Predictive bandwidth allocation method and apparatus

• US 6,208,640 B1
• Filed: 02/27/1998
• Issued: 03/27/2001
• Est. Priority Date: 02/27/1998
• Status: Expired due to Fees

• Current Assignee: Seiko Epson Corporation (Seiko Group), Turnkey Solutions Corp.
• Original Assignee: Seiko Epson Corporation (Seiko Group), Turnkey Solutions Corp.
• Inventors: Spell, David, Ellis, Dale, Mikami, Yukihiro, Rai, Ajay, Roland, Dave
• Primary Examiner(s): Pham, Chi H.
• Assistant Examiner(s): Pham, Brenda H.
• Application Number:
  US09/031,993
• Time in Patent Office: 1,124 Days
• Field of Search: 370/359, 370/352, 370/354, 370/460, 370/463, 370/468, 370/465, 370/522, 370/358, 370/493, 370/389
• US Class Current: 370/358
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04Q 11/0457 Connection protocols
• CPC Class Codes: H04Q 2213/1305 Software aspects
• CPC Class Codes: H04Q 2213/13054 Expert system
• CPC Class Codes: H04Q 2213/13103 Memory
• CPC Class Codes: H04Q 2213/13106 Microprocessor, CPU
• CPC Class Codes: H04Q 2213/13141 Hunting for free outlet, circuit or channel
• CPC Class Codes: H04Q 2213/13176 Common channel signaling, CCS7
• CPC Class Codes: H04Q 2213/13204 Protocols
• CPC Class Codes: H04Q 2213/13209 ISDN
• CPC Class Codes: H04Q 2213/1327 Release and resetting of connection
• CPC Class Codes: H04Q 2213/13298 Local loop systems, access network
• CPC Class Codes: H04Q 2213/13332 Broadband, CATV, dynamic bandwidth allocation
• CPC Class Codes: H04Q 2213/13335 Simulation, emulation
• CPC Class Codes: H04Q 2213/13343 Neural networks
• CPC Class Codes: H04Q 2213/13389 LAN, internet

Abstract

Allocation of telecommunication systems bandwidth is provided preferably in a predictive fashion. Packets are identified with particular data streams and characteristics of the data streams are used to predict probable future bandwidth requirements. Such predictions are used to allocate high-bandwidth channels, such as ISDN B channels and to close or switch channels as in accordance with predicted needs. Preferably the system is self-learning and can modify a rules base for making allocation decisions e.g. based on actual use statistics.

Claims

1. A computer-implemented process, in a telecommunications system for switching at least some data in a telecommunications stream, comprising:

identifying a first characteristic of said stream, wherein said characteristic is at least partially predictive of likely data size of said stream;

deciding, based on at least said first characteristic, whether to switch at least some of said data in said stream so as to provide different data transmission properties, wherein said deciding is performed using a first stored rules base;

automatically evaluating the effectiveness of said step of deciding.

2. A process, as claimed in claim 1, further comprising modifying said rules base, based on said step of evaluating.

3. A process, as claimed in claim 2, wherein said step of modifying said rules base is performed automatically, to provide a self-learning computer implemented telecommunications switching process.

4. A process, as claimed in claim 1, wherein said step of deciding is implemented using a pegboard process.

5. A process, as claimed in claim 1, wherein said step of evaluating comprises evaluating both cost and level of service.

6. A method, as claimed in claim 1, wherein said different data transmission property is selected from the group consisting of different bandwidth properties and different effective data transmission speed properties.