Multiplex communications – Channel assignment techniques – Polling
Reexamination Certificate
1997-12-01
2002-07-02
Yao, Kwang Bin (Department: 2664)
Multiplex communications
Channel assignment techniques
Polling
Reexamination Certificate
active
06414964
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method and apparatus for performing a multipoint polling protocol and, more particularly, to a method and apparatus for performing a multipoint polling protocol which uses silence as a polling response and to control other aspects of circuit operation.
BACKGROUND OF THE INVENTION
Traditional polling protocols for multipoint circuits require that a polled tributary always respond with a message regardless of whether the data terminal equipment (DTE) associated with the polled tributary has any real data to send. A control device located at one end of the communication line manages the bandwidth on the subscriber line by polling each tributary. When a polled tributary fails to respond to a poll, the control device assumes that the poll message was corrupted.
Responding to polls is a function of the link layer. The link layer normally is implemented in the DTE, which is physically separated from the device which performs the modulation layer, i.e., the data communication equipment (DCE). Normally, the DCE does not interpret the contents of received messages and does not know the address assigned to the DCE. Therefore, the DCE relies entirely on the DTE to determine if and when to transmit a message to the control device. The standard handshake between DTE and DCE is such that there is no way for the DTE to tell the DCE before a poll arrives whether or not it has data for the DCE to send. Furthermore, there has not been a recognized standard or specification prescribing the maximum time delay between the time that the control device issues a poll and the time that the poll message was corrupted or that the DTE is otherwise unable to respond. In order to accommodate unknown processing delays in the DTE when responding to a poll, the silent time interval would have to be very large relative to the time needed to send a short non-silent message. Otherwise, a long processing delay in the DTE would be incorrectly interpreted as indicating that the DTE has no response. The accumulation of these delays when polling several drops would significantly degrade the response times seen by network users. All of these factors have made it impractical, if not impossible, to use silence as a valid response to indicate that the DTE has no data to send.
The use of silence as a poll response, if it could be utilized in a practical manner, would have several advantages over current multipoint polling protocols. As stated above, with typical multipoint protocols, a tributary is required to send a message in response to every poll. If the poll response is corrupted, the control device is obligated to poll again, even if the tributary actually had no data to send. A silence period is less likely to be corrupted and more likely to be correctly interpreted than is a no-response data message. Similarly, if any part of a message containing a poll is corrupted, the entire message must be disregarded. Interpreting a brief silence period as a no-data message allows the control device to quickly transmit again, thereby keeping the line busy and avoiding lengthy idle time-outs.
Another advantage of using silence as a no-data poll response is that it can minimize the amount of bandwidth required to be used for polling overhead. For example, a large number of terminal devices connected to the subscriber line that are not completely inactive but have low levels of activity can be efficiently polled by a single, group polling message. Since these low-activity devices typically do not have data to send, using silence as the poll response would allow all of these devices to respond correctly at the same time. In contrast, if the devices were required to respond with a non-silent message, only one device would be able to respond at a time. Therefore, using silence as a poll response could reduce the amount of line bandwidth required for polling tributaries which are unlikely to have data to send.
A further advantage of using silence as a polling response is that detecting the presence or absence of silence requires less processing time than to do all of the receiving functions of the physical and link layers of the line protocol. The functions of the DCE and/or DTE are typically executed on a programmable processor. In some implementations, this processor may be shared among several tasks in addition to those required just to support the DCE and/or DTE functions. When silence is used as a polling response, less of the available processing time of the shared processor is required to perform the DCE and/or DTE functions especially when “nothing to send” is the most frequent response, which is typically the case. It is possible to use some auxiliary hardware circuitry to detect the transition from silence to “data” so that the shared processor is only required to do the demodulation functions when something other than “nothing to send” is received in response to a poll.
The use of silence in a multipoint circuit can be extended to control other aspects of circuit operation. This makes use of the advantage that the length of silence can be easily measured and the measured length can provide useful information to the receiving station. In some cases it is convenient to use a shared bus circuit to provide the connection between the stations on the multipoint line. Examples of such a circuit are a telephone subscriber line or the wiring used in some types of local area networks (e.g., Ethernet). In these types of circuits, only one tributary station transmits on the shared line at a time and all connected tributary stations receive the signal transmitted from any of the other connected stations. While the tributary stations can receive signals from each other, they may not be able to demodulate these signals and typically the tributaries communicate only with the control station and not directly with each other. In this case, it is necessary for the tributaries to be able to determine whether or not the next transmission is from the control station rather than from another tributary station without being able to demodulate messages from other tributaries. If the transmission is from the control station, the DCE and DTE functions must be performed for the currently received message. If the transmission is from another tributary station, the signal can be disregarded, thereby making the processor available for other functions. As will be shown below in the discussion of the present invention, measuring the length of the silence time provides a convenient way for the tributary stations to make this distinction.
Yet another advantage of using silence to control circuit operations in a multipoint communications system is that it can be used to indicate the type of modulation that will be used next on the line. In some multipoint communications systems there may be a benefit to using more than one type of modulation. For example, a modulation capable of transmitting very high data rates may be used by stations requiring these rates. However, other stations which do not require high data rates may use a lower speed modulation which can be more economically implemented. These stations would, of course, not be able to use the high speed modulation. In accordance with the present invention, it has been determined that when the multipoint circuit is provided by a shared bus, silence can be used to indicate the type of modulation that will be used on the line next. The control station, which is capable of using either type of modulation, can indicate, via the length of the silence interval between the last and the next transmission, which of the two types of modulation it will be using in the next transmission.
SUMMARY OF THE INVENTION
In accordance with the present invention, a method and apparatus is provided for performing a multipoint polling protocol which uses silence for controlling circuit operation. In accordance with the present invention, when a tributary station which has no data to send is polled by a control station, the tributary station does not send any
Betts William L.
Holmquist Kurt E.
Paradyne Corporation
Thomas Kayden Horstemeyer & Risley LLP
Yao Kwang Bin
LandOfFree
Method and apparatus for performing a multipoint polling... 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 performing a multipoint polling..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for performing a multipoint polling... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2871989