Multiplex communications – Generalized orthogonal or special mathematical techniques – Particular set of orthogonal functions
Reexamination Certificate
1999-12-28
2003-09-30
Hsu, Alpus H. (Department: 2665)
Multiplex communications
Generalized orthogonal or special mathematical techniques
Particular set of orthogonal functions
C370S210000, C375S260000, C375S343000
Reexamination Certificate
active
06628606
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an orthogonal frequency division multiplexing (OFDM) receiver, and more particularly, to a coarse frequency offset estimator in an OFDM receiver, and a method of estimating a coarse frequency offset.
2. Description of the Related Art
FIG. 1
is a block diagram showing the structure of a conventional OFDM receiver. Referring to
FIG. 1
, the conventional OFDM receiver includes an OFDM demodulator
10
and a coarse frequency offset estimator
12
. The OFDM demodulator
10
includes a radio frequency (RF) receiver
101
, an analog-to-digital converter (ADC)
102
, an in-phase/quadrature (I/Q) separator
103
, a frequency corrector
104
, a fast Fourier transformer (FFT)
105
, and a Viterbi decoder
106
. The coarse frequency offset estimator
12
includes a register
121
, a complex data multiplier
122
, an inverse fast Fourier transformer (IFFT)
123
, a maximum value detector
124
, a counter
125
, and a reference symbol generator
126
.
In the operation of the receiver having such a configuration, first, the RF receiver
101
. receives an RF wave and outputs an RF signal. The ADC
102
quantizes the RF signal. The I/Q separator
103
separates an in-phase (I) component and a quadrature (Q) component from the quantized RF signal. The frequency corrector
104
performs frequency-correcting. The FFT
105
receives a frequency-corrected signal and performs Fourier-transforming of the frequency-corrected signal, thereby performing demodulation. The Viterbi decoder
106
decodes a demodulated signal.
Meanwhile, the demodulated signal is stored in the register
121
of the coarse frequency offset estimator
12
, and is output as a received signal X. A reference symbol which is output by the reference symbol generator
126
is represented by Z. When a received symbol has a frame synchronization offset of &ohgr;, and frequency offset does not exist, if the k-th sub-carriers of the received symbol X and the reference symbol Z are X
k
and Z
k
, respectively, X
k
and Z
k
have a relationship as X
k
=Z
k
e
−J2&pgr;&ohgr;|N
. The complex data multiplier
122
multiplies the conjugate value of X
k
by the conjugate value of Z
k
. The signal output from the complex data multiplier
122
is inversely Fourier-transformed by the IFFT
123
, and the following signal h
n
is output:
h
n
=
IFFT
{
XZ
*
}
=
1
/
N
∑
k
=
0
N
-
1
X
k
Z
k
*
ⅇ
j2π
kn
/
N
=
1
/
N
∑
k
=
0
N
-
1
Z
k
ⅇ
-
j
2
π
k
ω
/
N
Z
k
*
ⅇ
j
2
π
kn
/
N
=
1
/
N
∑
k
=
0
N
-
1
&LeftBracketingBar;
Z
k
&RightBracketingBar;
2
ⅇ
j
2
π
k
(
n
-
ω
)
/
N
=
δ
(
n
-
ω
)
(
1
)
A received symbol X
k
having an integer multiple &Dgr;f
i
of a frequency offset with respect to a transmitted symbol Z
k
can be expressed as Z
k−&Dgr;f
i
e
−J2&pgr;k&ohgr;|N
, so that Equation 1 can be expressed as the following Equation 2:
h
n
=
IFFT
{
XZ
*
}
=
1
/
N
∑
k
=
0
N
-
1
X
k
Z
k
*
ⅇ
j
2
π
kn
/
N
=
1
/
N
∑
k
=
0
N
-
1
Z
k
-
Δ
f
i
*
ⅇ
-
j2π
k
ω
/
N
Z
k
*
ⅇ
j2π
kn
/
N
(
2
)
The result of Equation 2 is the same as a process for obtaining the convolution of two signals in the time domain, and the result value h
n
is a channel impulse response (CIR). As described above, an OFDM system which has reference symbol in a frequency domain can obtain the CIR using a received symbol. Here, because the reference symbol Z consists of a pseudo noise (PN) sequence, maximum peak value exists only if frequency offset &Dgr;f
i
, is zero, and if otherwise noisy peaks with small value are gained. Using such a relationship, the coarse frequency offset estimator
12
shifts a received symbol with respect to &Dgr;f
i
, and the maximum value detector
124
detects a maximum peak value according to Equation 2. The counter
125
obtains a shift amount &Dgr;f
i
in which the maximum peak value is generated. The shift amount &Dgr;f
i
becomes an offset value F
o
which is an integral multiple of a frequency offset intended to be detected, which establishes a relationship expressed as the following Equation 3:
F
o
=
max
Δ
f
i
{
max
amp
[
IFFT
{
X
R
Z
}
]
}
(
3
)
wherein Z denotes a reference symbol, and X
R
is expressed as a symbol X
((k+&Dgr;f
i
))N
, obtained by shifting an individual sub-carrier X
k
of a received symbol X by &Dgr;f
i
in the frequency domain.
