Back to results
US 2020/0092175 A1Published application
Machine learning algorithms for quality of service assurance in network traffic
AssigneeCable Television Laboratories, Inc.
InventorsBernardo 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 future network traffic comprising:
- measuring network traffic volume at multiple points within a first time window to create a time series, each time series being associated with a measured future traffic rate within a second time window;
- multiplying the time series by a sparse, random binary matrix to obtain a vector of higher dimensionality than the time series;
- encoding the vector;
- mapping the measured future traffic rate to the encoding;
- repeating the steps of measuring, multiplying, encoding and mapping for different time windows to create a lookup table comprising a plurality of encodings and mappings;
- measuring, multiplying and encoding a current time series of network traffic;
- and
- matching the encoding for the current time series to one or more of the plurality of encodings within the lookup table.
5 more elements
A non-transitory computer-readable medium having a plurality of non-transitory instructions executable with a processor to predict future network traffic, the plurality of non-transitory instructions being executable for:
- measuring network traffic volume at multiple points within a first time window to create a time series, each time series being associated with a measured future traffic rate within a second time window;
- multiplying the time series by a sparse, random binary matrix to obtain a vector of higher dimensionality than the time series;
- encoding the vector;
- mapping the measured future traffic rate to the encoding;
- repeating the steps of measuring, multiplying, encoding and mapping for different time windows to create a lookup table comprising a plurality of encodings and mappings;
- measuring, multiplying and encoding a current time series of network traffic;
- and
- matching the encoding for the current time series to one or more of the plurality of encodings within the lookup table.
5 more elements
Dependencies are inferred from the claim text.