Balancing loads on telecommunications links

• US 6,731,739 B2
• Filed: 12/11/2000
• Issued: 05/04/2004
• Est. Priority Date: 12/11/2000
• Status: Expired due to Term

• Current Assignee: Unwired Planet LLC (JP Morgan Chase & Co)
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Riihinen, Wesa, Söderberg, Johan
• Primary Examiner(s): Deane, Jr., William J.
• Application Number:
  US09/732,883
• Publication Number:
  US 20020150229A1
• Time in Patent Office: 1,240 Days
• Field of Search: 379/221.01, 379/221.15
• US Class Current: 379/221.01
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04W 28/082 among bearers or channels
• CPC Class Codes: H04W 8/04 Registration at HLR or HSS [Home Subscriber Server]

Abstract

A telecommunications network (110) has a unit (100) and technique to alleviate congestion on a congested link as can occur, e.g., when a new connection seeks admission to the network. The congestion avoidance technique/unit ascertains, from a set of candidate connections (C₀, C₁, C₂, C₃), a best candidate to route in order to avoid the congestion. Further, the congestion avoidance technique/unit determines a best candidate path to which the best candidate connection is to be routed. The best candidate connection and the best candidate path are determined so that, after routing the best candidate connection to the best candidate path, a load of the weakest link in the network is as low as possible.

Claims

1. A method of operating a telecommunications network to alleviate congestion on a congested link, the telecommunications network having plural nodes including a common node, a plurality of connections being carried by the telecommunications network, each of the plurality of connections being carried over one of plural paths of the network, each path including at least one link, each link connecting two of the plural nodes, each of the plural paths being connected to the common node by one of the plural links, wherein the method comprises:

determining, for routing purposes, a best candidate connection from a set of candidate connections;

determining a best candidate path to carry the best candidate connection;

the best candidate connection and the best candidate path being determined so that, after routing the best candidate connection to the best candidate path, (congestion on the congested link is alleviated and) a load of the weakest link in the network is as low as possible;

developing, for each candidate connection, a network topology connection subtraction load scenario, each connection subtraction load scenario having a load for the corresponding candidate connection subtracted therefrom;

determining a weakest link of each connection subtraction load scenario, whereby for each candidate connection there is a weakest link of its connection subtraction load scenario.

2. The method of claim 1, further comprising forming the set of candidate connections with a plurality of connections which use a congested link.

3. The method of claim 1, further comprising:

(a) for a particular candidate connection, developing a set of network topology path load scenarios, each path load scenario being developed by augmenting the connection subtraction load scenario for the particular candidate connection with a unique path assignment for the particular candidate connection;

(b) determining, for each network topology path load scenario of the set, a weakest link comprising the unique path assignment;

(c) determining, from the entire set of network topology path load scenarios, a strongest one of weakest links determined in step (b) for the particular candidate connection.

4. The method of claim 3, further comprising:

(d) determining a weakest link for the for the network topology path load scenario having the strongest one of the weakest links as determined in step (c); and

(e) for the particular candidate connection, using the weakest link determined at step (d) as the weakest link of its connection subtraction load scenario if the strength of the weakest link determined at step (d) exceeds the strength of the weakest link of its connection subtraction load scenario as previously determined.

5. The method of claim 4, wherein the candidate connection having a strongest weakest link of its connection subtraction load scenario is chosen as the best candidate connection, and wherein for the best candidate connection the unique path assignment determined as having the strongest one of the weakest links is used as the best candidate path.

6. The method of claim 1, wherein the plural nodes include radio network controller (RNC) nodes of a radio access network (RAN).

7. The method of claim 6, wherein the common node is a core network node.

8. The method of claim 1, wherein the steps of claim 1 are performed at the common node.

9. A telecommunications network which carries a plurality of telecommunication connections, the network comprising:

plural nodes including a common node;

plural paths, each of the plurality of connections being carried over one of plural paths, each path including at least one link, each link connecting two of the plural nodes, each of the plural paths being connected to the common node by one of the plural links;

a connection handler which alleviates congestion on a congested link by performing the operations of;

determining, for routing purposes, a best candidate connection from the plurality of connections;

determining a best candidate path to carry the best candidate connection, the best candidate connection and the best candidate path being determined so that, after routing the best candidate connection to the best candidate path, a load of the weakest link in the network is as low as possible;

developing, for each candidate connection, a network topology connection subtraction load scenario, each connection subtraction load scenario having a load for the corresponding candidate connection subtracted therefrom;

determining a weakest link of each connection subtraction load scenario, whereby for each candidate connection there is a weakest link of its connection subtraction load scenario.

10. The network of claim 9, wherein the connection handler performs the further operations of:

(a) for a particular candidate connection, developing a set of network topology path load scenarios, each path load scenario being developed by augmenting the connection subtraction load scenario for the particular candidate connection with a unique path assignment for the particular candidate connection;

(b) determining, for each network topology path load scenario of the set, a weakest link comprising the unique path assignment;

(c) determining, from the entire set of network topology path load scenarios, a strongest one of weakest links determined in step (b) for the particular candidate connection.

11. The network of claim 10, wherein the connection handler performs the further operations of:

(d) determining a weakest link for the for the network topology path load scenario having the strongest one of the weakest links as determined in step (c); and

(e) for the particular candidate connection, using the weakest link determined at step (d) as the weakest link of its connection subtraction load scenario if the strength of the weakest link determined at step (d) exceeds the strength of the weakest link of its connection subtraction load scenario as previously determined.

12. The network of claim 11, wherein the connection handler performs the further operations of:

choosing as the best candidate connection the candidate connection having a strongest weakest link of its connection subtraction load scenario; and

for the best candidate connection, using the unique path assignment determined as having the strongest one of the weakest links as the best candidate path.

13. The network of claim 9, wherein the plural nodes include radio network controller (RNC) nodes of a radio access network (RAN).

14. The network of claim 13, wherein the common node is a core network node.

15. The network of claim 9, wherein the connection handler is situated at the common node.

16. The network of claim 15, wherein the common node is a core network node.