 |
Claims  |
|
|
What is claimed is:
1. A method of determining a set of a plurality of signal parameters for
use in an adaptive filter signal processor to process a plurality of input
signals to generate a plurality of processed signals, said method
comprising:
a) generating a fixed plurality of sets;
b) storing said fixed plurality of sets;
c) generating a plurality of cumulative performance values, with a
cumulative performance value corresponding to each of said fixed plurality
of sets;
d) comparing said plurality of cumulative performance values,
e) choosing one of said plurality of cumulative performance values, based
upon said comparison;
f) processing said plurality of input signals for a duration of time using
one set, corresponding to said chosen one cumulative performance value,
from said stored fixed plurality of sets to generate the plurality of
processed signals;
g) generating a new set of a plurality of signal parameters;
h) generating a plurality of cumulative performance values for said new set
and for each of said stored fixed plurality of sets;
i) comparing said plurality of cumulative performance values generated in
step (h),
j) choosing one of said plurality of cumulative performance values
generated in step (h), based upon the comparing step of (i);
k) based upon the comparing step of (i), either:
i) replacing one of said stored fixed plurality of sets, by said new set;
or
ii) deleting said new set;
l) processing said plurality of input signals for a duration of time using
one set, corresponding to said chosen one cumulative performance value,
from said stored fixed plurality of sets to generate the plurality of
processed signals; and
m) periodically reverting to steps (g)-(l).
2. The method of claim 1, wherein said generating step (h) comprising
arithmetically combining random values from a pseudorandom number
generator, one set in said plurality of sets, and recent values in said
plurality of input signals.
3. The method of claim 1 wherein:
said comparing step (i) compares the cumulative performance value of said
new set with the cumulative performance values of one of said stored fixed
plurality of sets having a least performance value;
and wherein said replacing step (k)(i) replaces one of said stored fixed
plurality of sets having said least performance value.
4. The method of claim 1 wherein:
said generating step (h) further comprises generating an initial cumulative
performance value for said new set by subtracting a fixed value from a
cumulative performance value of one of said stored fixed plurality of sets
having the greatest cumulative performance value; and
wherein said choosing step (j) chooses one of said plurality of cumulative
performance values generated in step (h) having the largest cumulative
performance value.
5. The method of claims 1, 2, 3, or 4 wherein said generating steps of (c)
and (h) further comprises:
generating a plurality of instantaneous performance values, each at a
different time, for each set; and
combining said plurality of instantaneous performance values for each set
to generate said plurality of cumulative performance values; and wherein
said plurality of input signals is characterized as x.sub.1 (t), x.sub.2
(t), . . . , x.sub.n (t), said plurality of processed signals is
characterized as y.sub.1 (t), y.sub.2 (t), . . . , y.sub.n (t), and each
of said choosing steps (e) and (j) selects a set having associated
instantaneous performance values that are highest for the sets that would
generate processed signals y.sub.i (t) and y.sub.j (t), that are most
statistically independent for different i and j.
6. The method of claim 5 wherein said plurality of input signals and said
plurality of processed signals are assumed to have zero mean and to
fluctuate in power over a few clock cycles, and said choosing steps (e)
and (j) selects a set for which the associated instantaneous performance
values are highest for the sets that would generate processed signals
y.sub.i (t) and y.sub.j (t) that most closely satisfy the relation
E›y.sub.i (t)y.sub.j (t)!=E›y.sub.i (t)!E›y.sub.j (t)! for any different
indices i and j, where the operation E is a weighted average over a period
of time of said value of x.sub.1 (t), x.sub.2 (t), . . . , x.sub.n (t),
y.sub.1 (t),y.sub.2 (t), . . . , y.sub.n (t).
7. The method of claim 6 wherein said choosing step of (e) and (h) computes
an instantaneous performance value in accordance with:
C(t)=-log (.SIGMA..sub.ij {E›y.sub.i (t)y.sub.j (t)!}.sup.2),
and wherein the logarithm computation is protected against numerical
overflow.
