Back to results
US 8,554,746 B2Granted
Multiple-source data compression
AssigneeHewlett-Packard Development Company, L.P.
InventorsMarcelo Weinberger, Raul Herman Etkin, Erik Ordenllich, Gadiel Seroussi
Life of this patent
Estimating the term…
Abstract
One embodiment of the present invention is directed to a method for compressing data generated by multiple data sources. The method includes steps of partitioning data generated by the multiple data sources into data partitions, the data included in each data partition containing inter-data-source redundancies and, for each data partition, compressing the data in the data partition to remove the inter-data-source redundancies.
Claims17
A method for compressing data messages generated by multiple data sources, the data messages containing encoded signals, the method comprising:
- extracting and partitioning, by an electronic computing system, the encoded signals into data partitions, the data included in each data partition containing inter-data-source redundancies;
- and
- for each data partition, compressing, by an electronic computing system, the data in the data partition to remove the inter-data-source redundancies,
- wherein compressing the data in the data partition to remove the inter-data-source redundancies further includes:
- when the data partition includes multiple encoded data source signals,
- normalizing and aligning the encoded data source signals;
- ordering the multiple encoded data source signals to produce an ordering of the encoded data source signals that facilitates detection of inter-data-source redundancies during data compression;
- and
- applying a two-dimensional compression method to compress the ordered multiple encoded data source signals.
6 more elements
A distributed-sensor network comprising:
- a plurality of sensor nodes;
- and
- a data-collection entity to receive data messages from the sensor nodes and compress encoded sensor signals extracted from the data messages to remove inter-sensor-node data redundancy from the encoded sensor signals;
- wherein the data-collection entity is to:
- collect the data messages received from the sensor nodes during a time interval of interest;
- extract the encoded sensor signals from the received data messages;
- invoke a partitioning component to partition the encoded sensor signals received from the sensor nodes into data partitions containing inter-data-source redundancies;
- and
- invoke a multiple-data-source data-compression component to compress the encoded sensor signals in each data partition to remove the inter-data-source redundancies;
- and
- wherein, to compress the encoded sensor signals in a data partition to remove the inter-data-source redundancies, the multiple-data-source data-compression component is to:
- normalize and align the encoded sensor signals;
- order the normalized and aligned encoded sensor signals to produce an ordering of the encoded sensor signals to facilitate a detection of inter-data-source redundancies during data compression;
- and
- apply a two-dimensional compression method to compress the ordered encoded sensor signals.
12 more elements
A router that compresses data generated by multiple data sources, the router comprising:
- a partitioning component to partition encoded signals received from the data sources into data partitions;
- a multiple-data-source data-compression component to compress the encoded signals of each data partition to remove the inter-data-source redundancies;
- and
- a hardware processor to implement the partitioning component and the multiple-data-source data-compression component,
- wherein the partitioning component is to:
- apply, to each of the encoded signals received from one of the data sources, a partitioning function that returns a relatedness value;
- and
- for each relatedness value, group the encoded signals for which the partitioning function returned the relatedness value into a data partition associated with the relatedness value;
- wherein the relatedness value reflects a relatedness of data with respect to inter-data-source redundancies;
- and
- wherein the data-compression component is to:
- normalize and align the encoded signals;
- order the normalized and aligned encoded signals to produce an ordering of the encode signals to facilitate a detection of inter-data-source redundancies during data compression;
- and
- apply a two-dimensional compression method to compress the ordered encoded signals.
12 more elements
A router that compresses data generated by multiple data sources, the router comprising:
- a partitioning component to partition encoded signals received from the data sources into data partitions;
- a multiple-data-source data-compression component to compress the encoded signals of each data partition to remove the inter-data-source redundancies;
- and
- a hardware processor to implement the partitioning component and the multiple-data-source data-compression component,
- wherein the multiple-data-source data-compression component is to:
- normalize and align the encoded signals;
- order the normalized and aligned encoded signals to produce an ordering of the encoded signals to facilitate a detection of inter-data-source redundancies during data compression;
- and
- apply a two-dimensional compression method to compress the ordered encoded signals.
6 more elements
A distributed-sensor network comprising:
- a plurality of sensor nodes;
- and
- a data-collection entity to receive data messages from the sensor nodes and compress the encoded sensor signals extracted from the data messages to remove inter-sensor-node data redundancy from the encoded sensor signals;
- wherein the data-collection entity includes:
- a partitioning component to partition the encoded sensor signals received from the sensor nodes into data partitions;
- and
- a multiple-data-source data-compression component to compress the encoded sensor signals of each data partition to remove the inter-data-source redundancies;
- wherein the partitioning component is to:
- apply, to each of the encoded sensor signals received from one of the sensor nodes, a partitioning function that returns a relatedness value;
- and
- for each relatedness value, group the encoded sensor signals for which the partitioning function returned the relatedness value into a data partition associated with the relatedness value;
- wherein the relatedness value reflects a relatedness of data with respect to inter-data-source redundancies;
- and
- wherein the multiple-data-source data-compression component is to:
- normalize and align the encoded sensor signals;
- order the normalized and aligned encoded sensor signals to produce an ordering of the encoded sensor signals to facilitate a detection of inter-data-source redundancies during data compression;
- and
- apply a two-dimensional compression method to compress the ordered encoded sensor signals.
15 more elements
A distributed-seismic-sensor network including:
- a plurality of sensor nodes to detect, record, and transmit signals related to ground movement;
- a sensor router to receive signals from the plurality of sensor nodes and accumulate sensor data;
- and
- a data-storage and signal processing facility to receive the sensor data from the sensor router,
- wherein the sensor router is to:
- accumulate sensor messages from the plurality of sensor nodes over a time interval,
- partition the accumulated sensor messages to form a set of message partitions,
- normalize and align the sensor messages in each message partition;
- order the normalized and aligned sensor messages to produce an ordering of the sensor messages to facilitate a detection of inter-data-source redundancies during data compression, and
- apply a two-dimensional compression method to compress the ordered sensor messages, and
- transmit the compressed message to the data-storage and signal processing facility.
8 more elements
Dependencies are inferred from the claim text.