System and method for management of connection oriented networks

• US 6,078,946 A
• Filed: 09/10/1996
• Issued: 06/20/2000
• Est. Priority Date: 09/10/1996
• Status: Expired due to Fees

• Current Assignee: Verado Holdings, Inc.
• Original Assignee: FirstWorld Communications Inc. (Verado Holdings, Inc.)
• Inventors: Johnson, William A.
• Primary Examiner(s): Levy, Stuart S.
• Assistant Examiner(s): KUPSTAS, TOD A
• Application Number:
  US08/707,949
• Time in Patent Office: 1,379 Days
• Field of Search: 395/200.36, 395/200.68, 395/200.7, 395/200.71, 395/23, 395/22, 364/242.94, 364/284.4, 364/274.9, 370/238
• US Class Current: 709/200
• CPC Class Codes: H04L 41/0806 for initial configuration or provisioning, e.g. plug-and-play
• CPC Class Codes: H04L 41/5051 Service on demand, e.g. definition and deployment of services in real time
• CPC Class Codes: Y04S 40/00 Systems for electrical power generation, transmission, distribution or end-user application management characterised by the use of communication or information technologies, or communication or information technology specific aspects supporting them

Abstract

The system and method for provisioning resources in a network provides real time, parallel evaluation of the best path within the network using neural network principles. Elements of the network having a plurality of paths are assigned relative values according to a network user'"'"'s requirements. Attributes may include factors such as reliability, cost, speed, distance, expandability, etc., and may be static or dynamic. Selection of a best path from the plurality of paths comprises application of fuzzy logic, using a threshold function to identify a best relative path value by providing an input to the function which is a combination of the attribute values of the elements within each path. The input to the function is the sum of weighted attribute values, where each attribute value is multiplied by a weight which is a relative value determined in accordance with the network user'"'"'s priorities; the higher the priority, the greater the weight applied to that attribute. Computation of the threshold function is performed for each path and the resulting values are compared to a pre-determined threshold value to determine if the threshold has been crossed. If a single optimal path has not been identified during this step, the weights of the various attributes are adjusted in order of their priority; upward if no paths have crossed the threshold and downward if multiple paths have crossed the threshold. The process is continued with increasingly smaller incremental changes in the weights until a single combination of elements provides a path value which crosses the threshold, indicating the best path for meeting the network user'"'"'s criteria. The threshold function may be any algorithm which provides a threshold, including sigmoid, linear, exponential and quadratic functions.

Claims

1. A computer-based system for management of a connection oriented network according to a user'"'"'s pre-determined priorities, said network having a plurality of trails interconnecting a plurality of points, each trail of said plurality having a plurality of physical attributes (A, B, C, . . . ), the system comprising:

means for assigning a plurality of attribute values (A1, A2, A3, . . . ) (B1, B2, B3, . . . ) (C1, C2, C3 . . . ) for each physical attribute (A, B, C, . . . );

a memory means for receiving and storing a data model of each said trail, said data model including a set of attribute values {A1, B1, C1, . . . } for each said trail representative of said plurality of physical attributes;

data entry means for entering an attribute weight {W_A, W_B, W_C, . . . } for each physical attribute (A, B, C, . . . ) according to the user'"'"'s priorities, wherein a higher priority of the user'"'"'s priorities has a greater weight than a lower priority of the user'"'"'s priorities; and

a processing means for;

(a) multiplying each attribute value of said set of attribute values by the corresponding attribute weight to provide a set of weighted attribute values {A1*W_A, B1*W_B, C1*W_C, . . . } for each said trail;

(b) summing said set of weighted attribute values for each said trail to generate an initial input value IN_T for each said trail,

(c) performing a transfer function on said initial input value according to a first equation OUT_T =f(IN_T) to determine a threshold for each said trail,

(d) adjusting one of said attribute weights of the set of weighted attribute values {A1*W_A, B1*W_B, C1*W_C, . . . } for each said trail according to the user'"'"'s priorities to provide a set of adjusted weighted attribute values for each said trail,

(e) summing said set of adjusted weighted attribute values to provide an adjusted input value IN_T (adjusted) for each said trail,

(f) performing said transfer function on said adjusted input value for each said trail to generate an adjusted output value for each said trail where OUT_T (adjusted)=f(IN_T (adjusted)-IN_T),

(g) comparing said adjusted output value OUT_T (adjusted) for each said trail to said threshold for each said trail to identify a single adjusted output value that crosses its corresponding said threshold, and, if no said single adjusted output value is identified which crosses its corresponding said threshold,

(h) repeating steps (d)-(g) until said single adjusted output value is identified; and

(i) generating an output signal indicating an identity of said trail of said plurality of trails having said single adjusted output value that crosses its corresponding threshold.

