Distributed switching architecture

• US 4,792,948 A
• Filed: 08/08/1985
• Issued: 12/20/1988
• Est. Priority Date: 08/08/1985
• Status: Expired due to Term

• Current Assignee: Data General Corp. (Dell Technologies Inc.)
• Original Assignee: Data General Corp. (Dell Technologies Inc.)
• Inventors: Lee, Fred S., Hangen, Joaquin J., Whelpley, William A., Swithers, Robert A.
• Primary Examiner(s): Olms, Douglas W.
• Application Number:
  US06/763,701
• Time in Patent Office: 1,230 Days
• Field of Search: 370/95, 370/58, 370/85, 370/104, 370/66
• US Class Current: 370/376
• CPC Class Codes: G06F 15/17375 One dimensional, e.g. linear array, ring
• CPC Class Codes: H04W 74/00 Wireless channel access
• CPC Class Codes: H04W 84/00 Network topologies
• CPC Class Codes: H04W 84/14 WLL [Wireless Local Loop]; RLL [Radio Local Loop]

Abstract

A communication network made up of distributed switching stations controlled by a network control center. The distributed switches communicate with each other over long haul links and locally with users over short haul communication channels to user switches. Each user switch interfaces a distributed switch to a plurality of individual users. Communication is carried out in a Demand Assigned Multiple Access fashion with the distributed switches communicating with differential inbound and outbound data frame lengths. The outbound frames being shorter and in time division multiplexed (TDM) format. The inbound data frames from the user switches are longer and in time division multiple access (TDMA) format.

Claims

1. A distributed switching architecture for a communication network comprising:

a plurality of groups of user ports for message transmissions;

a plurality of user node means for receiving message transmissions from user ports, composing data frames from said message transmissions, transmitting the data frames in TDMA bursts over an associated communication channel, receiving continuous TDM signals including data intended for individual user ports, and distributing the data to the individual user ports, said user node means being connected to all of the user ports in a group;

a plurality of distributed switch means, each connected to at least one user node means by respective associated communication channels and at least one other distributed switch means by communication links, for receiving data frames in TDMA bursts from its associated user node means and receiving data frames from associated distributed switch means, temporarily storing the data frames in a storage device located within said distributed switch means, and transmitting TDM data frames to associated user node means in accordance with a time plan and transmitting data frames to associated distributed switch means; and

a means for network control connected to all distributed switch means for generation and distribution of said time plan.

2. A distributed switching architecture as in claim 1 wherein each of said user node means is located proximately to a group of user ports and each of said distributed switch means is located centrally to one or more socket assignment means.

3. A distributed switching architecture as in claim 1 wherein the data frames transmitted by the user node are inbound frames and the data frames transmitted to the user node are outbound data frames,

said inbound data frames and outbound data frames are of different lengths wherein said outbound data frames are shorter than said inbound data frames.

4. A distributed switching architecture as in claim 3 wherein said user node means comprises:

frame controller means connected to said communication channel for time socket assignment of said data frames; and

processor means connected to said frame controller and all of said user ports in a group by a high speed transmit/receive processor bus for controlling said user ports and said frame controller; and

a channel address bus means connected to said frame controller and said user ports for enabling said ports.

5. A distributed switching architecture as in claim 4 wherein said frame controller means comprises:

sequencer means connected to said processor bus for storing a burst time plan received from said processor means and assigning time sockets by sequentially outputting plan port addresses contained in said burst time plan to said channel address bus;

means for scrambling outbound data frames connected to said processor bus;

means for generating preambles and adding said preambles to scrambled inbound frames responsive to said sequencer means and connected to said communication channel;

means connected to said communication channel for detecting preambles of said outbound data frames and providing a preamble detect signal to said sequencer means;

means connected to said processor bus responsive to said communication channel and said preamble detect signal for descrambling said outbound data frames.

6. A distributed switching architecture as in claim 5 wherein each user port comprises:

buffer means coupling said processor bus with a user termination device for data compression and expansion;

register means connected to said processor means by said processor bus for storing an assigned port address;

means responsive to said register means and said channel address bus for comparing said assigned port address to said plan port address and connected to enable said buffer means based on a match.

7. A distributed switching architecture as in claim 6 wherein said user port further comprises means for squelching a failed signal connected to said processor bus means.

8. A distributed switching architecture as in claim 6 wherein said user port further comprises means for idle pattern suppression connected to said buffer means.

9. A distributed switching architecture as in claim 6 wherein said user termination device comprises data processing terminals.

