Baseband data slicing method and apparatus

Pulse or digital communications – Receivers – Automatic baseline or threshold adjustment

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C327S306000, C327S307000

Reexamination Certificate

active

06349121

ABSTRACT:

TECHNICAL FIELD
The present invention relates generally to communication systems and, in particular, to a method and apparatus for performing baseband data slicing.
BACKGROUND OF THE INVENTION
Communications systems are well known in the art. In many such systems, information (e.g., voice or data information) to be conveyed from a transmitting communication unit to a receiving communication unit is represented as a baseband signal that, in turn, is used to modulate a carrier signal. At the receiving communication unit, a demodulation process is employed to extract the baseband signal from the carrier signal. Where the baseband signal is a representation of digital data, so-called data slicing is performed at the receiver in order to determine what binary digits have been received. In general, data slicing refers to the process whereby the recovered baseband signal is compared against a threshold to decide whether, for a given data period, a binary one or zero value has been sent. Various techniques for performing such baseband data slicing are known in the art.
FIG. 2
illustrates a prior art apparatus for performing AC-coupled baseband data slicing. In the present context, the term AC-coupled refers to the fact that only higher frequency components of the baseband signal are compared against the threshold. As shown in
FIG. 2
, a demodulator
202
using known demodulation techniques outputs a recovered baseband signal. The baseband signal output by the demodulator
202
includes, as is often the case, low frequency components including a DC component or offset. An exemplary baseband signal V
in
is illustrated in
FIG. 3
as a voltage varying over time. For the sake of simplicity, the baseband signal
302
shown in
FIG. 3
is represented as a single sinusoidal waveform. In practice, however, such baseband signals typically comprise a plurality of frequency components, resulting in a more complex time domain waveform. As shown, the baseband signal comprises an offset component V
DC
304
. In an ideal system, the baseband signal is assumed to adhere to a 50% duty cycle. That is, a binary one value (represented, for example, by a positive voltage) has a duration in the time domain baseband signal equal to a binary zero value (represented, for example, by a negative voltage). In order to accurately compare the received baseband signal against a threshold, and still assuming the 50% duty cycle constraint, the AC-coupled data slicer filters out all low frequency components, including the offset V
DC
using a high pass filter
204
. In effect, the high pass filter causes the baseband signal to become centered upon a zero voltage value, rather than a DC offset value. This is illustrated in
FIG. 4
, where the high pass filtered version of baseband signal
402
is shown over time.
As known in the art, such high pass filters are characterized by a settling time constant which causes DC components in the input signal to be removed from the filter output according to a decaying exponential curve. This effect is illustrated in
FIG. 4
where the DC component
404
is gradually removed from the filtered baseband signal
402
. Progressively greater attenuating effects by the high pass filter are typically characterized by a correspondingly increased settling time constant that implies, in turn, that the filter output will increasingly lag behind the filter input. As such, filter design must typically strike a balance between the desired attenuation and the settling time constant that the system is able to endure.
Referring again to
FIG. 2
, the high pass filtered baseband signal is provided to a comparator
206
that compares the filtered signal against a threshold V
th
. Because the high pass filter
204
has the effect of centering the baseband signal on a zero voltage or ground level, V
th
is preferably set to that level. Typically, the comparator will output a predetermined positive voltage for any input signal above the threshold and a predetermined negative voltage (of same magnitude) for any input below the threshold. For example, assuming that the filtered baseband signal
402
of
FIG. 4
is applied to the comparator of
FIG. 2
, an exemplary output V
out
502
is shown as the heavy dashed line in FIG.
5
. Note that the output
502
of the comparator
206
does not accurately track the baseband signal
302
output by the demodulator
202
due to the settling time constant of the high pass filter. In an ideal system, in which the settling time constant where negligible, the output of the comparator
206
would correspond to the curve having reference numeral
504
in FIG.
5
. In a typical application, however, a significantly long predetermined delay
406
(typically several times the length of the settling time constant) must pass before the data output of the comparator may be considered reliable. As a result, data may be lost at the beginning of a received signal. To combat this problem, a sufficient amount of dummy data may be inserted as a preamble to the baseband signal being transmitted such that the processing of the dummy data at the receiver allows the predetermined delay to pass before actual baseband data is demodulated and sliced. However, this approach builds in a fixed signal delay that may not be acceptable in all applications. A further difficulty with this approach is that long periods of zeros or ones represented by the baseband signal are seen by the high pass filter as a DC component to be filtered out, thereby attenuating the desired signal, which results in decreased system performance.
An alternative to the AC-coupled method is the DC-coupled method of FIG.
6
. In this method, an unfiltered version of the received baseband signal is provided to the comparator and a filtered version of the DC offset is provided as the threshold provided to the comparator. In a sense, instead of filtering the baseband signal to the correct threshold level as in the AC-coupled method, the DC-coupled method filters the baseband signal to determine the necessary threshold level. To this end, a low pass filter
604
is used to attenuate all higher frequency components from the baseband signal, preferably leaving only the DC offset to be used as the comparator threshold input. However, as with the high pass filter
204
, the low pass filter
604
is characterized by a settling time constant, implying that there is a lag between the time that the DC offset is applied to the low pass filter and the time that it is reflected in the output signal. This is illustrated in
FIG. 3
where the output of the low pass filter V
lpf
306
is shown as a dotted line. Because of the settling time constant, the output of the comparator in
FIG. 6
will be similar to the filtered output
502
shown in FIG.
5
. That is, the comparator output will not be reliable until after the predetermined delay has passed. Furthermore, the long strings of ones or zeros in the baseband signal will cause the output of low pass filter
604
to drift, causing a corresponding change in the threshold level. In essence, the DC-coupled approach illustrated in
FIG. 6
suffers from the same disadvantages as the AC-coupled approach.
A variant of the DC-coupled approach is to set the threshold to an initial, predetermined value equal to the expected offset value in the demodulator output, rather than low pass filtering to determine the threshold. This approach suffers, however, in the event that perturbations resulting from the transmission channel or the receiver front-end cause the actual offset value to differ from the assumed offset value, thereby resulting in the use of a non-optimal threshold value. To combat the possible drift of the offset value, yet another approach is to continuously adjust the threshold level as illustrated in FIG.
7
.
The threshold is initially set to a predetermined value V
th
(0). As shown in
FIG. 7
, the output of the comparator is passed through a low pass filter
708
that has the effect of averaging the duty cycle of the comparator output. If the threshold is, for example, too low,

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Baseband data slicing method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Baseband data slicing method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Baseband data slicing method and apparatus will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2949564

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.