International Business Machines Corporation

Duke University

Erol Gelenbe

Gerald Arnold Marin

Raif O Onvural

Yutao Feng

Xiaowen Mang

Methods, systems and computer program products are provided for evaluating requests for a network connection combining the requested network connection with existing connections so as to provide a representation of the total network connections and accepting the request for a network connection if sufficient resources are available to accept the request for a network connection based upon the representation of the total network connections. Particular embodiments of the present invention determine if sufficient resources are available to accept the request for a network connection based upon diffusion based representations of the existing network connections and the requested network connection and accepting the request for a network connection if sufficient resources are available to accept the request for a network connection. The diffusion based representation may be used to determine the capacity required for existing connections and the requested connection based upon a predefined maximum loss ratio (L).

Statistical call admission control

Image content classification methods, systems and computer programs repeatedly scan an image having an array of image pixels, with at least one random neural network. Each scan corresponds to one of multiple texture patterns. A corresponding texture pattern is compared to each of multiple image portions for each of the multiple scans. A value is assigned to each image portion, corresponding to the texture pattern having the highest coincidence. An array of pixels corresponding to the assigned values for the image portions may then be displayed. Highly accurate results may be obtained, at high speed, without the need for lengthy expert analysis.

Growing community of inventors

Durham, NC, United States of America

Erol Gelenbe