8. The method of claim 7 further comprising the step of:
generating said plurality of input signals by a plurality of transducer
means, based upon waves received by said plurality of transducer means
from a plurality of sources.
9. The method of claim 8 wherein said plurality of input signals is
generated by a plurality of transducer means, based upon acoustic waves
received by said plurality of transducer means from a plurality of
sources.
10. The method of claim 8 wherein said plurality of input signals is
generated by a plurality of transducer means, based upon electromagnetic
waves received by said plurality of transducer means from a plurality of
sources.
11. The method of claim 8 wherein said transducer means are directional and
positioned to minimize relative propagation delays.
12. The method of claim 11 wherein each set of said plurality of signal
parameters represents values of relative gains in transduction of said
waves from said waves by said plurality of transducer means.
13. The method of claim 12 wherein said choosing steps (e) and (j) computes
said instantaneous performance value in accordance with: Y(t)=inv G*X(t),
where X and Y are the vector representations of the signals x.sub.i (t)
and y.sub.i (t), G is the matrix representation of said relative gains,
and inv G is its inverse matrix.
14. The method of claim 13 wherein said choosing steps (e) and (j)
comprises computing Y(t) explicitly from the currently available input
signals X(t); and
calculating said instantaneous performance value by operating on the values
Y(t) for at least a duration of time equal to the averaging time of the
expectation operator E› !.
15. The method of claim 13 further comprising the step of storing said
plurality of input signals X(t) for at least a duration of time equal to
the averaging time of the expectation operator E› !; and
wherein said choosing steps (e) and (j) computes Y(t) explicitly from said
storage of input signals X(t), and subsequently said instantaneous
performance value.
16. The method of claim 13 wherein said choosing steps (e) and (j)
comprises the steps of:
precalculating, once per clock cycle, the quantity
e.sub.ij =E{x.sub.i (t)x.sub.j (t)}
for every (i,j) pair; and
computing said instantaneous performance value C(t) as required, from the
quantities e.sub.ij.
17. The method of claim 8 wherein
said plurality of signal parameters represent the coefficients of filters
that model the transfer function of propagation of said waves from said
plurality of sources to said plurality of transducer means; and
said choosing steps (e) and (j) computes said instantaneous performance
value in accordance with Y(t)=inv H*X(t), where X and Y are the vector
representations of the signals x.sub.i (t) and y.sub.i (t), H is the
matrix representation of said filters, and inv H is its inverse matrix.
18. The method of claim 17 wherein said choosing steps (e) and (j)
comprises the steps of:
computing Y(t) explicitly from the currently available input signals X(t);
and
calculating said instantaneous performance value by operating on the values
Y(t) for at least a duration of time equal to the averaging time of the
expectation operator E› ! plus the longest duration of said filters.
19. The method of claim 17 further comprising the step of storing said
plurality of input signals X(t) for at least a duration of time equal to
the averaging time of the expectation operator E› ! plus the longest
duration of said filters; and wherein said choosing steps (e) and (j)
computes Y(t) explicitly from said storage of input signals X(t), and
subsequently said instantaneous performance value.
20. The method of claim 17 wherein said choosing steps (e) and (j)
comprises the steps of:
precalculating, once per clock cycle, the quantity e.sub.ij
(k.sub.i,k.sub.j)=E{x.sub.i (t-k.sub.i .DELTA.t)x.sub.j (t-k.sub.j
.DELTA.t)} for every (i, j) pair, and for each such pair, for every
(k.sub.i,k.sub.j) pair; and
computing C(t) from the quantities e.sub.ij (k.sub.i,k.sub.j) every time
the instantaneous performance value is required.
21. The method of claim 8 wherein said transducer means are spaced apart
and omnidirectional; and wherein said plurality of signal parameters
represent values of relative delays in propagation of said waves from said
plurality of sources to said plurality of transducer means; and wherein
said choosing steps (e) and (j) computes said instantaneous performance
value using the definition Y(t)=adj D*X(t), where X and Y are the vector
representations of the signals x.sub.i (t) and y.sub.i (t), D is the
matrix representation of said relative delays, and adj D is its adjugate
matrix.