2. A computer-based system as in claim 1 wherein said network is an optical telecommunications network and said trails include fiber optic cables.

3. A computer-based system as in claim 2 wherein said plurality of physical attributes of said plurality of trails includes installation method.

4. A computer-based system as in claim 2 wherein said plurality of physical attributes of said plurality of trails includes cable length.

5. A computer-based system as in claim 2 wherein said plurality of physical attributes of said plurality of trails includes cost to expand.

6. A computer-based system as in claim 1 wherein said transfer function is a sigmoid function.

7. A computer-based system as in claim 1 wherein said processing means is in communication with said plurality of trails and receives information from said plurality of trails representing changes in said plurality of physical attributes and wherein said processing means adjusts said plurality of attribute values in response to said changes in said plurality of physical attributes.

8. A computer-based system as in claim 1 wherein said processing means is in communication with said plurality of trails and said output signal generates a control signal to said plurality of trails for selecting said trail having said single adjusted threshold value that crosses its corresponding threshold.

9. A computer-based system for allocating resources in a communications network comprising a plurality of interconnected resources forming a plurality of optional pathways for communicating between a source point and a destination point within said communications network, wherein each pathway of said plurality of optional pathways comprises at least one resource and each resource of said plurality of interconnected resources has a plurality of physical attributes A, B, . . . to n, the computer-based system comprising:

means for assigning each physical attribute of the plurality of physical attributes with an attribute value chosen from a set of attribute values A(i=1 to x), B(j=1 to y), . . . to n(k=1 to z) for each said physical attribute;

a memory means for receiving and storing a data representation of each said pathway, said data representation including a set of attribute values {A(i), B(j), . . . to n(k)} for each said pathway representative of said plurality of physical attributes;

data entry means for entering into the memory means a plurality of numerical weight values W_A, W_B, . . . to W_N representing pre-determined priorities for each of said plurality of physical attributes, wherein a higher priority of the pre-determined priorities has a higher numerical weight value than a lower priority of the pre-determined priorities, and for entering a designation of said source point and said destination point;

a processing means connected to said memory means and said data entry means, the processing means for;

(a) multiplying each attribute value of said set of attribute values by a corresponding weight of said plurality of numerical weight values to provide a set of weighted attribute values {A(i)·

w_A, B(j)·

w_B, . . . n(k)·

w_N } for each said pathway,

(b) summing said set of weighted attribute values for each said pathway to generate an initial input value for each said pathway given by IN(pathway)=A(i)·

w_A +B(j)·

w_B + . . . +n(k)·

w_N,

(c) performing a transfer function on said initial input value to determine a threshold for each said pathway,

(d) adjusting at least one numerical weight value W_N of said plurality of numerical weight values in an order according to the pre-determined priorities to provide a set of adjusted weighted attribute values for each said pathway,

(e) summing said set of adjusted weighted attribute values to provide an adjusted input value for each said pathway given by IN(adjusted),

(f) performing said transfer function on said adjusted input value for each said pathway to generate an adjusted output value for each said pathway given by OUT=f(IN(adjusted)),

(g) comparing said adjusted output value for each said pathway to said threshold for each said pathway to identify a single optional pathway of said plurality of optional pathways having an adjusted output value that crosses its corresponding said threshold, and, if no said single optional pathway is identified,

(h) repeating steps (d)-(g), until said single optional pathway of said plurality of optional pathways is identified, and

(i) generating an output signal indicating an identity of said single optional pathway.

10. A computer-based system as in claim 9 wherein each said resource is a fiber optic cable terminated at each end by a device for transmitting/receiving a signal conducted through said fiber optic cable.

11. A computer-based system as in claim 9 wherein each said resource is an electrical power transmission line terminated at each end by a device for transmitting/receiving electrical power.

12. A computer-based system as in claim 9 wherein each said resource is an airline route terminated at each end by an airport.

