Network connectivity control by artificial intelligence

• US 4,999,833 A
• Filed: 12/20/1988
• Issued: 03/12/1991
• Est. Priority Date: 05/06/1985
• Status: Expired due to Fees

• Current Assignee: ITT Corporation (ITT, Inc.)
• Original Assignee: ITT Corporation (ITT, Inc.)
• Inventors: Lee, William C.
• Primary Examiner(s): Safourek, Benedict V.
• Assistant Examiner(s): Hsu, Alpus H.
• Application Number:
  US07/287,742
• Time in Patent Office: 812 Days
• Field of Search: 370/60, 370/94.1, 370/58.1, 370/58.2, 370/58.3, 370/60.1, 370/94.2, 370/94.3, 455/62, 364/513, 340/825.02, 340/825.06
• US Class Current: 370/312
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: H04L 45/08 Learning-based routing, e.g. using neural networks or artificial intelligence
• CPC Class Codes: H04W 4/12 Messaging; Mailboxes; Announcements
• CPC Class Codes: H04W 40/02 Communication route or path selection, e.g. power-based or shortest path routing
• CPC Class Codes: H04W 76/10 Connection setup
• CPC Class Codes: H04W 88/02 Terminal devices
• CPC Class Codes: H04W 88/04 adapted for relaying to or from another terminal or user
• CPC Class Codes: H04W 92/18 between terminal devices
• CPC Class Codes: Y10S 706/909 Communication

Abstract

A communications system utilizes artificial intelligence to select connectivity paths among various locations in a communications network. An embodiment shown is that of a packet radio network, wherein an artificial intelligence module, located at one or more of the radio sites in the network, applies a set of heuristic rules to a knowledge base obtained from network experience to select connectivity paths through the network. The artificial intelligence module comprises an inference engine, a memory for storing network data obtained from a radio receiver and transmitting it to the inference engine, a memory connected to the inference engine which stores a set of heuristic rules for the artificial intelligence system, and a knowledge base memory which stores network information upon which the inference engine draws. The knowledge base memory is also capable of feeding back network information to the rule base memory, which can thus update its rules. Also shown is an embodiment of a multimedia communications network.

Claims

1. In a communications system of the type including a plurality of individual radio frequency transceivers wherein a communications link can be set up between a first one of said transceivers to any other one of said transceivers via various paths which paths may include other ones of said transceivers which other ones act as relays in said communications link, each of said transceivers capable of receiving and transmitting messages, the improvement in combination therewith comprising:

each of said plurality of transceivers having message transmitting and receiving means for transmitting and receiving messages between transceivers,

selected ones of said plurality of transceivers each having an artificial intelligence apparatus coupled to its message receiving means for receiving messages which are transmitted from a transmitting transceiver to the selected transceiver or broadcasted within range of the selected transceiver for ultimate transmission to a receiving transceiver through a relaying transceiver which may include the selected transceiver itself,

wherein said artificial intelligence apparatus coupled to selected ones of said transceivers in said system includes;

a microprocessor for processing message data indicative of selected attributes of transmitted messages between transceivers, as received by the selected transceiver, and selected attributes of the communications links in which said messages between transceivers are transmitted,

a data base memory for storing therein said message data indicative of transmitted messages and communications links between said transceivers,

a rule base memory for storing rules for establishing minimum criteria for said selected attributes of transmitted messages and communications links in order for particular ones of the communications links to be deemed as desired communications links by said artificial intelligence apparatus for radio transmissions between transceivers,

an inference computing means connected to said data base memory and to said rule base memory and operative to scan the message data stored in said data base memory and to match the message data with the rules for establishing minimum criteria in said rule base memory in order to generate output conclusion data indicative of desired communications links,

a knowledge base memory coupled to said inference computing means for storing said output conclusion data indicative of desired communications links, whereby the selected transceiver is enabled to relay transmitted messages on the desired communications links based on the conclusion data stored in its knowledge base memory.

2. The communication system according to claim 1, wherein said rule base memory is a programmable read only memory (PROM).

3. The communication system according to claim 1, wherein said knowledge base memory comprises:

a first memory section for storing all possible communications links,

a second memory section for storing only one optimum communication link from a particular one transceiver to another transceiver in said system.