22. The method of claim 21 wherein said choosing steps (e) and (j)
comprises the steps of: computing Y(t) explicitly from the currently
available input signals X(t); and calculating said instantaneous
performance value by operating on the values Y(t) for at least a duration
of time equal to the averaging time of the expectation operator E› ! plus
the maximum value of said relative delays.
23. The method of claim 21 further comprising
the step of storing said plurality of input signals X(t) for at least a
duration of time equal to the averaging time of the expectation operator
E› ! plus the maximum value of said relative delays; and
wherein said choosing steps of (e) and (j) computes Y(t) explicitly from
said storage of input signals X(t), and subsequently said instantaneous
performance value.
24. The method of claim 22 wherein said choosing steps (e) and (j)
comprises the steps of:
precalculating, once per clock cycle, the quantity e.sub.ij
(k.sub.i,k.sub.j)=E{x.sub.i (t-k.sub.i .DELTA.t)x.sub.j (t-k.sub.j
.DELTA.t)} for every (i,j) pair, and for each such pair, for every
(k.sub.i,k.sub.j) pair; and
computing C(t), as is required, from the quantities e.sub.ij
(k.sub.i,k.sub.j) by implementing the necessary time delays using
linear-phase, non-causal FIR filters with a truncated sinc-shaped impulse
response.
25. A method of processing waves from a plurality of sources, comprising:
receiving said waves, including echoes and reverberations thereof, by a
plurality of transducer means;
converting said waves, including echoes and reverberations thereof from
said plurality of sources, by each of said plurality of transducer means
into a signal, thereby generating a plurality of signals;
calculating a set of a plurality of signal parameters by:
generating a fixed plurality of sets of a plurality of signal parameters;
storing said fixed plurality of sets;
generating a plurality of instantaneous performance values for each set,
with each instantaneous performance value generated at a different time;
combining said plurality of instantaneous performance values for each set
to produce a plurality of cumulative performance values, with a cumulative
performance value produced for each set;
storing said plurality of cumulative performance values;
periodically generating a new set of a plurality of signal parameters;
generating a new cumulative performance value, based upon said new set;
comparing said new cumulative performance value to said plurality of stored
cumulative performance values;
based upon said comparing step, either:
i) replacing one of said stored plurality of cumulative performance values,
by said new cumulative performance value, and the corresponding stored set
by said new set; or
ii) deleting said new cumulative value and said new set;
comparing said stored plurality of cumulative performance values;
choosing one of said stored plurality of cumulative performance values,
based upon said comparison;
choosing one set, corresponding to said chosen one cumulative performance
value;
supplying said chosen one set to a first processing means for operation
thereon;
first processing said plurality of signals, using said chosen one set,
corresponding to said chosen one cumulative performance value, to generate
a plurality of first processed signals, wherein each of said first
processed signals represents waves from one source, and a reduced amount
of waves from other sources; and then
secondly processing said plurality of first processed signals to generate a
plurality of second processed signals, wherein in the presence of echoes
and reverberations of said waves from said plurality of sources, each of
said second processed signals represents waves from only one different
source.
26. The method of claim 25 wherein said transducer means are spaced apart
omnidirectional microphones, and said chosen one set of plurality of
signal parameters has a set of relative delay parameters associated
therewith; said first processing step further comprises:
delaying said plurality of signals using said set of relative delay
parameters and generating a plurality of delayed signals in response
thereto; and
combining each one of said plurality of signals with at least one of said
plurality of delayed signals to produce one of said first processed
signals.
27. The method of claim 26 further comprising the step of:
filtering each of said second processed signals to generate a plurality of
third processed signals.
28. The method of claim 27 further comprising the step of:
sampling and converting each one of said plurality of signals and for
supplying same to said plurality of delay means and to said plurality of
combining means, as said signal.