According to this coarse frequency offset estimation method, a conventional OFDM receiver as described above can theoretically and practically estimate an almost exact frequency offset in all cases regardless of a channel environment or a frame synchronization error. However, this method has significantly many calculation processes. Thus, a very complicated inverse fast Fourier transformation (IFFT) module is required to estimate an accurate frequency offset within a predetermined short period of time. Also, a long response time causes an excessive time delay.
To solve the problem, another conventional coarse frequency offset estimation method is disclosed. According to the method, first, to find out the influence of a frequency offset on the received symbol, f
k
and f
off
are defined. Here, f
k
denotes the frequency of a k-th sub-carrier, and f
off
denotes an actual frequency offset. The frequency offset is expressed as a multiple of a subcarrier frequency interspacing. Generally, the frequency offset includes an offset expressed as an integral multiple of the subcarrier frequency interspacing, and an offset expressed as a prime multiple thereof, and the multiples are individually processed. Thus, each term in: f
k
+f
off
can be defined as the following Equation 4:
f
k
=
k
T
s
f
off
=
Δ
f
1
T
s
=
(
Δ
f
i
+
Δ
f
f
)
1
T
s
(
4
)
wherein &Dgr;f denotes the frequency offset of a subcarrier expressed by a multiple of a subcarrier frequency interspacing. Also, &Dgr;f is expressed as the sum of an integer number &Dgr;f
i
and a floating number &Dgr;f
f
which satisfies a condition −½<&Dgr;f
f
<½. Under this condition, a received symbol of an n-th symbol is expressed as the following Equation. However, for convenience of the development of the following Equation 5, it is assumed that no noise exists.
r
n
(
m
)
=
∑
k
=
0
N
-
1
C
n
,
k
ⅇ
j
2
π
[
k
T
s
+
(
Δ
f
i
+
Δ
f
f
)
1
T
s
]
T
s
N
m
=
∑
k
=
0
N
-
1
C
n
,
k
ⅇ
j
2
π
[
(
k
+
Δ
f
i
+
Δ
f
f
)
m
/
N
(
5
)
wherein C
n,k
denotes a k-th sub-carrier of an n-th symbol in a frequency domain, and N denotes the number of OFDM sub-carriers.
Meanwhile, when an integral multiple of a frequency offset &Dgr;fi among frequency offsets is zero, a demodulated signal Ĉ′
n,p
is expressed as the following Equation 6:
C
^
n
,
p
′
=
1
N
∑
m
=
0
N
-
1
r
n
(
m
)
ⅇ
-
j
2
π
m
p
/
N
=
1
N
∑
m
=
0
N
-
1
∑
k
=
0
N
-
1
C
n
,
k
ⅇ
j
2
π
(
k
+
Δ
f
i
)
m
N
ⅇ
-
j2π
m
p
/
N
=
1
N
∑
k
=
0
N
-
1
C
n
,
k
∑
m
=
0
N
-
1
ⅇ
j2π
(
k
+
Δ
&em
Bang Keuk-joon
Hong Dae-sik
Park Hyun-cheol
Park Kyoung-shin
Hsu Alpus H.
Nguyen Toan D.
Samsung Electronics Co,. Ltd.
LandOfFree
Coarse frequency offset estimator in orthogonal frequency... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Coarse frequency offset estimator in orthogonal frequency..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Coarse frequency offset estimator in orthogonal frequency... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3058319