13. A method for allocating resources in a connection oriented network based upon a pre-determined set of priorities, wherein each resource has a plurality of physical attributes, the method comprising:

assigning a plurality of numerical attribute values corresponding to said plurality of physical attributes for each said resource, said plurality of numerical attribute values for each resource representable as a numerical attribute set={a1, b1 . . . to n1]; and

storing said plurality of numerical attribute values into a computer memory;

entering into the computer memory a plurality of numerical weight values w(a), w(b), . . . to w(n), . . . for each physical attribute representative of said pre-determined set of priorities;

allocating a resource of the connection oriented network by processing the numerical weight values and the numerical attribute values, the method of processing comprising the steps of;

multiplying each numerical attribute value of said plurality of numerical attribute values by a corresponding numerical weight value of said plurality of numerical weight values to provide a set of weighted attribute values for each resource given by a weighted attribute set={a1*w(a), b1*w(b), . . . , n1*w(n)} for each resource;

adding said set of weighted attribute values for each said resource to generate an initial input value for each said resource;

performing a transfer function on said initial input value to determine a threshold for each said resource;

adjusting one of said plurality of numerical weight values according to said pre-determined set of priorities to provide a set of adjusted weighted attribute values;

adding said set of adjusted weighted attribute values to provide an adjusted input value for each said resource;

performing said transfer function on said adjusted input value for each said resource to generate an adjusted output value for each said resource;

comparing said adjusted output value to said threshold for each said resource to identify a single resource having said adjusted output value that crosses its corresponding said threshold;

if no said single resource is identified which has an adjusted output value that crosses its corresponding said threshold, repeating the steps of adjusting one of said plurality of numerical weight values, adding said set of adjusted weighted attribute values, performing said transfer function on said adjusted input value, and comparing said adjusted output value to said threshold for each said resource until said single adjusted output value is identified; and

generating an output signal indicating an identity of said single resource.

14. The method of claim 13 further comprising the step of adjusting one of said plurality of numerical attribute values when a corresponding one of said plurality of physical attributes changes.

15. The method of claim 14 further comprising the step of monitoring the network to determine whether said one of said plurality of physical attributes has changed.

16. The method of claim 13 wherein said network is an optical telecommunications network and said resources include fiber optic cables.

17. The method of claim 16 wherein said plurality of physical attributes includes installation method.

18. The method of claim 16 wherein said plurality of physical attributes includes cable length.

19. The method of claim 16 wherein said plurality of physical attributes includes cost to expand.

20. The method of claim 16 wherein the steps of performing said transfer function comprise applying a sigmoid function to said input values and adjusted input values.

21. The method of claim 13 wherein said network is a ground transportation grid and said resources include roadways.

22. The method of claim 13 wherein said network is an air transportation grid and said resources include airline flight paths.

23. A method for allocating a plurality of resources in a connection oriented network based upon a pre-determined set of priorities, wherein each resource r of said plurality has a plurality of physical attributes representable by a set of numerical values {A1, A2, . . . Ax}, the method comprising:

creating a representation of said resources of said connection oriented network;

storing said representation into a computer memory;

storing said set of numerical attribute values into said representation in the computer memory;

entering into said representation in the computer memory a plurality of initial numerical weight values w(A1), w(A2), . . . W(Ax) representative of said pre-determined set of priorities, with one initial numerical weight value of said plurality of initial numerical weight values corresponding to each physical attribute of said plurality of physical attributes;

multiplying each said initial numerical weight value with its corresponding said numerical attribute value to generate a plurality of weighted attribute values A1*w(A1), A2*w(A2), . . . Ax*w(Ax);

summing the plurality of weighted attribute values for each said resource r to generate a summed weighted attribute value given by a first equation IN(r)=A1*w(A1)+A2*w(A2)+ . . . +Ax*w(Ax);

applying a threshold function to the summed weighted attribute value to provide a relative resource value for comparison to a threshold value;

using a neural network training method to adjust at least one initial numerical weight value of said plurality of initial weight values and repeating the steps of multiplying, summing and applying a threshold function until the single relative resource value exceeds the threshold value; and

outputting a signal indicating at least one resource of said plurality of resources that corresponds to the single relative resource value.

24. The method of claim 23 wherein the neural network training method is backpropagation.