" listing-type="equation">I'"'"'(t)=cos φ

I(t)+sin φ

Q(t),

space="preserve" listing-type="equation">Q'"'"'(t)=sin φ

I(t)-cos φ

Q(t).

20. In an FM/CW radar system, wherein a transmitter signal and a receiver signal are mixed to yield a target data signal having a frequency spectrum the instantaneous frequency of which is representative of the range of a target and the amplitude of which at said instantaneous frequency is representative of the size of said target at said range, wherein said system is subject to operation over a random nonlinear sweep of transmitter frequency, the improvement comprising a linearization network compensating for nonlinearity, said network comprising:

a. transmitter/receiver mixer means for developing a complex, two channel target data signal represented vectorially as (I(t)_n, Q(t)_n) for times t_n, wherein the vector rotates at times t₁, t₂, t₃ . . . at a constant angular rate under linear transmitter sweep frequency conditions;

b. phase error means receiving a sampling signal characteristic of the linearity of said transmitter sweep for developing a phase error signal proportional to any nonlinearity in sweep frequency at times t_n ; and

,

c. phase rotation means receiving said target data signal and said phase error signal for rotating said vector by an amount proportional to any phase error to maintain said angular rate constant.

21. The linearization network of claim 20, wherein said phase error means is comprised of a phase counter for detecting the phase of said transmitter frequency over a predetermined time period and developing a measured phase signal characteristic thereof, a phase comparison signal generator for developing a comparison signal characteristic of the phase of a transmitter operating under linear sweep conditions and a comparator receiving as inputs said measured phase signal and said phase comparison signal for developing a phase error signal proportional to the difference between said inputs.

22. The linearization network of claims 20 or 21, wherein said transmitter/receiver mixer means is comprised of a first transmitter/receiver mixer receiving transmitter and receiver signals and developing a first target data signal and a second transmitter/receiver mixer receiving transmitter and receiver signals, one of which is phase shifted in a phase shifting circuit to develop a second target data signal, rotated 90°

from said first target data signal.

23. The linearization network of claim 22, further comprising a transmitter mixer receiving said transmitter signal and a standard frequency signal for developing a low-frequency transmitter characteristic data signal characteristic of the sweep frequency of said transmitter.

24. The linearization network of claims 20 or 21, wherein said transmitter/receiver mixer means is comprised of a mixer network including a first transmitter/receiver mixer receiving transmitter and receiver signals and developing a first target data signal, a second transmitter/receiver mixer receiving said first target data signal and a standard frequency signal for developing a first low-frequency target data signal, and a third transmitter/receiver mixer receiving said first target data signal and said standard frequency signal, one of which is phase shifted, to develop a second low-frequency target data signal rotated from said first low-frequency data signal by 90°

.

25. The linearization network of claim 24, further comprising a transmitter mixer receiving said transmitter signal and a standard frequency signal for developing a low-frequency transmitter characteristic data signal characteristic of the sweep frequency of said transmitter.

26. In an FM/CW radar system, wherein a transmitter signal and a receiver signal are mixed to yield a target data signal having a frequency spectrum the instantaneous frequency of which is representative of the range of a target and the amplitude of which at said instantaneous frequency is representative of the size of said target at said range, wherein said system is subject to operation over a random nonlinear sweep of transmitter frequency, the improvement comprising a linearization network compensating for nonlinearity, said network comprising:

a. transmitter/receiver mixer means for developing a raw target data signal;

b. transmitter characteristic data mixer means receiving said transmitter signal and a standard frequency signal for developing a transmitter characteristic signal representative of the transmitter sweep;

c. counter means receiving said transmitter characteristic signal for counting a predetermined number of cycles thereof and developing a measured time signal proportional to the time period of said predetermined number of cycles;

d. comparison signal means for developing a time comparison signal proportional to the time expected for said predetermined number of cycles were the transmitter frequency sweep linear;

e. comparator means receiving the measured and expected time signals and developing a time error signal proportional to the difference thereof;

f. phase converter means receiving said time error signal and a signal proportional to the instantaneous frequency expected for linear operation, for developing a phase error signal representative of any nonlinearity in the sweep of transmitter frequency;

g. converter means for converting said phase error signal to a sample interval control signal;

h. linearizing means receiving said raw target data signal and said sample interval control signal, wherein the latter adjusts the sampling time of said raw data signal by a time proportional to any error due to nonlinearity in said sweep, thereby normalizing said target data signal; and

,

i. processing means receiving the normalized target data signal for discriminating said normalized signal into a plurality of frequency bands, wherein the instantaneous frequency of each band is characteristic of the range of said target and the amplitude at said instantaneous frequency is characteristic of the size of said target at said range.

