Machine learning algorithms for quality of service assurance in network traffic
AssigneeCable Television Laboratories, Inc.
InventorsBernado HUBERMAN, Scott H. CLEARWATER
Life of this patent
Estimating the term…
Abstract
The present disclosure generally relates to apparatus, software and methods for predicting future network traffic. The disclosed apparatus, software and methods alleviate congestion and/or increase overall traffic flow by providing methods for reallocating future idle capacity.
Claims20
A method for predicting a probability of successful transmission from a device connected to a common node comprising:
- determining a state of each of a plurality of devices connected to a common node, the state comprising at least a time of transmission and whether an attempted transmission is successful or not;
- constructing a vector of the state of each device whenever any of the plurality of devices attempts to transmit to the common node;
- multiplying the vector by a sparse, random binary matrix to obtain an encoding;
- repeating the steps of determining, constructing, and multiplying to generate a lookup table of encodings;
- producing a current encoding by multiplying a vector for a current device state by the sparse, random binary matrix;
- matching the current encoding to one or more encodings in the lookup table;
- and
- providing a probability of successful transmission for the current encoding based on results of the attempted transmissions of the matching encoding(s).
A non-transitory computer-readable medium having a plurality of non-transitory instructions executable with a processor for predicting a probability of successful transmission from a device connected to a common node, the plurality of non-transitory instructions being executable for:
- determining a state of each of a plurality of devices connected to a common node, the state comprising at least a time of transmission and whether an attempted transmission is successful or not;
- constructing a vector of the state of each device whenever any of the plurality of devices attempts to transmit to the common node;
- multiplying the vector by a sparse, random binary matrix to obtain an encoding;
- repeating the steps of determining, constructing, and multiplying to generate a lookup table of encodings;
- producing a current encoding by multiplying a vector for a current device state by the sparse, random binary matrix;
- matching the current encoding to one or more encodings in the lookup table;
- and
- providing a probability of successful transmission for the current encoding based on results of the attempted transmissions of the matching encoding(s).
Dependencies are inferred from the claim text.