10. A distributed switching architecture as in claim 6 wherein said user termination device comprises telephones.

11. A distributed switching architecture as in claim 6 wherein said user terminating device comprises means for receiving said message transmissions.

12. A distributed switching architecture as in claim 6 wherein said user termination device comprises means for transmitting said message transmission.

13. A distributed switching architecture as in claim 6 wherein said user termination device comprises means for transmitting and receiving said message transmissions.

14. A distributed switching architecture as in claim 3 wherein said distributed switch means comprises:

a plurality of frame controller means each assigned to a different communication channel for time socket assignment of outbound data frames and interfacing with said communication channel;

a processor means connected to all of the frame controllers by high speed transmit/receiver processor bus for controlling said frame controllers;

a time slot interchange means for temporary storage of message transmissions connected to said processor bus; and

a time shared address bus means for addressing said time slot interchange means, each address bus connected to an individual frame controller means and said time slot interchange means.

15. A distributed switching architecture as in claim 14 wherein said distributed switch means further comprises:

a means for interfacing with a long haul truck connected to said processor bus and an additional address bus means which is connected to said time slot interchange means.

16. A distributed switching architecture as in claim 14 wherein said frame controller comprises:

sequencer means connected to said processor bus for storing a burst time plan received from said processor means and assigning time sockets by sequentially outputting plan channel addresses contained in said burst time plan time slot interchange by said address bus means;

means for scrambling outbound data frames connected to said processor bus;

means for generating preambles and adding said preambles to scrambled outbound frames responsive to said sequencer means and connected to said communication channel;

means connected to said communication channel for detecting preambles of said inbound data frames and providing a preamble detect signal to said sequencer means;

means connected to said processor bus responsive to said communication channel and said preamble detect signal for descrambling said inbound data frames.

17. A distributed switching architecture as in claim 16 wherein said time slot interchange means further comprises:

means for receiving inbound data frames from all said plurality of frame controller means and storing said data frames in a predetermined location; and

means for outputting said data frames to said plurality of frame controllers in accordance with the burst time plan provided to the time slot interchange in the form of plan channel addresses over said address bus means.

18. A distributed switching architecture as in claim 17 wherein said distributed switch means further comprises means for time division multiplexed interfacing with a long haul trunk connected to said processor bus and an additional address bus means which is connected to said time slot interchange means;

and

wherein said means for receiving further comprises means for receiving and storing time division multiplexes data frames from said means for interfacing with a long haul trunk; and

means for receiving reformatting and storing time division multiple access data frames from said frame controller means.

19. A distributed switching architecture as in claim 3 further comprising means for extending transmission range by repeating message transmissions between user nodes and a distributed switch.

20. A point-to-multipoint communication system comprising:

a plurality of remote communication terminals;

a central switching terminal for communicating with said remote terminals over a plurality of communication channels including;

a plurality of means for interfacing said central switching terminal with said communication channels, receiving inbound data frames from said remote terminals implementing a stored burst time plan for outputting outbound data frames to said remote terminals in a demand assigned fashion;

processor means for controlling said means for interfacing and generating and providing said burst time plan to said means for interfacing;

time slot interchange means for temporary storage of said data frames wherein inbound frames are stored in a predetermined order based upon the channel they are received from and outbound data are output in a sequence prescribed by said burst time plan;

processor bus means for connecting said processor means to said means for interfacing and time slot interchange means to said means for interfacing;

a time shared address bus means, connecting the time slot interchange means to a means for interfacing for providing said time slot interchange means with plan addresses contained in said burst time plans.

21. A point-to-multipoint communication system as in claim 20 wherein said inbound data frames and said outbound data frames are of different lengths, said outbound data frames being shorter than said inbound data frames.

22. A method of point-to-multipoint communication between a plurality of remote terminals and a central station which receives inbound data frames and transmits outbound data frames made up of a plurality of time sockets with differing sources and destinations within said remote terminals comprising the steps;

for inbound communication;

each remote terminal successively transmitting time division multiple access inbound data frames to the central station;

the central station receives the inbound data frames and loads them into a time slot interchange for temporary storage;

for outbound communication;

resequencing the stored time sockets into outbound data frames in accordance to a burst time plan;

transmitting the burst time plan to the remote terminals;

transmitting the outbound data frames in time division multiplexed fashion to all associated remote terminals;

each remote terminal accessing the outbound data frames in accordance to the burst time plan and receiving only the data of those time sockets with a destination within said remote terminal.

23. A method of point-to-multipoint communications as in claim 22 wherein said inbound data frames and outbound data frames are of different lengths, said outbound data frames being shorter than said inbound data frames.