Multiplex communications – Pathfinding or routing – Combined circuit switching and packet switching
Reexamination Certificate
1999-09-24
2004-01-06
Ton, Dang (Department: 2666)
Multiplex communications
Pathfinding or routing
Combined circuit switching and packet switching
C370S352000, C370S353000, C370S447000, C370S462000
Reexamination Certificate
active
06674750
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to data communications, and more particularly, to a communications system that communicates both time division multiplexed data (synchronous data) and packet data (variable bit rate) on a shared bus.
2. Discussion of the Related Art
In recent years, there has been a notable migration of digital communications from time-division-multiplexed transmission systems to packet-based transmission systems. This migration has led to the development of communications equipment having the ability to communicate both types of digital transmissions.
In this regard, a communications device known as a “network access unit”(NAU) or “network access switch” (NAS) typically provides network switching services between multiple local and one or more network interfaces. Examples of local interfaces are RS232, DSX-1, EIA-530, etc. Examples of network interfaces are T1, T3, or SONET facilities. A network access unit provides transmission between multiple local communication networks, and a single network facility. The network access switch is a more general device, which provides communication between multiple local and network interfaces. The NAU typically resides at a home office, small office, branch office, or network edge. The NAS typically resides in the network cloud or at a large regional office fed by multiple branches.
To provide the most flexibility, it is preferable that the NAS/NAU support two types of data: synchronous TDM data and packet data. For example, the support of synchronous data provides the ability to make telephone (i.e., voice) calls, while the support of packet data provides access to public or private network packet services. However, the asynchronous nature of packet data at the logical level combined with the requirements of synchronous data causes design tradeoffs in both the complexity and cost of a NAS/NAU.
Network access switching requirements have progressed from providing TDM only type switching services, to packet only type switching services, to hybrid TDM/packet type switching services. The architecture of the NAS/NAU has evolved in order to support the more recent requirements for packet, and hybrid TDM/packet type services.
Early NAU/NAS architectures were based on a TDM bus architecture (See FIG.
1
A). The TDM bus provides for static switching of synchronous traffic between multiple interface modules
110
-
1
,
110
-N. The TDM bus architecture is best suited for offering TDM services. The TDM bus design offers lowest complexity, a minimum and constant transmission delay, guaranteed network bandwidth, and provides a robust ability to switch reconfigurations. The TDM bus allows modules to exchange data based on time slots, which repeat over a fixed frame time. A synchronous data stream is transmitted over the bus by transmitting the required number of bits in each frame time to match the transmission rate of the synchronous data stream. For example, a synchronous data stream of 64 kbps transmits 8 bits across the TDM bus for an 8 KHz bus frame. The number of 64 kbps data streams that can be transmitted across the bus is a function of the bus speed (throughput). Transmission of data across the bus is controlled by a time slot map, which is typically resides in each of the modules. The time slot map defines the times when a pair of modules exchange data across the bus. Essentially two atomic operations are required in the pair of TDM modules to begin exchanging 64 kbps or n×64 kbps data (where n is an integer value). While the configuring of this channel affects the transmission of data for this channel until configuration is complete, traffic transmitted in other channels are not affected by the reconfiguration. The time slot maps are typically structured so that a single 64 kbps or n×64 kbps channel can be added or removed in the network access switch without corrupting or disrupting the transmission of other 64 kbps data streams in the system. The ability to add or remove connections without disrupting existing connections in the NAS is usually a critical service requirement for voice and data networks. Finally, traffic delay across the bus is largely defined by the bus frame rate. Minimizing the switching delay is often an important requirement for delay sensitive TDM traffic such as voice. Frame rates of 8 Khz are typical in TDM based systems resulting in traffic delays of 125 &mgr;s to 250 &mgr;s. While the TDM bus is ideally suited for the switching of TDM traffic in the NAS, the classical TDM bus is not suited for the switching of packet traffic.
TDM based network access switches were originally designed for switching voice traffic, and later used to switch data traffic. TDM switches required static connections to be provisioned for the switching of data traffic. The major consideration for data traffic is that it is typically bursty in nature. A high volume of traffic will flow for a short period of time followed by idle periods. The TDM bus does not provide the ability for traffic to be dynamically switched from one module to another module over the bus in time. For a given configuration, the same two TDM modules continue to exchange information during the same time slot, until a reconfiguration occurs. For example, packet traffic from packet module
1
may be switched to packet module
2
in one instant, and then switched to packet module
3
in the next instant. Another consideration for supporting data traffic is that the actual traffic flow and connections are typically highly bursty as well. One machine may be required to exchange data with several machines, requiring multiple static connections in the TDM NAS between the one machine and its peers. Transmission of packet type applications through a TDM NAS typically results in an over provisioning and hence under utilization of NAS bus bandwidth. While information within the packet itself could be used to indicate which of the destination modules on the bus should receive the traffic, the standard TDM bus provides no effective means to use this information.
In order to address the shortcomings and limitations of the TDM switch for packet applications, a new type of NAS emerged to better manage packet data communications. The packet data switch (see FIG.
1
B), Frame Relay switch, and multi-protocol router type architectures, provide a more suitable means for switching of packet data traffic. These devices are based on a packet switching bus. The key attribute of the packet switching bus is that the data packet is prefixed with a switching or routing tag which includes an identifier or address which is used by the system to switch (or route) traffic to the intended destination bus module (and interface). Another key attribute is that the bus bandwidth is more dynamically shared between all the modules communicating over the bus. Modules
112
-
1
,
12
-n not requiring transmission across the bus during a particular time, allow the packet bus bandwidth to be dynamically allocated to modules requiring transmission. Some of the key elements in a packet switching architecture are dynamic allocation (or arbitration) of the bus bandwidth, formation and identification of packet addressing information, and identification of the beginning and end of a packet. While the packet bus is ideally suited for packet applications, it is not well suited for the switching of delay sensitive TDM traffic. First of all, since information is transmitted across the bus as packets, TDM data first must be accumulated into a packet before it can be transmitted across the bus. Packetization delay of TDM data is a function of the data rate and packet size. The larger the packet and lower the data rate, the larger the delay. A 64 kbps voice stream packetized into a 200 byte packet, incurs a 25 ms packetization delay, which is likely to exceed end-to-end network delay requirements. Further, packet traffic is bursty in nature. Consequently, switching delays across the bus may vary considerably based on traffic load. Not only
Abelson Ron
Paradyne Corporation
Thomas Kayden Horstemeyer & Risley
Ton Dang
LandOfFree
Apparatus and method for communicating time-division... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for communicating time-division..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for communicating time-division... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3225168