Electricity: measuring and testing – Measuring – testing – or sensing electricity – per se – Frequency of cyclic current or voltage
Reexamination Certificate
2000-05-30
2002-11-19
Oda, Christine (Department: 2858)
Electricity: measuring and testing
Measuring, testing, or sensing electricity, per se
Frequency of cyclic current or voltage
C324S076380, C324S076290, C702S075000
Reexamination Certificate
active
06483286
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention is applicable to the implementation of digital demodulation of analog frequency modulated signal in which the frequency of the modulating signal is relatively small compared to the sampling frequency. A typical application of this invention is demodulation of the frequency modulated TV sound sampled at high frequency.
2. Description of the Prior Art
An FM modulated signal y(t) can be expressed as:
y
(
t
)=
A
sin[&psgr;(
t
)]=
A
sin[2
&pgr;f
c
t
+&phgr;(
t
)]
where A is the amplitude of the modulated signal, f
c
is the carrier frequency and &phgr;(t) is a function of the baseband (modulating) signal m(t) as follows
φ
(
t
)
=
k
∫
-
∞
t
m
(
t
)
ⅆ
t
where k is the modulation constant. The instantaneous frequency f(t) of the modulated signal y(t) is a linear function of m(t) as follows
f
(
t
)
=
1
2
π
ⅆ
ψ
(
t
)
ⅆ
t
=
f
c
+
k
2
π
m
(
t
)
The FM demodulation is to recover m(t) from the FM modulated signal y(t). There are several methods to demodulate an analog FM modulated signal using digital techniques. The arc tangent FM demodulation method (see “Demodulator of sampled data FM signals from sets of four successive samples”, by J. J. Gibson, U.S. Pat. No. 4,547,737, and “Digital Frequency discriminator”, by F. G. A. Goupe, Electronics Letters, vol. 15, no. 16, August 1979) uses the Hilbert transform to decompose the carrier into two signals a(t) and b(t) which are orthogonal to each other, i.e.
y
(
t
)=
a
(
t
)+
jb
(
t
)
Thus, the baseband signal m(t) can be obtained by
m
(
t
)
=
α
ⅆ
{
tan
-
1
[
b
(
t
)
/
a
(
t
)
]
}
ⅆ
t
where &agr; is a constant. Another method is proposed based on an arc cosine system (see “Digital FM demodulation apparatus demodulating sampled digital FM modulated wave” by Y. Ishikawa and S. Nomura, U.S. Pat. No. 5,444,416). In this arc cosine FM demodulation method, the FM modulated signal is first sampled at a sampling interval T. Consecutive samples y
t−T
, y
t
and y
t+T
at time t−T, t and t+T can be expressed as follows:
y
t−T
=A
sin[2&pgr;
f
(
t−T
)+&thgr;]
y
t
=A
sin[2&pgr;
ft+&thgr;]
y
t−T
=A
sin[2&pgr;
f
(
t+T
)+&thgr;]
where A, f and &thgr; are an amplitude, an instantaneous frequency and an initial phase of y
t
, respectively. It has been proved that the instantaneous frequency f at time t is given by:
f
=
1
2
π
T
cos
-
1
(
y
t
-
T
+
y
t
+
T
2
y
t
)
Once the instantaneous frequency of a FM modulated i signal is obtained, the demodulated signal can be obtained from the instantaneous frequency using a linear equation.
As described above, Hilbert transform is used in the arc tangent digital demodulation method is a phase circuit employing the Hilbert transform, The required circuit is large and the computation process introduces long delay. In order to avoid this problem, the sampling frequency must be carefully chosen to be four times the carrier frequency rate. This is not practical. Firstly, different FM systems use different carrier frequencies. The demodulation system will be too complex if multiple frequencies are introduced. Secondly, the sampling frequency is sometimes restricted by requirements of other modules of an FM demodulation and signal processing system. The arc cosine digital FM demodulation method is simpler than the arc tangent FM digital demodulation system. However, there is still an inconvenient division operation involved in the procedure for computing the instantaneous frequency. This is not practical in hardware implementation since the values of y, are sometimes close or equal to zero. Thus, special processing is needed to prevent the demodulation system from overflow if this situation happens.
