Pulse or digital communications – Receivers – Particular pulse demodulator or detector
Reexamination Certificate
1999-11-19
2003-07-22
Bayard, Emmanuel (Department: 2631)
Pulse or digital communications
Receivers
Particular pulse demodulator or detector
Reexamination Certificate
active
06597749
ABSTRACT:
TECHNICAL FIELD
This invention relates to pulse monitoring systems and in particular to systems for detecting the pulse width of received signals.
BACKGROUND ART
Transmitted digital signals need their pulse widths to remain within a prescribed tolerance in order for the receiving circuitry to be able to ascertain the data content of the transmitted signals. A transmitted digital signal may become distorted by frequency response effects of the transmitting medium, by loss of energy over distance, electromagnetic interference, etc.
With reference to
FIG. 1
, two sample pulse sequences are shown. Both begin with a logic low and transition to a logic high. In the case of a logic low, TLmin denotes the minimum allowed pulse length of a logic low pulse and TLmax denotes the maximum allowed pulse length for a logic low pulse. In order for a receiving circuit to accept a logic low to high transition, the incoming pulse low signal should transition within a window of time defined by the difference of TLmax and TLmin. This window of time is defined as logic low transition window, W
0
, in FIG.
1
. Similarly, a logic high pulse also needs to have a pulse length defined by a minimum logic high pulse length, THmin, and a maximum logic high pulse length, THmax. In order for a logic high to low transition to be accepted, the transition should occur within a window of time defined by the difference of THmax and THmin. This window of time is defined as a logic high transition window, W
1
. As shown with the two sample pulse sequences of
FIG. 1
, the time windows W
0
and W
1
need not be the same for different applications, although they typically should remain constant within a single application.
The receiving circuit includes a pulse detecting sub-circuit to observe the pulse length of received pulse signals. Prior art pulse detecting circuit typically observe only one of the logic high or logic low pulse of a received signal, but not both. This means that the pulse detecting circuit cannot determine the signal period of the received pulse signal, and thus cannot observe shifts in the signal's frequency. Frequency shifts in a received pulse signal can therefore lead to undetected errors.
Prior art pulse detecting circuits typically use two one-shot circuits to determine if a received signal is valid for a predetermined transition window. Both one-shot circuits are typically triggered on the leading edge of an received pulse signal. A first one-shot circuit issues a first pulse having a length equal to the minimum pulse length requirement, and the second one-shot circuit issues a second pulse having a length equal to the maximum pulse length requirement. At the end of the second pulse, the received signal is compared with the first and second pulse. If the length of the received pulse does not lie within the first and second pulse, then no detection signal is issued and the received pulse signal is ignored. One shot-circuits, however, are difficult to control or adjust in integrated circuitry. Additionally, they do not provide for a simple way of adjusting the required transition window for different applications.
This prior art approach can also slow down a system since the pulse detecting circuit waits for the lapse of the maximum pulse length before verifying the received pulse signal. In effect, it waits the maximum amount of delay time for each received pulse signal even if the received pulse signal transition earlier and do not require any additional wait time. U.S. Pat. No. 3,735,271 to Leibowitz shows a pulse width detecting circuit that does not wait for the maximum allowable pulse length before testing for a valid received pulse signal. The '271 circuit, however, requires three one-shot circuits and adds a delay to each received pulse signal. This added delay may actually cause additional errors. Assuming the circuit receives a pulse signal having a pulse length slightly shorter than the minimum require pulse length, the added pulse delay introduced by '271 may actually cause the circuit to think that the received signal did meet the minimum pulse length requirement.
It is an object of the present invention to provide a pulse detecting circuit that observes both a received pulse signal's pulse length and its frequency.
It is another object of the present invention to provide a pulse detecting circuit that observes both logic high and logic low pulse of a received pulse signal.
It is yet another object of the present invention to provide a pulse detecting circuit that lends itself to simple integration onto an IC circuit and that permits its allowable pulse transition windows W
0
and W
1
to be independently, or jointly, adjusted from inside and from outside the IC circuit.
SUMMARY OF THE INVENTION
The above objects are met in a pulse detection circuit that uses a pair of linear voltage ramp generators and a reference voltage source for establishing a preferred pulse length window for a received pulse signal. The present pulse detection circuit monitors both the positive duty cycle and the negative duty cycle of a received pulse signal. Therefore, it can detect frequency shifts in the received pulse signals. If a positive or negative duty cycle is found to be either too short or too long, it is categorized a “bad” duty cycle and an error signal is issued that disables the internal circuitry of a chip. If the duty cycle lies within a predetermined time window, i.e. a preferred pulse transition window, it is categorized as a “good” duty cycle. If an error signal is issued due to encountering a bad duty cycle, the internal circuitry of the chip remains disabled until two consecutive good duty cycles are observed. When the present pulse detection circuit receives two consecutive good pulses, i.e. good high pulse following a good low pulse, or a good low pulse following a good high pulse, it issues an enable signal to re-enable the internal circuitry of the chip.
The present duty cycle monitoring circuit includes two monitoring sub-circuits. A first sub-circuit monitors the positive duty cycle of an incoming pulse signal, and a second sub-circuit monitors the negative duty cycle of the incoming pulse signal. Since the positive and negative duty cycles are monitored by separate sub-circuits, the present pulse detection circuit can support different preferred pulse transition windows for positive duty cycles and for negative duty cycles. For the sake of brevity, only the first sub-circuit for monitoring a positive duty cycle described here. The circuit structure for the second sub-circuit is similar to the first sub-circuit and is discussed in detail in the best mode description below.
A preferred pulse transition window, W
1
, for a “good” positive duty cycle is first defined. This pulse transition window is established by means of a slower linear voltage ramp and a faster linear voltage ramp working in unison. As time passes, the voltage difference between the slower and faster linear voltage ramps increases, and this voltage difference becomes a measure of the a duty cycle duration. A desired duration window is established by noting the voltage value of the slower linear voltage ramp when the voltage difference between the slower and faster linear voltage ramps becomes representative of the desired pulse transition window. This noted voltage value becomes a reference voltage to which an incoming clock is compared. In other words, the length of a pulse transition window can be adjusted or shifted in time by changing the reference voltage or by adjusting the slope of the linear voltage ramp generators. The reference voltage may be generated from within the circuit or may be supplied from a source external to the circuit.
As logic state of an incoming clock changes from low to high, the slower and faster voltage ramps are reset and re-triggered. A first comparator is used to compare the rising voltage value of the faster voltage ramp with the reference voltage. The first comparator signals when the faster voltage ramp rises above the reference voltage. A second com
Atmel Corporation
Bayard Emmanuel
Schneck Thomas
LandOfFree
Digital frequency monitoring does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Digital frequency monitoring, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Digital frequency monitoring will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3047487