Pulse or digital communications – Receivers – Automatic baseline or threshold adjustment
Reexamination Certificate
1998-04-30
2002-03-19
Pham, Chi (Department: 2631)
Pulse or digital communications
Receivers
Automatic baseline or threshold adjustment
Reexamination Certificate
active
06359941
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to digital data receivers for receiving burst mode digital data, and more particularly to a system and method for reducing the burst mode penalty and output signal duty cycle distortion suffered by the receiver during digital data transmissions by automatically selecting an advantageous reference threshold.
2. Description of the Related Art
Digital optical communication between modem computer systems may be accomplished using either continuous or burst mode data transmissions. Conventional AC-coupled optical receivers are typically used for continuous data transmissions, while DC-coupled optical receivers are used for burst mode transmissions. Passive optical networks utilizing burst mode data transmission have proliferated in recent years. Because AC-coupled receivers are generally superior in sensitivity and performance to DC-coupled receivers, attempts have been made to adapt AC-coupled receivers for use with burst mode data transmissions. Typically this has been accomplished by encoding burst mode transmitted data to enable an AC-coupled receiver to interpret and process the transmissions. However, in passive optical networks the transmission media must be shared by several users—each user is allowed to transmit in a dedicated time slot only and is required to be “silent” outside the dedicated time slot. Thus, data encoding is not possible in passive optical networks because an encoded data signal would not exhibit the required time slot information.
DC-coupled receivers that receive burst mode data transmissions suffer from a decrease in sensitivity and signal power that is often referred to as a “burst mode penalty”. Typically, by the time a digital data signal is received by a DC-coupled receiver, the signal pulse shapes are degraded to an analog-type pulse shapes having uncertain amplitudes. In previously known DC-coupled receivers, the digital data signal is compared to a fixed reference threshold voltage in a decision circuit of the receiver to recover the pure digital signal. Thus, when uncertain and widely varying signal amplitudes are compared to the fixed reference threshold voltage, identification of logic ONEs and ZEROes is erratic and results in a high burst mode penalty and distortion. Accordingly, in recent years attempts have been made to develop techniques to improve the ability of a DC-coupled receiver decision circuit to identify logic ONEs and ZEROEs with greater certainty to improve receiver performance characteristics. In particular, industry efforts have been directed to reducing the burst mode penalty.
One such technique, described in U.S. Pat. No. 5,025,456 to Ota et al., provides a DC-coupled receiver with adaptive threshold circuitry for providing a varying reference threshold voltage that adapts to the amplitudes of the received digital data signal. The reference threshold voltage amplitude is set to one half of the minimum and maximum excursion of the data signal. Thus, the reference threshold voltage automatically follows the changes of amplitudes in the data signal and provides improved identification of logic ONEs and ZEROes, resulting in a significant reduction in the burst mode penalty suffered by the DC-coupled receiver.
However, the adaptive threshold approach suffers from a significant disadvantage. All digital data receivers are subject to noise that is generated by a variety of internal and external sources. Thus, one of the requirements in a DC-coupled burst mode receiver is that the receiver output must be “silent”, i.e. at zero amplitude without a signal at the input of the receiver, to reduce or eliminate the noise. This is functionally accomplished by applying an extra offset voltage at the threshold level of sufficient amplitude to overcome the input and internal noise. Accordingly, the adaptive threshold is actually a sum of the offset voltage amplitude and one half of the difference between the maximum and minimum excursions of the received data signal. Thus, the reference threshold level is not actually in the desirable middle position between the maximum and minimum excursions of the data signal, but is then always above the middle position. As a result, a DC-coupled receiver equipped with adaptive threshold circuitry still suffers from degradation in sensitivity/signal power (i.e. burst mode penalty) of at least 3 dB. Another disadvantage of the adaptive threshold approach is that significant duty cycle distortion is present at the receiver output when low amplitude data signals are received at the receiver input.
Thus, it would be desirable to provide a DC-coupled receiver with the ability to automatically set an advantageous reference threshold that reduces the burst mode penalty suffered by the receiver and that reduces duty cycle distortion at the receiver output when low amplitude data signals are received at the receiver input.
SUMMARY OF THE INVENTION
In accordance with the present invention, a system and method for automatically setting an advantageous reference threshold in a burst mode receiver to reduce the burst mode penalty associated with burst mode optical data transmissions, and to reduce duty cycle distortion at the receiver output, are provided.
The system of the present invention is implemented in a burst mode digital data receiver having an input and an output. The system includes an optional transimpedance preamplifier, connected to the receiver input, for amplifing a received digital data signal; an offset generator, connected to the preamplifier, for generating an offset voltage of sufficient amplitude to eliminate noise at the receiver output, when the data signal is absent at the receiver input; a signal processor, connected to the offset generator, for automatically setting the reference threshold voltage to an advantageous value in accordance with the invention; and an optional output amplifier connected to the preamplifier and the signal processor for amplifying the digital data signal before the signal is processed by other receiver circuitry.
In a preferred embodiment of the invention, the signal processor compares the amplitude (i.e. maximum excursion) of the received data signal to an offset threshold equal to approximately twice the offset voltage amplitude. If the data signal amplitude is less than the offset threshold, then the signal processor sets the reference threshold voltage approximately equal to the offset voltage. If, on the other hand, the data signal amplitude is greater than or approximately equal to the offset threshold, then the signal processor sets the reference threshold to exactly one half of the difference between the maximum and minimum excursions of the data signal. Thus, when the data signal amplitude is very low (i.e. lower than twice the offset voltage amplitude) the data signal is assumed to be generated by noise and the reference threshold at the output will be substantially equal to the offset voltage, producing a silent receiver output. When, on the other hand, the data signal substantially equals or exceeds the offset threshold, the reference threshold is set to exactly one-half of the maximum and minimum excursions of the data signal, without the addition of the offset voltage to the threshold as is done in the previously known adaptive threshold approach. This arrangement virtually eliminates the burst mode penalty, thus improving receiver sensitivity, and furthermore eliminates duty cycle distortion.
In another embodiment of the invention, the signal processor sets the reference threshold to a fractional value of the following expression:
[(maximum data signal excursion−minimum data signal excursion)+offset voltage]
The fractional value is selected so that the resulting reference threshold is approximately equal to one-half of the difference between the maximum and minimum excursions of the data signal. As a result, the reference threshold is set to the desirable mid-point position between the maximum and minimum excursions without the undesira
Burd Kevin M
Lucent Technologies - Inc.
Pham Chi
LandOfFree
System and method for improved reference threshold setting... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for improved reference threshold setting..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for improved reference threshold setting... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2834022