Comparative path replacement

Details Comparative path replacement Comparative path replacement

1997-04-18

2001-02-27

Kizou, Hassan (Department: 2632)

Multiplex communications

Data flow congestion prevention or control

Flow control of data transmission through a network

C370S248000, C370S351000

Reexamination Certificate

active

06195336

ABSTRACT:

TECHNICAL FIELD OF THE INVENTION
The present invention relates to improving the efficiency of paths used in ISDNs (Integrated Services Digital Networks).
BACKGROUND OF THE INVENTION
In a public or private ISDN network where features such as call transfer and call redirection (“call forwarding”) may occur, highly inefficient connections may result. For example, suppose a person in San Francisco calls a person in New York who subsequently transfers the call to a person in Los Angeles. The resulting connection will traverse the continent twice and be more than 6,000 miles long, even though the parties actually in the call are only a few hundred miles apart. In such an environment, it may be advantageous for the network to automatically improve (e.g., shorten) the connection between the parties by selecting a new path through the network and by switching to the new path through the network even as the parties continue talking.
This process is known as path replacement. The protocol and procedure for path replacement have been standardized by ISDN standards bodies. However, in current networks, when a path is replaced, e.g., because of a redirection of a call such as a transfer, there is no guarantee that the replacement path is better than the original one. The path may in fact be worse (e.g., more costly).
What is needed is an ISDN path replacement procedure that is driven by comparisons of path quality.
SUMMARY OF THE INVENTION
A networking procedure collects network statistics pertaining to two paths for a network call and determines the better path through the network for the particular call. The procedure collects network data in a first data set containing first data fields. The first data set relates to the path of the current connection. A second data set is collected about a potential replacement path. The first mad second data sets are compared, and the better of the two paths determined.
The procedure provides a means for determining which path through the network is more desirable. Desirability is determined using a predefined set of prioritized criteria customized by each client network. For example, the administrator can indicate that the delay is more important than the number of gateway PTNs (public or private telephone networks), and that trunk priority level should not be considered in the analysis.
Steps in the procedure include: 1) requesting path replacement, which initiates the procedure; 2) collecting statistical routing data link by link for a current path and a proposed replacement path; 3) comparing paths based on the customized criteria; and 4) retaining the better path and releasing the other path.
1. Requesting path replacement
As discussed supra, standardized protocols exist for requesting path replacement. Triggers for path replacement are defined based on preselected criteria. For example, a change in relative costs of different links can arise at certain times of day; accordingly, particular times may trigger a path replacement request. Calls created by means of supplementary service, as for example by “transfer by join” procedures, can trigger a path replacement request, since such a link will pass through the transferring user's PTN, which might not be the least-cost path.
Termination of an add-on conference can also trigger a path replacement search, because the call will always go through the private telecommunications network/exchange (PTN/X) of the conference bridge. Similarly, a call which has been forwarded by Call Forwarding using Forward Switching, or which has undergone Line Hunting, will be routed through the forwarding PTN/X. Any call routed through a particular PTN/X may not follow the most desirable path.
Many other circumstances can lead to a call being routed in an undesirable path.
These circumstances can be selected to trigger a request for a replacement path. In some cases, a gateway PTN can “invite” a PTN to request a new path.
2. Collecting statistical routing data link by link on a per-call basis
A request typically includes a unique “correlation” number to allow any new connection found to be associated with the old connection. An endpoint sends a message requesting a path replacement, along with a data set to be filled. The collected data about the first path is compared with collected data about a candidate replacement path. The more desirable path is retained and the less desirable path is released.
Each data set comprises evaluations of the same standard criteria. The first data set is transmitted from the requesting PTN (the PTN making the path replacement request) in a facility message to the cooperating PTN. Each intermediate tandem or gateway PTN recognizes the facility message, updates the first data set with its information, and passes the message on toward the cooperating PTN. When the cooperating PTN receives the message, it also updates the first data set. The cooperating and requesting PTNs may additionally function as gateways. When the data collection process cannot be internetworked, the gateway acts on behalf of the user PTN as a requesting PTN and/or cooperating PTN for the collection of the data.
After all the route information has been gathered (e.g., the first data set has arrived at the cooperating PTN and been updated), the first data set is sent from the cooperating PTN back to the requesting PTN within a facility acknowledgment message. In the return direction, the first data set is not updated. In a preferred embodiment, the first data set in the return direction is sent along the original path. In alternative embodiments, the first data set in the return direction could be sent along the candidate replacement path.
The second data set is transmitted along the candidate replacement path to the requesting PTN. Each intermediate PTN updates the data set with its information. The first and second data sets each have unique identifiers.
3. Path analysis based on administrable parameters
In a first embodiment, criteria are compared pairwise. For each criterion selected by the user, either the first data set or the second data set wins (excepting ties). For each data set, the number of wins is counted. The most desirable path (the path that “wins” on the most criteria) is selected. The other path is released.
In a preferred embodiment, each criterion is evaluated at each PTN on a scale of zero to fifteen, with zero being more desirable. In one alternative, the path with the smaller sum wins. In another alternative, the path with the smaller maximum individual criterion score wins.
In a second embodiment, the user selects numerical weights for the criteria. Each criterion is multiplied by its respective numerical weight and these products are summed for each path. The path corresponding to the data set with the smaller sum is selected, and the other path is released.


REFERENCES:
patent: 4472798 (1984-09-01), Hafer
patent: 5263025 (1993-11-01), Torii et al.
patent: 5311572 (1994-05-01), Friedes et al.
patent: 5517550 (1996-05-01), Kobayashi
patent: 5661779 (1997-08-01), Lee
patent: 5751706 (1998-05-01), Land et al.
patent: 5884179 (1996-09-01), Patel

Affiliated with

Also associated with

Rating

Comparative path replacement does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Comparative path replacement, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Comparative path replacement will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2573190