SUMMARY OF THE INVENTION
The problem to be solved by this invention is to develop a simple method for digital FM demodulation having less complexity than the methods described in the prior art.
The new method of calculating the instantaneous frequency based on the sampled values is described as follows. Assume that an FM modulated signal is sampled at a sampling interval of T. If the sampling frequency is high enough, we can assume that the instantaneous frequency values at time t−T, t and t+T are approximately the same, i.e.
f
t−T
≈f
t
≈f
t+T
Hence, three consecutive samples y
t−T
, y
t−T
, and y
t−T
at time t−T, t and t+T can be respectively expressed as follows:
y
t−T
=A
sin[2&pgr;
f
t
(
t−T
)+&thgr;]
y
t
=A
sin[2&pgr;
f
t
t
+&thgr;]
y
t−T
=A
sin[2&pgr;
f
t
(
t+T
)+&thgr;]
Let &phgr;=2&pgr;f
t
t+&thgr; and &phgr;
0
=2&pgr;f
t
T, we have
y
t−T
=A
sin(&phgr;−&phgr;
0
)
y
t
=A
sin &phgr;
y
t−T
=A
sin(&phgr;+&phgr;
0
)
Since
y
t
-
T
y
t
+
T
=
A
2
sin
(
ϕ
-
ϕ
0
)
sin
(
ϕ
+
ϕ
0
)
=
A
2
2
[
cos
(
2
ϕ
0
)
-
cos
(
2
ϕ
)
]
and
cos
(
2
ϕ
)
=
1
-
2
sin
2
ϕ
=
1
-
2
A
2
y
t
2
Therefore
cos
(
2
ϕ
0
)
=
2
A
2
(
y
t
-
T
y
t
+
T
-
y
t
2
)
+
1
Thus, the instantaneous frequency value at time t can be computed by the following equation:
f
t
=
1
4
π
T
cos
-
1
[
2
A
2
(
y
t
-
T
y
t
+
T
-
y
t
2
)
+
1
]
Based on the equations described here, an apparatus is set up for measuring instantaneous frequency of FM-modulated signal. Said apparatus comprises a sampling means, an instantaneous frequency computing means and a lowpass filtering means. Said instantaneous frequency computing means further comprises two delay means, a multiplier, a square means, a bit-shifting means, a subtracting means, a scaling means, an adding means and an inverse cosine computing means.
The description will now be made on the operation of the apparatus invented for digital FM demodulation. Said sampling means samples the input analog FM modulated signal at a prescribed sampling interval T to obtain three consecutive samples y
t−T
, y
t
and y
t+T
at times t−T, t and t+T, respectively. Said signal instantaneous frequency computing means computes said instantaneous frequency f
t
at time t based on an equation
f
t
=
1
4
π
T
cos
-
1
(
s
t
)
,
where
s
t
=
2
A
2
(
y
t
-
T
y
t
+
t
-
y
t
2
)
+
1
and A is the amplitude of an FM modulated signal. The instantaneous frequency is then filtered by said lowpass filter to reduce the noise introduced by said data sampling means and said instantaneous frequency computing means.
The operations of said instantaneous frequency computing means are now explained. The first delay means receives the digitized FM signal and delays said digitized FM signal by one prescribed interval to provide a first delayed FM signal. The second delay means receives said first delayed FM signal and further delays said first delayed FM signal by said prescribed interval to provide a second delayed FM signal. The square means receives said first delayed signal and computes the squared values of said first delayed FM signals. The multiplier computes the product values of digitized FM signal and said second delayed FM signal. Said squared values are subtracted from said product values to provide the difference signal. Each value of said difference signal is left-shifted by one bit to the left by said bit-shifting means to obtain bit-shifted signal. The scaled signal is then obtained by scaling said bit-shifted signal by the factor A
2
Bi Mi Michael
Kiew Ming Kwong Peter
Ng Chin Hon
Deb Anjan K.
Greenblum & Bernstein P.L.C.
Oda Christine
LandOfFree
Method and apparatus for measuring the instantaneous... 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 and apparatus for measuring the instantaneous..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for measuring the instantaneous... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2993688