Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1996-09-09
2001-06-05
Chin, Wellington (Department: 2664)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S428000
Reexamination Certificate
active
06243380
ABSTRACT:
FIELD OF THE INVENTION
The invention relates generally to computer networking systems, and, in particular, to addressing frames of data.
BACKGROUND OF THE INVENTION
Typically, when a remote node
152
uses a service provider to establish a remote connection with a separate network, the remote node
152
will dial into the service provider's local Network Access Server (NAS)
154
. As shown in
FIG. 1
, the NAS
154
usually includes several port interfaces. Each remote node dials into the NAS
154
through a separate port interface. Data received through interfaces may in turn be multiplexed over one synchronized line
156
providing a connection to a Wide Area Network (WAN) access device
260
.
The data is usually transmitted between the remote nodes and the NAS
154
over an asynchronous line (e.g. telephone line). Therefore, the NAS typically needs to frame the packets of data received from the remote node into data-link control protocol frames (e.g. High-level Data Link Control (HDLC)) in order to transmit the packets over the synchronous line
156
. (References herein to the data-link layer and the network layer are references to the Open Systems Interconnection (OSI) model developed by the International Standardization Organization).
As shown in
FIG. 1
, the synchronous line
156
between the NAS
154
and the WAN access device
160
is usually a point-to-point connection. That is, a connection with out any intervening nodes or switches.
As a result of the point-to-point nature of the connection, the address field of the data-link frame is not used because the frame has only one possible destination once it is transmitted. Usually, the address field of a data-link frame is used to indicate the physical destination of the frame so that any switches between a frame's source and destination will be able to read the frame's address field and know where to forward the frame. However, when it is known that there will be no intervening switches between the frame's source and destination (e.g. point-to-point connection) no address is necessary. In fact, the RFC
1662
states that for point-to-point connections, the address field of the HDLC frame should contain 0×FF.
When the NAS
154
receives responses from the WAN access device
160
, the NAS
154
usually performs a routing function to determine where the response is to be forwarded. That is, the NAS
154
will strip the response packet from its data-link frame and read the packet's address information, which is usually provided at the network layer (e.g. the Internet Protocol (IP) address). The routing procedures, however, slow down the transmission of the responses and usually only work if the node to which a response is being delivered, has been given a network address.
In the case of a user dialing into a NAS
154
from a unit that does not have a network address (e.g. a terminal), the NAS will usually assign a network address to each packet received from that unit. The assigned network addresses will in turn will be used by the NAS to forward responses using the routing procedures described above. Assigning network addresses, however, has the disadvantage of preallocating several network addresses, when in fact, some of these addresses may not be used.
Therefore, what is needed is a way for an access server to forward data-link frames received from a point-to-point connection without having to perform routing operations.
SUMMARY OF THE INVENTION
The present invention provides a method and apparatus for transmitting packets. One embodiment of the present invention includes an access server receiving a packet from a first unit through a first interface of a plurality of interfaces. The access server maps the packet into a first frame to transmit the packet over a predetermined point-to-point connection. An interface number representing the first interface is stored in an address field of the first frame. The first frame is then forwarded via the point-to-point connection to a second unit.
The second unit maps a response to the packet into a second frame and stores the interface number representing the first interface in an address field of the second frame. The second unit then forwards the second frame to the access server.
Upon receipt, the access server reads the address field of the second frame to determine through which interface the response is to be forwarded.
REFERENCES:
patent: 5416771 (1995-05-01), Iwata
Blakely , Sokoloff, Taylor & Zafman LLP
Chin Wellington
Nortel Networks Limited
Yao Kwang B.
LandOfFree
Method and apparatus for mapping asynchronous ports to HDLC... 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 mapping asynchronous ports to HDLC..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for mapping asynchronous ports to HDLC... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2496701