The present invention relates to a method and device for setting the slice level in a binary signal.
The two signal levels of a binary signal are usually referred to as “high” and “low” and may represent a logical “1” and “0” respectively. The “high” level may correspond to a signal level of, for example, +5V while the “low” level may correspond to a signal level of, for example, −5V or ground. When recovering a transmitted binary signal it has to be decided which signal portions are high and which are low. To this end, a threshold is usually set approximately halfway between the high and the low signal levels. Any signal level exceeding this threshold or “slice level” is considered to represent a high level, other signal levels are categorized as low.
In the presence of noise it is possible that errors are introduced. Noise peaks present in the low level signal portions may exceed the slice level so as to incorrectly cause a high level to be detected. The prior art offers only a partial solution to this problem.
U.S. Pat. No. 4,707,740 discloses a sync detector for recovering a sync signal from a video signal. A slice level signal is adjusted during a low level (“sync tip”) portion of the video signal. To this end, a noise detector provides an output representative of the average noise during this low level signal portion. This noise detector output is used to generate a positive slice level offset during the low level signal portions and a negative offset of the same magnitude during the high level signal portions.
This prior art solution has several disadvantages. The slice level adjustment is discontinuous, thus compounding any detection errors. In addition, the adjustment is based on the noise level in the low level signal portions only. As the noise level increases, the slice level will be pushed up indiscriminately towards the high signal level and may even reach the high signal level, which is clearly undesirable. Moreover, the positive and negative offset of equal magnitude assume a symmetrical noise distribution, that is, the noise peaks in the high level signal and in the low level signal having (on average) the same magnitude. However, there are many applications in which the noise has an asymmetrical distribution, the noise peaks in one signal level having (on average) a different magnitude to those in the other signal level. An example of such an application is optical communication links.
It is therefore an object of the present invention to solve the problems of the prior art and to provide a method for setting the slice level in a binary signal which takes the noise level in both the low level and the high level signal portions into account.
It is another object of the present invention to provide a method for setting the slice level in a binary signal which appropriately deals with asymmetric noise distributions.
It is a further object of the present invention to provide a device for carrying out the method of the present invention, and a receiver provided with such a device.
The invention is defined by the independent claims. The dependent claims define advantageous embodiments.
By measuring the noise level during the second signal portion as well as during the first, it is possible to take both noise levels into account when adjusting the slice level, thus achieving a more balanced adjustment. By applying a continuous slice level adjustment in both the first and the second signal portions, the detrimental effects of any erroneous discontinuous adjustments are avoided.
In a preferred embodiment, the slice level is set at a value substantially equal to half the difference between the magnitudes of the first and the second signal levels minus half the difference between the magnitudes of the first and the second noise levels. That is, the slice level is preferably initially set halfway between the signal levels, the adjustment being equal to half the difference in the noise levels. It will be clear that if the signal levels are equal in magnitude but of opposite signs, their difference will be equal to zero. Similarly, if the noise levels are equal in magnitude, no adjustment will be made. However, if the noise levels are not equal in magnitude (asymmetrical noise), the slice level is adjusted accordingly, resulting in fewer detection errors.
It is preferred that measuring the respective noise levels involves detecting peaks in the binary signal. By using peak detection, a good indication of the noise levels can be easily achieved.
The present invention also provides device for detecting the noise level in a binary signal, comprising a noise peak level detection means for receiving input signals and producing a noise indication signal, characterized in that the noise peak level detection means comprise an RMS level detector for detecting the RMS level of the binary signal, a first differential amplifier for amplifying the difference of the first level of the binary signal and the RMS level to supply a first level compensated noise signal to the first peak detector, a second differential amplifier for amplifying the difference of the second level of the binary signal and the RMS level to produce a second level compensated noise signal to the second peak detector. A noise level detector of this kind has the advantage that it can detect very small noise peaks (smaller than 50 mV) which Prior art noise detectors often fail to detect.
The present invention will be further explained below with reference to exemplary embodiments illustrated in the accompanying drawings, in which:
The binary signal T shown in
Setting the slice level halfway between the average positive and the average negative signal level, which corresponds to signal level “ground” in
The signal T shown in
In accordance with the present invention the slice level SL is set so as to maximize the distance between the slice level and the noise peaks. When the slice level SL is set at the common signal level G, the distance between the noise peaks X and the slice level SL is smaller than the distance between the noise peaks Y and the slice level. As a consequence, there is an unnecessary large probability that a noise peak X will cross the slice level, resulting in a detection error. However, when the slice level SL is set as shown in
D=A−X+Z=B−Y−Z
from which it follows that:
Z=½(B−A)+½(X−Y).
When the signal magnitudes are equal (A=B) it follows that:
Z=½(X−Y).
In other words, when the signal magnitudes are equal, the offset Z is equal to half the difference in noise (peak) magnitudes. In the case of symmetrical noise (X=Y) this means that the offset is equal to zero, as before.
A device for setting the slice level in accordance with the present invention is depicted in block diagram form in
The noise indication signal Vcon is passed on to both the first level shifter 11 and the second level shifter 12. In the latter, this signal is used to subtract a DC level from the input signal components IN and INQ as described above, said DC level being equal to Vcon. In the first level shifter 11 the signal Vcon is used to add a DC level equal to Vcon to the input signal components IN and INQ resulting in output signal components IN′ and INQ′ at the output terminals 16 having magnitudes of IN′=IN+Vcon and INQ′=INQ−Vcon respectively. This is the signal T having the offset slice level shown in
As can be seen, the adjustment of the slice level using the noise indication signal Vcon is continuous. When the noise characteristics change, the slice level is adjusted accordingly. Even a change in the type of noise, e.g. from asymmetrical to symmetrical, is taken into account. The detected signal is not used to set the slice level, thus avoiding the possibility of compounding detection errors.
An advantageous embodiment of the device of
In
A particularly advantageous embodiment of the peak detection unit 13 is illustrated in
It will be understood by those skilled in the art that many modifications and additions may be made without departing from the scope of the invention as defined in the appending claims. It is remarked that the scope of protection of the invention is not restricted to the embodiments described herein. Neither is the scope of protection of the invention restricted by the reference numerals in the claims. The word ‘comprising’ does not exclude other parts than those mentioned in the claims. The word ‘a(n)’ preceding an element does not exclude a plurality of those elements. Means forming part of the invention may both be implemented in the form of dedicated hardware or in the form of a programmed purpose processor. The invention resides in each new feature or combination of features.
Number | Date | Country | Kind |
---|---|---|---|
02078148.0 | Jul 2002 | EP | regional |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB03/02934 | 6/27/2003 | WO | 1/26/2005 |