27. The linearization network of claim 26, wherein said linearizing means is an analog-to-digital converter, the sampling rate of which is adjusted in response to said interval control signal.

28. An FM/CW radar linearization network compensating for random variations in the linearity of the frequency sweep of the radar transmitter in the processing of the radar receiver signal characteristic of the range and physical size of a target to provide target identification data discriminating said target from background reflections, said network comprising:

a. a transmitter/receiver mixer means receiving transmitter and receiver signals and developing a raw target data signal as an output thereof;

b. a transmitter characteristic mixer means receiving said transmitter signal and a reference signal and developing a transmitter characteristic signal as an output thereof;

c. an error data generating means receiving said transmitter characteristic signal and developing an error signal characteristic of any nonlinearity in the sweep of the transmitter frequency;

d. a voltage controlled oscillator receiving said error signal and developing a linearizing signal as an output thereof, having a frequency variation proportional to any nonlinearity in said sweep; and

,

e. a linearizing mixer receiving said raw target data signal and said linearizing signal for developing a frequency-normalized target data signal as an output thereof.

29. The linearization network of claim 28, further comprising target data processing means receiving said normalized target data signal for frequency discrimination thereof into a plurality of frequency bands.

30. The linearization network of claim 29, wherein said target data processing means comprises filter means for discriminating said normalized target signal into a plurality of frequency bands, wherein the frequency of each band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

31. A method for compensating for random variations in the linearity of the sweep of transmitter frequency of an FM/CW radar system in the processing of the receiver signal characteristic of the range and physical size of a target and thereby linearizing said signal to provide target identification data discriminating said target from background reflections, said method comprising the steps of:

a. developing a raw target data signal containing frequency un-normalized target identification information, wherein the instantaneous frequency of said raw target data signal is characteristic of the range of a target and the amplitude of said raw target data signal is characteristic of the physical size of said target at that range;

b. sampling the transmitter signal and developing a sampling signal containing transmitter frequency sweep information;

c. generating an error signal in response to said sampling signal characteristic of any nonlinearity in the sweep of said transmitter frequency; and

,

d. linearizing said raw target data signal in response to said error signal to frequency normalize said target data signal by proportionate compensation for any random variation in the linearity of the frequency sweep of said transmitter.

32. The method of claim 31, wherein said step of sampling said transmitter signal and generating an error signal in response thereto comprises the steps of:

a. determining the phase of the transmitter signal over a predetermined time period to develop a measured phase signal;

b. generating a comparison phase signal characteristic of the phase of a transmitter over said predetermined period were said transmitter frequency swept linearly; and

,

c. comparing said measured phase signal with said comparison phase signal to develop a phase error signal.

33. The method of claim 31, wherein said steps of sampling said transmitter signal and generating an error signal in response thereto comprises the steps of:

a. counting a preselected number of cycles of said transmitter signal and developing a measured time signal representative of the time period therefor;

b. generating a comparison time signal characteristic of the time for said preselected number of cycles were the transmitter frequency swept linearly; and

,

c. comparing said measured time signal with said comparison time signal to develop a time error signal.

34. The method of claims 31, 32 or 33, wherein said step of developing a raw target data signal and linearizing the same comprise the steps of:

a. developing a first raw target data signal I(t);

b. developing a second raw target data signal Q(t) rotated 90°

from said first signal, wherein the phase angle φ

between vectors (I(t)_n, Q(t)_n) rotates in vector space I(t), Q(t); and

,

c. applying said error signal to rotate (I(t)_n, Q(t)_n) to a corrected signal (I'"'"'(t)_n, Q'"'"'(t)_n) in accordance with;

space="preserve" listing-type="equation">I'"'"'(t)=cos φ

I(t)+sin φ

Q(t),

space="preserve" listing-type="equation">Q'"'"'(t)=sin φ

I(t)-cos φ

Q(t);

wherein (I'"'"'(t)_n, Q'"'"'(t)_n) represents target data were the transmitter frequency swept linearly.