29. The method of claim 25 wherein said second processing step further
comprising:
subtracting by a plurality of combining means one of said first processed
signals received at said first input, and the sum of input signals
received at a second input, to produce an output signal, said output
signal being one of said plurality of second processed signals;
generating a plurality of adaptive signals, with each of said adaptive
signals being the output signal of one of said plurality of combining
means; and
supplying each of said plurality of adaptive signals to second input of
said plurality of combining means other than the associated one combining
means.
30. The method of claim 25, wherein said step of generating a new set
comprises arithmetically combining random values from a pseudorandom
number generator, one set in said plurality of sets, and recent values of
said plurality of input signals.
31. The method of claim 30 wherein:
said comparing step compares the cumulative performance value of said new
set with the cumulative performance value of one of said stored plurality
of sets having a least cumulative performance value;
and wherein said replacing step replaces one of said stored plurality of
sets having said least cumulative performance value.
32. The method of claim 31 wherein: said replacing step further comprises:
generating an initial cumulative performance value for said new set by
subtracting a fixed value from a cumulative performance value of one of
said stored plurality of sets having the greatest cumulative performance
value; and
wherein said comparing step chooses one of said stored plurality of sets
having the greatest cumulative performance value, for processing by said
processing step.
33. An adaptive filter for determining a set of a plurality of signal
parameters for use in an adaptive filter signal processor to process a
plurality of input signals to generate a plurality of processed signals,
said filter comprising:
means for generating a fixed plurality of sets;
means for storing said fixed plurality of sets;
means for generating a plurality of cumulative performance values, based
upon said fixed plurality of sets, with a cumulative performance value
generated for each set;
means for evaluating said plurality of cumulative performance values, and
choosing one of said plurality of cumulative performance values, based
upon said evaluation; and
means for processing said plurality of input signals for a duration of time
using one set, corresponding to said chosen one cumulative performance
value, from said fixed plurality of stored sets to generate the plurality
of processed signals.
34. The filter of claim 33, further comprises:
means for periodically generating a new set of a plurality of signal
parameters;
means for generating a new cumulative performance value, for each set of
signal parameters, including said new set generated;
means for comparing said new cumulative performance value corresponding to
said new set generated to said cumulative performance values for each set
of signal parameters; and
means for either i) replacing one of said fixed number of plurality of said
sets, by said new set; or
ii) deleting said new set, in response to said comparing means.
35. The filter of claim 34 further comprises:
means for generating a plurality of instantaneous performance values for
each set, with each instantaneous performance value generated at a
different time; and
means for combining said plurality of instantaneous performance values for
each set to produce a plurality of cumulative performance values, with a
cumulative performance value produced for each set.
36. The filter of claim 35, wherein said means for generating a new
cumulative performance value comprises means for arithmetically combining
random values from a pseudorandom number generator, one set in said
plurality of sets, and recent values in said plurality of input signals.
37. The filter of claim 36 wherein:
said evaluating means compares instantaneous performance value of said new
set with instantaneous performance value of one of said stored plurality
of sets having a least instantaneous performance value;
and wherein said replacing means replaces one of said stored plurality of
sets having said least instantaneous performance value.
38. The filter of claim 37 wherein:
said means for generating a new cumulative performance value further
comprises means for generating an initial cumulative performance value for
said new set and means for subtracting a fixed value from a cumulative
performance value of one of said stored plurality of sets having the
greatest cumulative performance value; and
wherein said evaluating means comprises means for choosing one of said
stored plurality of sets having the greatest cumulative performance value,
for processing by said processing means.
