Pulse or digital communications – Receivers – Interference or noise reduction
Reexamination Certificate
1999-10-12
2003-06-10
Chin, Stephen (Department: 2634)
Pulse or digital communications
Receivers
Interference or noise reduction
C375S142000, C375S213000, C375S316000, C375S346000, C455S132000
Reexamination Certificate
active
06577686
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a receiver used in reception of digitally modulated radio frequency signal, for receiving the signal by plural antennas.
BACKGROUND OF THE INVENTION
Recently, in the field of mobile communication, digitization of radio communication is being promoted from the viewpoint of improvement of security and privacy, degree of congeniality with ISDN (integrated service digital network) or computer systems, effective use of frequency resources, etc. For effective use of frequency resources, it is preferred to utilize the radio wave of same frequency (channel) repeatedly at a minimum distance. However, if the distance of frequency re-use is shortened, it causes to increase the radio interference (co-channel interference) from the neighboring mobile station or base station using the same channel. As a result, the transmission quality is degraded.
In mobile communication, meanwhile, since fading occurs, the transmission quality, which is error rate in digital communication, is extremely degraded. Usually, the deterioration of transmission quality due to fading is compensated by space diversity reception which is implemented with two or more antennas and receiver circuits for diversity branches.
As branch combining method, that is, as the method of combining the signals issued from the plural receiver circuits into a combined signal, post-detection selection combining is the most common in which the detected branch output signal with the highest receiving signal strength (RSSI) is selected as the combined output. As another combining method which can further improve the receiving characteristic, post-detection maximum ratio combining is known. With the maximum ratio combining, the combined base band signal is made as the summation of the detected base band signal weighted by the RSSI at every branch.
Diversity reception is known not only to compensate for fading but also to suppress deterioration of transmission quality due to co-channel interference. To realize better performances against co-channel interference, adaptive array diversity, which is sometimes called adaptive diversity, optimum combining diversity, minimum mean square error combining diversity or least mean square adaptive array, is proposed. Examples of such receivers are disclosed in Japanese Laid-open Patent No. 7-154129 and Japanese Laid-open Patent No. 9-820400. These adaptive array diversity techniques can reduce the deterioration due to co-channel interference so the spectrum efficiency is enhanced.
In the adaptive array diversity receivers, for each diversity branch, there is a receiver circuit converting the receiving signals from the antenna into a base band signal which is not detected coherently. For the k-th branch (k=1, 2, . . . , K), the base band signal Xk is multiplied by a complex weight Wk and summed with the weighted base band signals from the other branches for combining. The Wk is calculated to adjust the phase and the amplitude of each branch in order to intensify the desired signal and to cancel or reduce the interference signal and noise at the combining output. By comparing this combined base band signal with a proper threshold value, demodulated data is decided and extracted.
Herein, those weights. (W
1
, W
2
, . . . ) are updated iteratively with X
1
, X
2
, . . . and the error signal E between the reference signal and the signal after being combined so that the error becomes small, or that the absolute value of the combining becomes constant.
As the weight update algorithm, the least mean square (LMS) algorithm or the recursive least mean square (RLS) algorithm can be used as well as for the linear equalizers with a tapped delay line. These are disclosed in the publication, Simon Haykin, Adaptive Filter Theory, 3rd edition, Prentice Hall, Upper Saddle River, N.J., 1996, for example. Among those algorithms, the LMS algorithm according to formula 1 is the simplest in calculation and is used frequently.
W
m
(
n
)
=
W
m
(
n
-
1
)
+
μ
X
m
(
n
-
1
)
E
*
(
n
-
1
)
(
n
=
0
,
1
,
2
,
…
)
E
(
n
)
=
D
(
n
)
-
∑
m
=
1
M
W
m
*
(
n
)
X
m
(
n
)
W
:
complex
weight
μ
:
fixed
step
size
X
:
complex
base
band
signal
E
:
complex
error
signal
D
:
complex
reference
signal
m
:
number
of
branches
n
:
number
of
symbols
*
:
complex
conjugate
Formula
1
Herein, n is the time in the unit of the number of symbols from the beginning of the training sequence (training signals), and &mgr; is the step size. For example, International Laid-open Patent WO97/20400 discloses the diversity receiver having the LMS applied in the spectrum diffusion communication. It is known that the convergence speed becomes fast if &mgr; is set large, but the residual error after convergence and the stability deteriorates. To the contrary, the characteristic after convergence is good when &mgr; is small. Accordingly, &mgr; is generally set smaller although the convergence speed is somewhat sacrificed.
In such conventional receiver, however, the problem is that the weight convergence speed is slow when the LMS is used as the weight update algorithm. On the other hand, the RLS algorithm can be used for fast convergence in place of the LMS. Nevertheless, a fast and complicated operation circuit is needed because the computation of the RLS is extremely complicated.
SUMMARY OF THE INVENTION
The invention is devised in the light of the above prior art, and it is hence an object thereof to present a receiver of adaptive array diversity capable of converging the weight at high speed.
To achieve this object, the receiver of the invention comprises:
(a) converting means provided individually in each antenna for converting the signal received on the antenna into a base band signal,
(b) combining means for combining by multiplying this base band signal by the weight of complex number and adding the weighted base band signal,
(c) deciding means for deciding the transmission symbol from the combined base band signal added and combined by this combining means,
(d) reference signal generating means for generating a known symbol to be transmitted as a reference signal,
(e) error detecting means for issuing an error signal between the combined base band signal and reference signal, and
(f) weight calculating means for calculating the weight corresponding to each antenna from the error signal and base band signal, and updating this weight iteratively.
Thus constituted receiver of the invention is characterized by the operation in which:
(g) the updating amount of weight in the weight calculating means is the product of the error signal, step size function and each base band signal,
(h) this step size function is a function becoming smaller depending on the lapse of time in each time slot, and
(i) it is initialized every time the time slot is changed.
By thus constitution and operation, the receiver of the invention is small in the quantity of calculation, and is capable of converging the weight at high speed by using a simple algorithm.
REFERENCES:
patent: 5757845 (1998-05-01), Fukawa et al.
patent: 5838742 (1998-11-01), Abu-Dayya
patent: 5991273 (1999-11-01), Abu-Dayya
patent: 6069912 (2000-05-01), Sawahashi et al.
patent: 6289062 (2001-09-01), Wang et al.
patent: 07154129 (1995-06-01), None
patent: WO97/20400 (1997-06-01), None
Koga Hisao
Taromaru Makoto
Ahn Sam
Chin Stephen
RatnerPrestia
LandOfFree
Receiver does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Receiver, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Receiver will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3105595