Telecommunications – Receiver or analog modulated signal frequency converter – Measuring or testing of receiver
Reexamination Certificate
1999-09-10
2003-03-25
Vo, Nguyen T. (Department: 2682)
Telecommunications
Receiver or analog modulated signal frequency converter
Measuring or testing of receiver
C455S522000, C455S067700, C455S226100, C370S318000, C370S320000
Reexamination Certificate
active
06539214
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to a method of estimating the signal to noise ratio of a digital signal received by a radiocommunications receiver.
It concerns more particularly a method of estimating such a signal to noise ratio in a radiocommunications system for which the communication resources are shared according to a plurality of codes.
2. Description of the Prior Art
It is known that, in a digital type telecommunications system, a large number of different communications can be transmitted simultaneously. This simultaneity is obtained by resorting to different codes and/or frequencies and/or time slots.
There will be considered here, by way of example, a telecommunications system in which a plurality of terminals communicate with a control station, notably by means of retransmission means on board a satellite. The inter-terminal communications are performed by means of the control station. Thus, the latter communicates simultaneously with a set of terminals. It is this multiplicity of simultaneous communications which makes it necessary to resort to shared code and/or frequency and/or time slot resources.
In this system, the resources are limited by the retransmission capacity of the equipment on board the satellite. It is therefore necessary to allocate, to each transmitter, the power which is just sufficient to satisfy the communications requirements, that is to say the power allowing the bit error rate to always be less than a required rate. In order to satisfy this requirement, it is ensured that, at any instant, the signal to noise ratio of the received signal is greater than a predetermined value.
This measurement of the signal to noise ratio is performed continuously since the propagation conditions can vary, in particular as a result of variations in the meteorological conditions. For example, rain causes a high attenuation of the received signal compared with a transmission in clear weather. It can also be pointed out that the propagation conditions can be degraded as a result of jitter which has its origin in multiple signal paths causing additive and subtractive combinations as well as maskings which occur when an antenna is following a moving source (the satellite) and obstacles are interposed on the path of the transmitted signal.
The accuracy of the measurement of the signal to noise ratio assumes great importance, since a low-accuracy measurement will lead to an excessive power allocated to each transmitter, which will reduce the communications capacity. On the other hand, if the measurement is accurate, there will be assigned to each transmitter the power just necessary for it, which makes it possible to maximize the communication resources.
The various methods used until now for measuring the signal to noise ratio provide a relatively poor accuracy.
The methods of estimating the signal to noise ratio which provide the best result are, in the case of digital signals, on the one hand, a correlation method and, on the other hand, a method of direct measurement of the received signal.
The first method (correlation) consists in correlating the received binary signal with a signal which is decoded, and then recoded. This is because it is known that a transmitted binary signal contains redundant information in order to allow a robust transmission of the information. For example, an ATM cell containing 424 bits is transmitted in 848 bits. The decoding consists in extracting the 424 useful bits and the coding consists of transforming the 424 useful bits into 848 bits intended to be transmitted. Thus, the received binary signal and the signal at the output of the coder have the same format and the same number of bits.
The correlation consists in performing a multiplication of the received signal S by the signal X at the output of the coder. The signal to noise ratio {circumflex over (&ggr;)} is determined from the mean value and the variance of this product Z=X.S measured for a series of N samples. This signal to noise ratio is then the ratio between the square of the mean and twice the variance. This method is limited by the performance of the decoder. When the link is of poor quality, the decoder provides an erroneous result and the measurement provided is then not reliable. Thus, the estimated signal to noise ratio has a correct value only when its value is sufficiently high.
The second method (a direct measurement on the received signal S) consists in determining the absolute value |s| of the samples and in estimating the signal to noise ratio from the mean and the variance of this absolute value of the samples, for example with the help of the following formulae:
μ
^
&LeftBracketingBar;
s
&RightBracketingBar;
=
1
N
∑
i
&LeftBracketingBar;
si
&RightBracketingBar;
(
1
)
σ
^
&LeftBracketingBar;
s
&RightBracketingBar;
2
=
1
N
-
3
∑
i
(
&LeftBracketingBar;
si
&RightBracketingBar;
-
μ
^
&LeftBracketingBar;
s
&RightBracketingBar;
)
2
(
2
)
γ
^
=
μ
^
&LeftBracketingBar;
s
&RightBracketingBar;
2
2
σ
^
&LeftBracketingBar;
s
&RightBracketingBar;
2
(
3
)
In these formulae, {circumflex over (&mgr;)} is the mean, {circumflex over (&sgr;)}
2
is the variance and {circumflex over (&ggr;)} is the signal to noise ratio.
The signal to noise ratio being determined from an absolute value, the results are satisfactory only when this ratio has a sufficiently high value. This is because, as only the absolute value is considered, the negative values fold back on to the positive values and, if the signal to noise ratio approaches zero, this folding back leads to an impairment of the statistical properties of the original signal and the estimates of the mean (1) and variance (2) are no longer adapted to the resulting signal.
SUMMARY OF THE INVENTION
The invention provides a method of estimating the noise which reduces, to a great extent, the estimation variance where the received signal comprises a number of codes.
To that end, according to the invention, a noise power is estimated on all the received codes and the mean value of the noise is determined, this mean value being used for estimating the signal to noise ratio for each code.
The invention results from the observation that the noises observed for each code are uncorrelated although coming from the same random process.
If M is the spread factor, or number of codes, the noise estimation variance is divided by M compared with the case where the noise estimation is performed individually on each code, without taking the other codes into account.
Moreover, an estimation of the noise power can be performed even if the receiver for which the estimation is being performed does not receive any signals intended for it. This is because this estimation can be carried out on the received codes which are intended for other receivers.
In order to estimate the signal to noise ratio from the noise thus estimated, the known methods of estimating means of signals can be resorted to.
In the case of a phase-modulated signal, notably with two or four states, the invention, according to another of its aspects, makes provision for estimating the mean and the variance of the signal x
I
2
+x
Q
2
, or the mean and the variance of the signals x
I
2
and x
Q
2
considered as one and the same random variable.
The invention provides a method of estimating the noise power of a given digital signal assigned a code, this signal being received by a receiver simultaneously with a plurality of other digital signals assigned different codes, which is characterised in that the noise powers are estimated for each of the received digital signals assigned codes, and in that there is assigned, to the given signal, the mean noise power which is the ratio between, on the one hand, the sum of the estimated noise powers and, on the other hand, the total number M of received codes.
The codes are for example orthogonal.
According to one embodiment, at least certain of the simultaneously received digital signals are inte
Calot Guillaume
Lapaille Cédric
Alcatel
Orgad Edan
Sughrue & Mion, PLLC
Vo Nguyen T.
LandOfFree
Method of estimating the signal to noise ratio of a digital... 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 of estimating the signal to noise ratio of a digital..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of estimating the signal to noise ratio of a digital... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3063547