4. The communication system according to claim 1, further including an induction computing means connected to said knowledge base memory and to said rule base memory and operative to form new rules for storage in said rule base memory according to the data stored in said knowledge base memory.

5. The communication system according to claim 1, further including means coupling said knowledge base memory to said rule base memory wherein said data stored in said knowledge base memory can modify said stored rules.

6. The communication system according to claim 1, further including an expert port coupled to said rule base memory and operative to provide rules to said rule base memory, said expert port also coupled to said induction computing means for inducing rules from said expert port and to apply said induced rules to said rule base memory.

7. The communication system according to claim 1, wherein said transmitting and/or receiving transceivers are grouped in sub-groups of said transmitting and/or receiving means, at least two of said sub-group member means using different media, and further comprising:

an additional artificial intelligence module associated with said sub-group for selecting an optimum transmission path among best paths of each of said sub-groups.

8. The communication system according to claim 1, wherein said knowledge base and said data base memory consists of a single memory structure.

9. The communication system according to claim 1, wherein said inference computing means includes a microprocessor.

13. The method according to claim 7, wherein said matching step is based on traffic conditions and/or connection successfulness.

10. A method of communicating among a plurality of transceivers wherein a communication link can be set up between transceivers via various communications paths, comprising the steps of:

providing selected ones of said plurality of transceivers each with an artificial intelligence apparatus for monitoring messages which are transmitted from a transmitting transceiver to the selected transceiver or broadcasted within range of the selected transceiver for ultimate transmission to a receiving transceiver through a relaying transceiver which may include the selected transceiver itself,

processing message data indicative of selected attributes of transmitted messages between transceivers, as received by the selected transceiver, and selected attributes of the communications links in which said messages between transceivers are transmitted,

establishing a data base in memory of said message data processed in each respective one of said selected transceivers,

scanning said message data in said data base and matching the message data with criteria from a rule base established in memory for said selected attributes of transmitted messages and communications links in order for particular ones of the communications links to be deemed as desired communications links so as to generate transmission routing sequences for said transceivers,

storing said generated sequences in a knowledge base memory in the respective selected transceiver, and

establishing connectivity paths among said plurality of transceivers using said generated routing sequences stored in the knowledge base memories of the selected transceivers.

11. The method according to claim 10, further comprising the step of:

updating said routing sequences in said knowledge base memory in real time.

12. The method according to claim 10, wherein said step of establishing connectivity paths results in distributed routing of communications among said plurality of transceivers.

14. An artificial intelligence apparatus module for use with radio networks, of the type including a plurality of individual radio frequency transceivers wherein a communications link can be set up between a first one of said transceivers to any other one of said transceivers via various paths which paths may include other ones of said transceivers which other ones act as relays in said communications link, each of said transceivers having a radio transmitter and receiver, and at least one of said transceivers being a selected transceiver for generating routing sequences to be used among said plurality of transceivers, comprising:

input means for coupling said artificial intelligence apparatus module to the receiver of said selected transceiver for receiving messages which are transmitted among said plurality of transceivers;

a microprocessor for processing message data indicative of selected attributes of the transmitted messages as received by the selected transceiver, and selected attributes of the communications links in which said messages between transceivers are transmitted;

a data base memory for receiving said message data from said microprocessor;

a rule base memory capable of being programmed with a set of rules for establishing minimum criteria for said selected attributes of transmitted messages and communications links in order for particular ones of the communications links to be deemed as desired communications links by said artificial intelligence apparatus for radio transmissions between transceivers;

an inference engine connected to said data base memory and to said rule base memory and functioning to scan the message data in said data base memory and to match this message data with the rules for establishing minimum criteria in the rule base memory so as to generate information indicative of desired communications links;

a knowledge base memory connected to said inference engine which receives said information generated by said inference engine, whereby the artificial intelligence module can enable the selected transceiver to relay transmitted messages on the desired communications links based on the information stored in the knowledge base memory.

15. The artificial intelligence module of claim 14, further comprising a communication path from said knowledge base memory to said rule base memory.

16. The artificial intelligence module of claim 14, further comprising an induction engine, connected to said knowledge base memory, to said rule base memory, and to an expert input port, functioning to formulate rules for said rule base memory.

17. The artificial intelligence module of claim 14, wherein said rule base memory is a programmable read only memory (PROM).