39. A signal processing system for processing waves from a plurality of
sources, said system comprising:
a plurality of transducer means for receiving waves from said plurality of
sources, including echoes and reverberations thereof and for generating a
plurality of signals in response thereto, wherein each of said plurality
of transducer means receives waves from said plurality of sources
including echoes and reverberations thereof, and for generating one of
said plurality of signals;
means for calculating a set of a plurality of signal parameters, said
calculating means comprising:
means for generating a fixed plurality of sets of signal parameters;
means for storing said fixed plurality of sets of signal parameters;
means for generating a plurality of cumulative performance values, based
upon said fixed plurality of sets of signal parameters, with a cumulative
performance value generated for each set of signal parameters;
means for evaluating said plurality of cumulative performance values, and
choosing one of said plurality of cumulative performance values, based
upon said evaluation, and one of said sets of signal parameters
corresponding to said one of said plurality of cumulative performance
values chosen;
first processing means for receiving said plurality of signals, and said
plurality of signal parameters of said set chosen for generating a
plurality of first processed signals in response thereto, wherein each of
said first processed signals represents waves from one source, and a
reduced amount of waves from other sources; and
second processing means for receiving said plurality of first processed
signals and for generating a plurality of second processed signals in
response thereto, wherein each of said second processed signals represents
waves from only one source.
40. The system of claim 39, further comprising:
means for generating a direction of arrival signal for said waves;
wherein said first processing means for generating said plurality of first
processed signals, in response to said direction of arrival signal.
41. The system of claim 39, wherein the number of transducer means is two,
and the number of sources is two.
42. The system of claim 39, wherein said transducer means are spaced apart
omnidirectional microphones and wherein said chosen one set of plurality
of signal parameters has a set of relative delay parameters, and said
first processing means comprises:
a plurality of delay means, each for receiving one of said plurality of
signals and using said set of relative delay parameters for generating a
plurality of delayed signals in response thereto; and
a plurality of combining means, each for receiving at least one delayed
signal and one of said plurality of signals and for combining said
received delayed signal and said signal to produce one of said first
processed signals.
43. The system of claim 39 wherein said plurality of transducer means are
co-located directional microphones and wherein said one of said set of
signal parameters has a set of gain parameters associated therewith, and
wherein first processing means comprises:
a plurality of multiplying means, each for receiving different ones of said
plurality of signals and said set of gain parameters and for generating a
scaled signal in response thereto; and
a plurality of combining means, each for receiving at least one scaled
signal and one of said plurality of signals and for combining said
received scaled signal and said signal to produce one of said first
processed signals.
44. The system of claim 39, 40, 41, 42, or 43 wherein said second
processing means comprises:
a plurality of combining means, each combining means having a first input,
at least one other input, and an output; each of said combining means for
receiving one of said first processed signals at said first input, an
input signal at said other input, and for generating an output signal, at
said output; said output signal being one of said plurality of second
processed signals and is a difference between said first processed signal
received at said first input and the sum of said input signal received at
said other input;
a plurality of adaptive filter means for generating a plurality of adaptive
signals, each of said adaptive filter means for receiving said output
signal from one of said plurality of combining means and for generating an
adaptive signal in response thereto; and means for supplying each of said
plurality of adaptive signals to one of said other input of said plurality
of combining means other than the associated one combining means.
45. The system of claim 44 further comprising means for filtering each of
said second processed signals to generate a plurality of third processed
signals.
46. The system of claim 45 wherein said second processed signals are
characterized by having a low frequency component and a high frequency
component, and wherein said filtering means boosts the low frequency
component relative to the high frequency component of said second
processed signals.
47. A signal processing system for processing waves from a plurality of
sources, said system comprising:
a plurality of transducer means for receiving waves from said plurality of
sources, including echoes and reverberations thereof and for generating a
plurality of signals in response thereto, wherein each of said plurality
of transducer means receives waves from said plurality of sources
including echoes and reverberations thereof, and for generating one of
said plurality of signals;
an adaptive filter for generating a plurality of signal parameters, said
filter comprising:
means for generating a fixed plurality of sets of signal parameters;
means for storing said fixed plurality of sets of signal parameters;
means for generating a plurality of cumulative performance values, based
upon said fixed plurality of sets of signal parameters, with a cumulative
performance value generated for each set;
means for evaluating said plurality of performance values, and choosing one
of said plurality of performance values and its corresponding set of
plurality of signal parameters, based upon said evaluation;
first processing means for receiving said plurality of signals and said
plurality of signal parameters and for generating a plurality of first
processed signals in response thereto, wherein in the absence of echoes
and reverberations of said waves from said plurality of sources, each of
said first processed signals represents waves from only one different
source; and
second processing means for receiving said plurality of first processed
signals and for generating a plurality of second processed signals in
response thereto, wherein in the presence of echoes and reverberations of
said waves from said plurality of sources, each of said second processed
signals represents waves from only one source.
