Multiplex communications – Pulse position modulation
Reexamination Certificate
1999-05-06
2002-04-30
Hsu, Alpus H. (Department: 2662)
Multiplex communications
Pulse position modulation
C370S445000, C370S461000, C375S239000
Reexamination Certificate
active
06381213
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to the field of data networking. More specifically, the present invention relates to the detection of collisions between transmitted signals on a network.
BACKGROUND
Ethernet networks implement a Carrier Sense, Multiple Access/Collision Detect (CSMA/CD) protocol to ensure that only one node (or station) of the relevant Ethernet network is propagating a signal over a carrier medium of the network at any given time. As implied by the “Carrier Sense” aspect of this protocol, prior to or during signal transmission, a station checks or listens on the relevant carrier medium to determine whether another station is transmitting. Merely for example, an inquiring station may sense a voltage on a network line, and then assume that another station is transmitting.
In view of the distances between stations on a network and finite signal propagation times, it often occurs that a station wishing to transmit will sample the relevant carrier medium, and sense no traffic thereon, even though another station may have begun transmitting. Accordingly, two stations will simultaneously transmit, resulting in a collision of transmitted signals. The delay in detection of transmission of a signal over the carrier medium because of the spacing of stations thereon has resulted in a so-called “slot time” being defined as a function of the time required for the transmission of a signal by a first station, the possible “Carrier Sense” detection of this signal by a second station of the network, and the return of a collision detect signal by the second station to the first station. It will be appreciated that the possibility of colliding transmission signals occurs during this slot time.
In Ethernet networks where the carrier medium comprises a regular and well-designed wiring system, constructed using a high-quality cable, the detection of collisions may be performed by detecting voltage swings on the carrier medium that exceed the voltage that can be generated by one transmitter alone. However, where the carrier medium of the network displays a number of undefined characteristics, the above collision detection methodologies may be unreliable. For example, where the carrier medium comprises existing telephone wiring within a residence (e.g., Plain Old Telephone Service (POTS) wiring), the wiring may comprise unshielded, twisted-pair (UTP) telephone cable and may be either Category One or Category Two cable as defined by the EIA/TIA 568 specification. Accordingly, the wiring may be susceptible to noise. Further, the use of residential telephone wiring as the carrier medium for the network poses a number of problems as the transmission of data signals must occur over an arbitrary, unterminated wiring system with unknown electrical characteristics. This results in impedance mismatches, and the lack of termination contributes significantly to the persistence of reflections of signals and to unpredictable peaks and valleys in frequency response within a band of interest. The power-splitting effect of star configurations, which are commonly implemented in residential wiring, increases these problems.
SUMMARY OF THE INVENTION
According to one aspect of the present invention, there is provided a method of detecting a collision on a network. A first identifier of a first network device is transmitted over a carrier medium. The first identifier comprises a first series of pulses transmitted over a first time interval. Each pulse has a respective temporal location within the first time interval. A second identifier, comprising a second series of pulses, is received over a second time interval at the network device. A determination is made as to whether the first identifier corresponds to the second identifier and, if not, then a collision is detected on the carrier medium. The determination includes determining whether the respective temporal locations, within the first time interval, of each pulse of the first series of pulses corresponds to a respective temporal location, within the second time interval, of a pulse within the second series of pulses.
REFERENCES:
patent: 4706246 (1987-11-01), Kume
patent: 4785396 (1988-11-01), Murphy et al.
patent: 4817115 (1989-03-01), Campo
patent: 5151698 (1992-09-01), Pophillat
patent: 5162791 (1992-11-01), Heegard
patent: 5339307 (1994-08-01), Curtis
patent: 5450596 (1995-09-01), Aden
patent: 5515035 (1996-05-01), Gut
patent: 5548614 (1996-08-01), Stoll et al.
patent: 5587692 (1996-12-01), Graham
patent: 5614901 (1997-03-01), Haas
patent: 5657326 (1997-08-01), Burns et al.
patent: 5696790 (1997-12-01), Graham
Jeffress Jeremiah
Webber, Jr. Harold H.
Blakely Sokoloff Taylor & Zafman LLP.
Hsu Alpus H.
Tut Systems, Inc.
LandOfFree
Method and apparatus for detecting collisions on a network does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for detecting collisions on a network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for detecting collisions on a network will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2823901