35. The method of claim 34, further comprising the step of discriminating the frequency-normalized target data signal into a plurality of frequency bands, the frequency of each of which is indicative of the range of said target and the amplitude of which is indicative of the physical size of said target at that range.

36. The methods of claims 31, 32 or 33, wherein said step of linearizing said raw target data comprises the steps of:

a. time-sampling said raw target data; and

,

b. applying said error signal to shift the time of time-sampling to that were the transmitter frequency swept linearly.

37. The method of claim 36, further comprising the step of discriminating the frequency-normalized target data signal into a plurality of frequency bands, the frequency of each of which is indicative of the range of said target and the amplitude of which is indicative of the physical size of said target at that range.

38. The method of claim 31, wherein said raw target data signal is applied as a first input to a linearizing mixer means, said method further comprising the steps of:

a. applying said error signal to a voltage controlled oscillator to develop a linearizing signal having a frequency variation characteristic of any nonlinearity in the sweep of said transmitter frequency; and

,

b. applying said linearizing signal as a second input to said linearizing mixer means to frequency-normalize said raw target data signal.

39. The method of claim 31, wherein said step of generating said error signal comprises the steps of:

a. counting a preselected number of cycles of said sampling signal and measuring the time period thereof to develop a measured time signal; and

,

b. comparing said measured time signal with a generated time signal characteristic of the time period for an identical number of cycles for a transmitter operating under linear sweep conditions.

40. The method of claim 31, wherein said step of generating said error signal comprises the steps of:

a. determining the phase of said sampling signal over a predetermined time period to generate a measured phase signal; and

,

b. comparing said measured phase signal with a generated phase signal characteristic of the phase over the identical time period were the transmitter sweep linear, to develop a phase error signal.

41. The method of claims 31, 39 or 40, further comprising the steps of:

a. applying said raw target data signal to a phase rotator; and

,

b. applying said error signal to said phase rotator to linearize said raw target data signal.

42. The method of claim 41, further comprising the step of discriminating the frequency-normalized target data signal into a plurality of frequency bands, the frequency of each of which is indicative of the range of said target and the amplitude of which is indicative of the physical size of said target at that range.

43. The method of claims 31, 39 or 40, further comprising the steps of:

a. applying said raw target data signal to a time-sampling converter; and

,

b. applying said error signal to adjust the time of sampling of said raw target data to that were the transmitter operating under linear conditions.

44. The method of claim 43, further comprising the step of discriminating the frequency-normalized target data signal into a plurality of frequency bands, the frequency of each of which is indicative of the range of said target and the amplitude of which is indicative of the physical size of said target at that range.

45. The method of claims 31, 32, 33, 38, 39, or 40, further comprising the step of discriminating the frequency-normalized target data signal into a plurality of frequency bands, the frequency of each of which is indicative of the range of said target and the amplitude of which is indicative of the physical size of said target at that range.

46. A method for compensating for random variations in the linearity of the sweep of transmitter frequency of an FM/CW radar system in the processing of the receiver signal characteristic of range and physical size of a target and thereby linearizing said signal to provide target identification data discriminating said target from background reflections, said method comprising the steps of:

a. mixing the transmitter and receiver signals in a first mixer means to develop a first raw target data signal I(t);

b. mixing said transmitter and receiver signals in a second mixer means to develop a second raw target data signal Q(t) rotated from said first signal by 90°

