Robust adaptive equalizer

Details Robust adaptive equalizer Robust adaptive equalizer

1999-11-12

2003-06-24

Corrielus, Jean (Department: 2631)

Pulse or digital communications

Equalizers

Automatic

C375S316000, C708S323000

Reexamination Certificate

active

06584151

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to signal processing as often used in communication systems. More specifically, the present invention relates to equalizers and permits adaptive control of an equalizer using robust techniques for operation in high noise environments such as is common in wireless, microwave, cable and other communication systems.
2. Discussion of Related Art
Communication systems communicate information between a transmitter and a receiver. A broad class of communication systems communicate information by modulating and demodulating data over a communication medium. The information is most frequently digital data, i.e., data that is represented as a sequence of binary digits (“bits” having values of 1 or 0). The digital information is transmitted as an analog signal over the communication medium. Such an analog signal may include electronic or electromagnetic signals as well as optical signals (light transmitted over optical communication media).
Typically, a carrier signal is transmitted over the communication medium and is modulated in such a manner to represent the digital data to be transferred between the devices. That is, an analog attribute of a carrier signal is modulated by the transmitter to encode data over the communication medium and demodulated by the receiving device to receive the data. For example, the amplitude or frequency of an electronic or electromagnetic signal may be modulated higher or lower from a carrier base level to encode the data. The receiving device then detects the modulations of the carrier signal attribute to decode (demodulate) the data encoded within the received signal.
The processing of the received signal to demodulate and decode the transmitted information is generally referred to as signal processing. Such signal processing is often performed by periodically sampling the received analog signal, converting the sampled signals to digital representations of the analog signal waveform and then processing the digital sampled values using digital signal processing techniques to demodulate and decode the information encoded in the received signal.
In many such communication systems analog noise signals or other interfering signals are superposed on the transmitted signals by environmental effects. For example, ambient electromagnetic signals, ubiquitous in our modern environment, may interfere with the effective transmission of information via the communication medium. Or, for example, electrical or other properties of the communication medium may degrade the quality of the received signal as data transmission rates are increased and/or as transmission distances are increased.
In such noisy environments, it is common in communication systems to process the signals in such a manner as to reduce the effects of such noise at the receiving end and to thereby improve the efficacy of data transmission between the transmitting and receiving devices.
It is known in the art to use a decision device (circuit), commonly referred to as a slicer, to detect the signal transmitted without the effect of superposed environmental noise. In general, a slicer detects that a sampled signal value has or has not made a transition to a predetermined signal threshold value. Such modulation transitions in the received signal are representative of the encoded digital information. Noisy environments or interfering signals as noted above can cause the slicer to erroneously detect the transmissions and therefore the encoded digital information.
To improve the efficacy of the slicer, it is known to use an equalizer—a filter adapted to improve the quality of the signal for purposes of slicer data decoding. An adaptive equalizer is one that may be adapted to modify its filtration characteristics in accordance with a control input signal applied to the equalizer. An equalizer and slicer are generally configured in a feedback configuration such that the equalizer is adapted to change its filtration based on an error calculated at the receiver by comparing the slicer input and perceived transmitted data with the actual transmitted information (expected information).
Conventional adaptive equalizers as known in the art are ineffective in high noise environments. Their stable operation depends on relatively reliable detection of transmitted signals by the slicer circuit that provides the feedback control signal to the equalizer (i.e., nearly correct detection as may be possible only in a low noise environment). Such conventional equalizers use estimates of the transmitted signal generated by the slicer to guide the adaptation of the equalizer filtration. In high noise environments the slicer estimates may be so inaccurate as to render the equalizer ineffective. The equalizer adaptation may diverge far from a desired optimum performance range due to such erroneous slicer estimates.
A number of approaches are known in the art to prevent such divergence of the equalizer. A first approach uses circuits downstream from the slicer, often referred to as a decoder, for more accurate estimates of the transmitted signal. The decoder circuit receives the slicer output and further decodes the detected information in accordance with known rules of encoding of information on the communication medium. The decoder therefore can more accurately detect the transmitted data because it understands the rules of data encoding—it distinguishes expected, legal sequences as compared to unexpected or illegal sequences. However, this approach has a negative aspect in that the decoder makes its decisions based on sequences of transitions as distinct from slicer decisions based on real time sampled values. The decisions from such a decoder are based on a sequence of samples and thus sample values must be buffered to adapt the equalizer before being passed on to the adaptive equalizer for filtration. In effect a first decoder is used with buffering in advance of applying the samples to the adaptive equalizer then a second decoder receives the delayed equalized output values from the equalizer. This added delay (latency) in adapting the equalizer filtration renders the equalizer ineffective for adapting to rapid changes in the noise levels. Further, this approach adds the need for a memory associated with the equalizer to store received sequences of sampled signals for later adaptation.
A second approach to avoid divergence of the equalizer filtration is to use only certain types of received signals for purposes of equalization adaptation. For example, where the received signals are known to be less reliable (i.e., high order modulation) adaptation is disabled. Adaptation is enabled only when received signals are known to be more reliable (i.e., low order modulation). Or for example, special data sequences, commonly known as training sequences, may be injected into the data transmission to allow adaptation at predetermined times with predetermined sequences known to be less sensitive to noise or interference. Such approaches suffer from problems adapting to rapidly changing noise or interference signals. In addition, injection of special adaptation sequences decreases the overall effective data rate for the communication channel by utilizing a portion of the available bandwidth for sequences carrying no user data. Further, such approaches are complex in that they must assure synchronization to distinguish the signals for which adaptation is enabled from signals for which adaptation is disabled. Lastly, such approaches require storage or generation circuits for the sequences to be used for enabling or disabling adaptation.
It is apparent from the above discussion that a need exists for improved circuits and techniques for adaptation of equalization filters that can improve equalizer adaptation in high noise environments.
SUMMARY OF THE INVENTION
The present invention solves the above and other problems, thereby advancing the state of the useful arts, by providing an equalizer adaptation circuit and method more immune to high noise environments

Affiliated with

Also associated with

Rating

Robust adaptive equalizer does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Robust adaptive equalizer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Robust adaptive equalizer will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3126425