48. The system of claim 47 wherein said waves are acoustic waves, and said
transducer means are microphones.
49. The system of claim 48 further comprising means for filtering each of
said second processed signals to generate a plurality of third processed
signals.
50. The system of claim 49 wherein said second processed signals are
characterized by having a low frequency component and a high frequency
component and wherein said filtering means boosts the low frequency
component relative to the high frequency component of said second
processed signals.
51. The system of claim 49 wherein said microphones are spaced apart
omnidirectional microphones and wherein said corresponding set of signal
parameters has a set of relative delay parameters associated therewith;
and
said first processing means comprises:
a plurality of delay means, each for receiving one of said plurality of
signals and said set of relative delay parameters and for generating a
delayed signal in response thereto; and
a plurality of combining means, each for receiving at least one delayed
signal and one of said plurality of signals and for combining said
received delayed signal and said signal to produce one of said first
processed signals.
52. The system of claim 48 wherein said microphones are co-located
directional microphones wherein said corresponding set of signal
parameters has a set of gain parameters associated therewith; and
said first processing means comprises:
a plurality of multiplying means, each for receiving different ones of said
plurality of signals and said set of gain parameters and for generating a
scaled signal in response thereto; and
a plurality of combining means, each for receiving at least one scaled
signal and one of said plurality of signals and for combining said
received scaled signal and said signal to produce one of said first
processed signals.
53. The systems of claims 47, 48, 49, 50, 51 or 52, wherein said second
processing means comprises:
a plurality of combining means, each combining means having a first input,
at least one other input, and an output; each of said combining means for
receiving one of said first processed signals at said first input, an
input signal at said other input, and for generating an output signal, at
said output; said output signal being one of said plurality of second
processed signals and is a difference between said first processed signal
received at said first input and the sum of said input signal received at
said other input;
a plurality of adaptive filter means for generating a plurality of adaptive
signals, each of said adaptive filter means for receiving said output
signal from one of said plurality of combining means and for generating an
adaptive signal in response thereto; and
means for supplying each of said plurality of adaptive signals to one of
said other input of said plurality of combining means other than the
associated one combining means.
54. The system of claim 53 wherein each of said adaptive filter means
comprises a tapped delay line.
55. An adaptive filter signal processing system for processing waves from a
plurality of sources, said system comprising:
a plurality of transducer means for receiving waves from said plurality of
sources, including echoes and reverberation thereof and for generating a
plurality of signals in response thereto, wherein each of said plurality
of transducer means receives waves from said plurality of sources
including echoes and reverberations thereof, and for generating one of
said plurality of signals;
an adaptive filter for generating a plurality of signal parameters, said
filter comprising:
means for generating a fixed plurality of sets of signal parameters;
means for storing said fixed plurality of sets signal parameters;
means for generating a plurality of performance values, based upon said
fixed plurality of sets of signal parameters, with a performance value
generated for each set;
means for evaluating said plurality of performance values, and choosing one
of said plurality of performance values, and its corresponding set of
signal parameters, based upon said evaluation;
first processing means comprises a beamformer for receiving said plurality
of signals and said plurality of signal parameters, and for generating a
plurality of first processed signals in response thereto, wherein each of
said first processed signals represents waves from one source, and a
reduced amount of waves from other sources; and
second processing means for receiving said plurality of first processed
signals and for generating a plurality of second processed signals in
response thereto, wherein each of said second processed signals represent
waves from only one source.
