Multiplex communications – Communication techniques for information carried in plural... – Adaptive
Reexamination Certificate
1998-04-10
2001-09-18
Olms, Douglas (Department: 2732)
Multiplex communications
Communication techniques for information carried in plural...
Adaptive
C370S395430, C370S471000
Reexamination Certificate
active
06292495
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of digital communications, and more particularly to managing virtual circuits that pass through a frame relay network.
BACKGROUND OF THE INVENTION
Frame relay is a broadband packet switching technology that is often used to implement wide area networks (WANs). Many local and inter-exchange carriers offer frame relay service with access rates ranging from fractional T1 (e.g., n×64Kb/s) to multimegabit (e.g., 44.736 Mb/s T3). Pricing is usually determined by the access line rate, the number of permanent virtual circuits (PVCs) managed by the network and the bandwidth consumed by each PVC. Frame relay is defined by American National Standards Institute (ANSI) specification T1.606, published in 1990 and entitled “Telecommunications—Integrated Services Digital Network (ISDN)—Architectural Framework and Service Description for Frame-Relay Bearer Service” (hereinafter, “the frame relay specification”).
FIG. 1
depicts a prior art network configuration
10
in which a frame relay network
12
is used to interconnect three local area networks (LANs)
16
a
,
16
b
,
16
c
. Each of the LANs
16
a
,
16
b
,
16
c
is used to interconnect a respective set of LAN stations
18
a
,
18
b
,
18
c
(e.g., personal computers, workstations or larger computers) and may employ any of a number of different data link layer protocols, including Ethernet, Fiber Distributed Data Interface, Token Ring, so forth. Each of the LANs
16
a
,
16
b
,
16
c
is coupled to the frame relay network
12
via a respective router
14
a
,
14
b
,
14
c
that typically includes a frame relay packet assembly/disassembly (PAD) function to assemble data received from various LAN stations
18
a
,
18
b
,
18
c
into one or more frame relay packets and to disassemble frame relay packets received from the frame relay network
12
into a format according to the LAN protocol. Although each router
14
a
,
14
b
,
14
c
is depicted as being coupled only between the frame relay network
12
and a respective LAN
16
a
,
16
b
,
16
c
, a router will typically be used interconnect a LAN to several different networks.
To support LAN-to-LAN communications across the frame relay network
12
, respective addresses called data link connection identifiers (DLCIs) are usually assigned to each of the LAN stations
18
a
,
18
b
,
18
c
. One DLCI is placed in the address field of each packet carried by the frame relay network to indicate the packet's destination. Because the DLCI effectively steers a packet through the frame relay network
12
to the indicated destination, the DLCI is said to establish a virtual circuit through the frame relay network
12
. Permanent virtual circuits (PVCs) are virtual circuits in which the connections between the routers
14
a
,
14
b
,
14
c
and the frame relay network
12
are configured by the provider of the frame relay network
12
and remain established thereafter. Switched virtual circuits, by contrast, require special setup and termination messages to be issued to the frame relay network
12
to establish and terminate a connection.
Still referring to
FIG. 1
, the connection between a router
14
a
,
14
b
,
14
c
and the frame relay network
12
is a demarcation point referred to as a User-Network Interface (UNI), with equipment on the user side of the UNI (e.g., the router, the LAN and the LAN stations) usually being customer premise equipment (CPE) and equipment on the network side of the UNI usually being WAN provider equipment. The router
14
a
,
14
b
,
14
c
is commonly referred to as a frame relay access device (FRAD) because it provides customer premise equipment access to the frame relay network
12
.
In 1990, a Consortium of companies including Cisco Systems, Inc., Digital Equipment Corporation, Northern Telecom, Inc. and StrataCom, Inc. developed a link monitoring interface over the UNI called the Local Management Interface (LMI) to allow customer premise equipment to monitor the status of PVCs in a frame relay network. The LMI protocol and its suite of messages are defined by an extension to the frame relay specification published by the Consortium on Sep. 18, 1990 and entitled “Frame Relay Specification with Extensions Based on Proposed T1S1 Standards, Document 001-208966”. A later published ANSI standard defines a modified version of LMI (“Integrated Services Digital Network(ISDN)—Signaling Specification for Frame Relay Bearer Service for Digital Subscriber Signaling System Number 1 (DSS1), ANSI T1.617 Annex D”, published in 1991). Fundamentally, the Consortium-specified LMI (hereinafter, Consortium LMI) and the ANSI T1.617 Annex D-specified LMI (hereinafter, Annex D LMI) are the same in that a FRAD issues status enquiry messages to the frame relay network
12
and the frame relay network
12
responds with status messages. Because LMI messages can become quite long and assume a one-to-one correspondence between DLCIs and PVCs, existing LMI implementations present obstacles to the transmission of voice and data over frame relay through a single UNI.
One characteristic of frame relay networks is that frame relay packets are permitted to vary in length from one packet to the next. This is in contrast to cell relay networks (e.g., FastPacket networks or Asynchronous Transfer Mode (ATM) networks) in which packets are fixed length cells. One advantage of permitting variable length packets is that, at least in larger packets, the ratio of overhead information (e.g., framing, addressing and error checking information) to payload is relatively small, meaning that a relatively small portion of network bandwidth is consumed by transmission of overhead information. By contrast, relatively short, fixed length cells (e.g., 24 or 53 octets) typically have a larger ratio of overhead to payload so that a larger portion of network bandwidth is consumed by transmission of overhead information. On the other hand, a significant disadvantage of permitting variable length packets to be transmitted on a frame relay network is that variable transmission delays are incurred as packets are queued behind one another in the network's various ingress and egress queues. As a result, data that requires a relatively fixed interval to be maintained between successive packets (e.g., packetized voice, video and other constant bit rate data) becomes distorted by the variable delays in the transmission path. This distortion is called jitter and is one reason that frame relay networks traditionally have not been used to carry voice and other constant bit rate data.
FIG. 2
illustrates a prior art network configuration
21
that allows packetized voice to be transmitted over a frame relay network
12
with significantly reduced jitter. Devices called fragmenters
22
a
,
22
b
,
22
c
receive variable length frame relay packets from respective routers
14
a
,
14
b
,
14
c
and decompose packets that are longer than a predetermined number of octets into two or more smaller packets called fragments. Each fragmenter
22
a
,
22
b
,
22
c
also receives voice inputs and packetizes them into fixed-length packets referred to herein as voice frames. The voice frames and the fragments adhere to the frame relay packet format and are carried by the frame relay network
12
to a destination (e.g., a LAN station
18
a
,
18
b
,
18
c
on a destination network
16
a
,
16
b
,
16
c
) indicated by their respective address fields. Because the voice frames and the fragments are transmitted to the frame relay network
12
on the same access line, long data packets would ordinarily introduce significant jitter to voice frames queued behind them. However, by decomposing long packets into relatively short fragments and then transmitting relatively short fragments across the frame relay network
12
, voice frame jitter is significantly reduced. Also, different PVCs can be allocated to carry the fragments and the voice frames through the frame relay network
12
and the PVC used to carry voice frames can often be tailored for voice support. For example, it
Simon Robert
Von Hammerstein Charles G.
Blakely , Sokoloff, Taylor & Zafman LLP
Cisco Technology Inc.
Olms Douglas
Vanderpuye Ken
LandOfFree
Segmented permanent virtual circuits does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Segmented permanent virtual circuits, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Segmented permanent virtual circuits will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2438556