Communications: directive radio wave systems and devices (e.g. – Directive – Beacon or receiver
Reexamination Certificate
2003-08-21
2004-09-14
Blum, Theodore M. (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Directive
Beacon or receiver
C342S156000
Reexamination Certificate
active
06791493
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to locating frequency agile emitters using RF interferometers, and more specifically, the present invention relates to using a long baseline interferometer (LBI) to make ambiguous and biased signal direction-of-arrival (DOA) phase measurements in a sequence of receiver dwells, possibly all at different signal frequencies, and to locate the emitter by forming phase change estimates by taking differences of the phase measurements between these dwells.
BACKGROUND OF THE INVENTION
FIG. 1
illustrates the relationship between interferometer baseline, emitter signal angle of arrival, emitter direction of arrival unit vector, and receiver phase measurement fundamental to the understanding of the present invention.
FIG. 1
relates LBI phase measurement to emitter DOA. The LBI baseline
102
is created by antennas
100
and
101
. The true phase
104
is the vector dot product of the emitter direction-of-arrival unit vector
106
onto the LBI baseline
102
scaled by the emitter RF carrier frequency f and speed-of-light c, as described by Equation 1:
φ
=
f
c
d
→
·
u
→
(
1
)
However, phase measurement
105
is only measured modulo one cycle by the phase detector
107
, i.e., as described by Equation 2:
φ
m
=
φmod
(
1
cycle
)
=
f
c
d
→
·
u
→
-
n
(
2
)
where the integer n is the number of cycles subtracted from the true phase so that the measured phase satisfies the inequality:
-
1
2
<
φ
m
≤
1
2
Determining n to recover the relationship in Equation 1 from Equation 2 is called resolving the phase measurement.
For long baseline interferometers, the inability to robustly and economically calibrate the large cable runs
109
from antennae
100
,
101
to receiver means a large unknown bias error
110
is typically present in the LBI phase measurement. But it is well established that accurate emitter range estimation requires only precise measurement of the emitter bearing rate-of-change. See for example, A. L. Haywood “Passive Ranging by Phase-Rate Techniques” (Wright-Patterson AFB Tech. Report ASD-TR-70-46 December 1970). In practice, discrete time phase differences rather than phase rates are used, i.e., as described by Equation 3:
φ
m
(
t
2
)
-
φ
m
(
t
1
)
=
f
c
d
→
2
·
u
→
2
-
f
c
d
→
1
·
u
→
1
+
n
1
-
n
2
(
3
)
Here, the phase differences are determined based on measurements typically made one second or more apart.
Using the phase difference, rather than phase, to locate the emitter means the bias errors
110
on the individual measurement from antennae
100
,
101
cancel. It also means differential, and not absolute, phase ambiguity resolution of the LBI baseline is required. Kaplan, in “Passive Ranging Method and Apparatus”, U.S. Pat. No. 4,734,702, described how to resolve the differential phase measurement ambiguity m=n
2
−n
1
utilizing a short baseline interferometer (SBI). The SBI can be a planar interferometer that measures {right arrow over (u)}
sbi
. Then, using the procedure outlined in Kaplan, the SBI DOA unit vectors measured at two different points in time, times 1 and 2, are dotted onto the LBI baseline to predict the LBI unambiguous phase and allow m to be found according to Equation 4:
m
=
n
1
-
n
2
=
nint
(
φ
m
(
t
2
)
-
φ
m
(
t
1
)
-
f
c
d
→
2
·
u
→
sbi2
+
f
c
d
→
1
·
u
→
sbi1
)
(
4
)
where nint is the nearest integer function.
FIG. 2
depicts a realization of the SBI/LBI method using only a linear SBI
200
. A linear or one-dimensional interferometer measures AOA (
108
,
FIG. 1
) not {right arrow over (u)}
sbi
(equivalent to
106
, FIG.
1
). In this case, Equation 4 is utilized in a special sensor-oriented coordinate system, such as the ijk set
103
depicted in FIG.