56. The system of claim 55, wherein said transducer means are spaced apart
omnidirectional microphones and said corresponding set of signal
parameters has a set of delay parameters associated therewith, and wherein
said first processing means comprises:
a plurality of delay means, each for receiving one of said plurality of
signals and said set of delay parameters, and for generating a delayed
signal in response thereto; and
a plurality of combining means, each for receiving at least one delayed
signal and one of said plurality of signals and for combining said
received delayed signal and said signal to produce one of said first
processed signals.
57. The system of claim 55 wherein said second processing means comprises:
a plurality of combining means, each combining means having a first input,
at least one other input, and an output; each of said combining means for
receiving one of said first processed signals at said first input, an
input signal at said other input, and for generating an output signal, at
said output; said output signal being one of said plurality of second
processed signals and is a difference between said first processed signal
received at said first input and the sum of said input signal received at
said other input;
a plurality of adaptive filter means for generating a plurality of adaptive
signals, each of said adaptive filter means for receiving said output
signal from one of said plurality of combining means and for generating an
adaptive signal in response thereto; and
means for supplying each of said plurality of adaptive signals to one of
said other input of said plurality of combining means other than the
associated one combining means.
58. The system of claims 55, 56, or 57, wherein said first processing means
comprises analog circuits.
59. The system of claims 55, 56, or 57, wherein said second processing
means comprises analog circuits.
60. The system of claims 55, 56, or 57, wherein said first processing means
are a part of a digital signal processor.
61. The system of claims 55, 56, or 57, wherein said second processing
means are a part of a digital signal processor.
62. The system of claims 55, 56, or 57, wherein said first processing means
are a part of a general purpose computer.
63. The system of claims 55, 56, or 57, wherein said second processing
means are a part of a general purpose computer.
64. The system of claims 55, 56, or 57, wherein said first processing means
are reconfigurable gate array circuits.
65. The system of claims 55, 56, or 57, wherein said second processing
means are reconfigurable gate array circuits.
66. The system of claim 55, further comprising:
a plurality of third processing means for receiving said plurality of
second processed signals and for removing frequency coloration therefrom.
67. The system of claim 56 further comprising:
a plurality of sampling digital converting means, each for receiving a
different one of said plurality of signals and for generating a digital
signal; said digital signal supplied to said plurality of delay means and
to said plurality of combining means, as said signal.
68. The system of claims 55, 56, or 57 further comprises:
means for periodically generating a new set of a plurality of signal
parameters;
means for generating a new cumulative performance value, for each of said
sets, including said new set;
means for comparing said new cumulative performance values; and
switch means for either i) replacing one of said fixed number of plurality
of said sets, by said new set; or ii) deleting said new set, in response
to said comparing means.
69. The system of claim 68 further comprises:
means for generating a plurality of instantaneous performance values for
each of said fixed plurality of sets, with each instantaneous performance
value generated at a different time;
means for combining said plurality of instantaneous performance values for
each set to produce a plurality of cumulative performance values, with a
cumulative performance value produced for each set.
70. The system of claim 69, wherein said means for generating a new
cumulative performance value for each of said sets, including said new
set, comprises means for arithmetically combining random values from a
pseudorandom number generator, one set in said plurality of sets, and
recent values in said plurality of input signals.
71. The system of claim 70 further comprises:
means for generating a plurality of instantaneous performance values for
each of said sets, including said new set, with each instantaneous
performance value generated at a different time; and wherein:
said comparing means compares instantaneous performance value of said new
set with instantaneous performance value of one of said stored plurality
of sets having a least instantaneous performance value;
and wherein said replacing means replaces one of said stored plurality of
sets having said least instantaneous performance value.
72. The system of claim 68 further comprises
means for generating an initial cumulative performance value for said new
set and means for subtracting a fixed value from a cumulative performance
value of one of said stored plurality of sets having the greatest
cumulative performance value; and
wherein said comparing means comprises means for choosing one of said
stored plurality of sets having the greatest cumulative performance value,
for processing by said processing means. |
|
 |
|
 |
Claims |
|