, and wherein the phase angle φ

between vectors (I(t)_n, Q(t)_n) rotates in vector space I(t), Q(t);

c. applying said raw target data signals to a phase rotator;

d. mixing said transmitter signal with a standard reference frequency signal to develop a transmitter characteristic signal;

e. applying said transmitter characteristic signal to a phase detector means to develop a measured phase signal representative of the phase of said transmitter frequency over a predetermined time period;

f. comparing said measured phase signal with a generated phase signal characteristic of the phase of a transmitter operating under linear frequency sweep conditions over said predetermined period, to develop a phase error signal indicative of and proportional to any nonlinear operation of said transmitter;

g. applying said phase error signal to said phase rotator to rotate said data signals I(t), Q(t) to yield frequency-normalized data signals I'"'"'(t), Q'"'"'(t) in accordance with the relationships;

space="preserve" listing-type="equation">I'"'"'(t)=cos φ

I(t)+sin φ

Q(t),

space="preserve" listing-type="equation">Q'"'"'(t)=sin φ

I(t)-cos φ

Q(t); and

,

h. processing said frequency-normalized signals in a filter network for discriminating said signals into a plurality of frequency bands, wherein the frequency of each band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

47. The method of claim 46, wherein said steps of mixing said transmitter and receiver signals include the steps of:

a. mixing said transmitter and receiver signals in a high frequency mixer means to develop a raw target data signal;

b. applying said raw target data signal to first and second frequency-reduction mixer means; and

,

c. applying a standard frequency signal as a second input to each of said frequency-reduction mixer means, wherein one of said inputs is rotated by 90°

, to develop said first and second raw target data signals I(t) and Q(t), respectively.

49. The methods of claims 46, 47 or 48, wherein said error signal is processed by applying same to a time delay line equivalent to the propagation time of said transmitter signal and then comparing the delayed signal with itself to yield a processed error signal.

48. A method for compensation for random variations in the linearity of the sweep of transmitter frequency of an FM/CW radar system in the processing of the receiver signal characteristic of the range and physical size of a target and thereby linearizing said signal to provide target identification data discriminating said target from background reflections, said method comprising the steps of:

a. mixing transmitter and receiver signals in a mixer means to develop a raw target data signal;

b. applying said raw target data signal to a time-sampling converter means;

c. mixing said transmitter signal with a standard frequency signal to develop a transmitter characteristic signal;

d. counting a preselected number of cycles of said transmitter characteristic signal and measuring the time period therefor to develop a measured time signal;

e. comparing said measured time signal with a generated time signal characteristic of the time for the identical number of cycles for a transmitter operating under linear frequency sweep conditions, to develop a time error signal;

f. applying said error signal and said generated time signal as inputs to a phase converter means to generate a phase error signal;

g. converting said phase signal into a time-sampling signal;

h. applying said time-sampling signal to said time-sampling converter to alter the sampling time of said raw data signal to conform to that were the transmitter frequency swept linearly to yield a frequency-normalized target data signal; and

,

i. processing said frequency-normalized signal in a filter network for discriminating said signal into a plurality of frequency bands, wherein the frequency of each band is characteristic of the range of said target and the signal amplitude within each band is characteristic of the size of said target.

50. A method for compensating for random variations in the linearity of the sweep of transmitter frequency of an FM/CW radar system in the processing of the receiver signal characteristic of the range and physical size of a target and thereby linearizing said signal to provide target identification data discriminating said target from background reflections, said method comprising the steps of:

a. applying transmitter and receiver signals to a first mixer means to develop a raw target data signal containing frequency un-normalized target identification information;

b. applying said raw target data signal as a first input to a linearizing mixer means;

c. mixing said transmitter signal with a standard frequency signal to develop a transmitter characteristic signal of reduced frequency containing transmitter frequency sweep information;

d. generating an error signal in response to said transmitter characteristic signal, characteristic of any nonlinearity of the sweep of said transmitter frequency;

e. applying said error signal as an input to a voltage controlled oscillator to develop a correction signal having a frequency variation characteristic of and proportional to any nonlinearity in the sweep of said transmitter frequency;

f. applying said correction signal as a second input to said linearizing mixer to develop a frequency-normalized target data signal; and

,

g. applying said normalized target data signal to a processor means wherein said signal is discriminated into a plurality of frequency bands, the frequency of each band being indicative of the range of said target and the amplitude of said signal within each band being indicative of the physical size of said target at that range.