1
. In such sensor coordinates, Equation 4 becomes Equation 5:
n
1
-
n
2
=
nint
(
φ
m
(
t
2
)
-
φ
m
(
t
1
)
-
f
2
c
L
cos
(
AOA
)
sbi2
+
f
1
c
L
cos
(
AOA
)
sbi1
)
(
5
)
where the obvious modification to handle different frequencies f
1
and f
2
at each phase measurement has also been incorporated, and L (
FIG. 1
,
111
) is the baseline length. The processing indicated by Equation 5 occurs in process step
202
, (FIG.
2
). The ambiguity integer is added at step
203
to the measured ambiguous phase determined at step
210
, and ambiguous phase change related to emitter range is resolved at step
204
, allowing the emitter location to be determined and output, i.e. emitter range and bearing found.
As Equation 1 indicates, to the first order, the resolved LBI phase difference may contain components due to baseline motion and frequency change, as well as phase change due to motion relative to the emitter as described by Equation 6: The component containing emitter-range information, and hence generating the phase change specifically considered by Haywood and Kaplan, is given by the second term, i.e., as described by Equation 7:
Δφ
range
=
f
c
d
→
·
Δ
u
→
(
7
)
This relation forms the basis for the step
204
processing in conventional SBI/LBI implementations used to locate frequency stable, or non-frequency agile emitters. Against frequency agile emitters, perturbations to the phase due to frequency changes must be accounted for before employing Equation (7).
As shown by the third term in Equation 6, emitter frequency agility alters the &Dgr; phase-range relation by introducing a frequency-change and DOA dependent factor. But, as Equation 5 demonstrates, this does not create a problem in SBI/LBI ambiguity resolution if the SBI itself can be resolved. However, SBI design techniques, e.g. as described by Robert L. Goodwin, in “Ambiguity-Resistant Three and Four-Channel Interferometers”, (Naval Research Laboratory, Washington, D.C. Report 8005, Sep. 9, 1976), typically assume the same RF carrier frequency for all pulses used to estimate the phase ambiguities and generate COS(AOA). This assumption is almost always valid if the phase across all baselines is measured on the same pulse. Emitter frequency may change intrapulse, e.g. chirped signals, but this is comparatively rare. However, monopulse measurements require a separate receiver pair and phase detector for each SBI baseline. These systems are expensive in terms of both weight and cost and do not exploit the fact that the LBI measurement requires only a single, two channel system,
205
(FIG.
2
).
Such a two channel system used with an SBI requires baseline switching, e.g., using an RF switch
201
to connect a single pair of receivers and phase detector
211
sequentially between SBI interferometer antennae
206
. Frequency is “simultaneously” obtained by the instantaneous frequency measurement (IFM) module
207
. In this method, the minimum number of pulses collected for a single emitter equals the number of interferometer baselines. But against pulse-to-pulse agile emitters, the IFM measures a different frequency for each pulse. Then, when employing conventional processing, such as described by Goodwin, the SBI ability to resolve phase ambiguities at process step
208
and subsequently use the resolved phase ambiguities to estimate COS(AOA) at step
209
totally fails. Hence, this most desirable two-channel SBI/LBI implementation cannot be used against frequency agile emitters.
Denton, in “Exploitation of Emitter RF Agility for Unambiguous Interferometer Direction Finding”, U.S. Pat. No. 5,652,590 (hereinafter referred to as the '590 patent), has demonstrated a way, specific to frequency agile signals, to overcome the above-described two-channel-system drawback by not using an SBI to estimate the COS(AOA). Denton considers the specialization of Equation 1 to sensor coordinates, e.g. ijk,
103
(FIG.
1
), as de
Blum Theodore M.
Lowe Hauptman GIlman & Berner, LLP
Northrop Grumman Corporation
LandOfFree
Method for using signal frequency change to differentially... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for using signal frequency change to differentially..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for using signal frequency change to differentially... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3214425