Methods, devices and systems enhance compression and decompression of data blocks of data values by selecting the best suited compression method and device among two or a plurality of compression methods and devices, which are combined together and which said compression methods and devices compress effectively data values of particular data types; said best suited compression method and device is selected using as main selection criterion the dominating data type in a data block by predicting the data types within said data block.

Techniques for general-purpose lossless data compression using a neural network including compressing an original content item to a baseline lossless compressed data format. The baseline lossless compressed data format is binarized to a binarized format. The binarized format is arithmetically coded based on probability estimates from a neural network probability estimator. The neural network probability estimator generates the probability estimates for current symbols of the binarized format to be arithmetically coded based on symbols of the binarized format that have already been arithmetically coded.

A smart sensing architecture (100) includes smart meters (102) and processing units (104). The smart meters (102) generate and transmit multidimensional data streams to the processing units (104). A processing unit (104) determines an optimum batch size for a multidimensional data stream and generates a multidimensional batch of data. The processing unit (104) reduces dimensionality of the multidimensional batch of data using principal component analysis to generate a low-dimensional batch of data and performs compressive sampling on the low-dimensional batch of data to generate a compressed batch of data, thereby saving bandwidth of transmission.

Video coding systems or apparatus utilizing context-based adaptive binary arithmetic coding (CABAC) during encoding and/or decoding, are configured according to the invention with an enhanced binarization of non-zero Delta-QP (dQP). During binarization the value of dQP and the sign are separately encoded using unary coding and then combined into a binary string which also contains the dQP non-zero flag. This invention capitalizes on the statistical symmetry of positive and negative values of dQP and results in saving bits and thus a higher coding efficiency.

Data may be efficiently analyzed and compressed as part of a data compression service. A data compression request may be received from a client indicating data to be compressed. An analysis of the data or metadata associated with the data may be performed. In at least some embodiments, this analysis may be a rules-based analysis. Some embodiments may employ one or more machine learning techniques to historical compression data to update the rules-based analysis. One or more compression techniques may be selected out of a plurality of compression techniques to be applied to the data. Data compression candidates may then be generated according to the selected compression techniques. In some embodiments, a compression service restriction may be enforced. One of the data compression candidates may be selected and sent in a response.

A data storage device includes at least one data storage medium and a controller that is operably coupled to the at least one data storage medium. The controller receives the bit stream in a data storage device and performs a first level of compression on the received bit stream to obtain a symbol frame including a plurality of symbols. The controller encodes an initial portion of the plurality of symbols contained in the symbol frame by a fixed encoding scheme. The controller also collects statistics for the initial portion of the symbol frame. The controller then selects at least one data compression algorithm based on the collected statistics. The controller then performs compression encoding on a remaining portion of the symbol frame with the selected at least one data compression algorithm.

Disclosed herein are systems and methods for compressing structured or semi-structured data in a horizontal manner achieving compression ratios similar to vertical compression. Collections include structured or semi-structured data include a number of fields and are described using a schema. Fields include information having semantic similarity and are compressed using methods suitable for compressing the type of data. Data of a collection is compressed after fragmentation or may be normalized prior to compression. Data with semantic similarity is compressed using token tables and/or n-gram tables, where higher weighted, consisting of the product of frequency and length, occurring values may be stored in the lower numbered indices of the data table. Records include record descriptor bytes, field descriptor bytes, zero or more array descriptor bytes, zero or more object descriptor bytes, or bytes representing the data associated with the record. Data is indexed or compressed by a suitable module.

A runtime bit-plane data-format optimizer for a processing element includes a sparsity-detector and a compression-converter. The sparsity-detector selects a bit-plane compression-conversion format during a runtime of the processing element using a performance model that is based on a first sparsity pattern of first bit-plane data stored in a memory exterior to the processing element and a second sparsity pattern of second bit-plane data that is to be stored in a memory within the processing element. The second sparsity pattern is based on a runtime configuration of the processing element. The first bit-plane data is stored using a first bit-plane compression format and the bit-plane second data is to be stored using a second bit-plane compression format. The compression-conversion circuit converts the first bit-plane compression format of the first data to be the second bit-plane compression format of the second data.

Systems and methods for adaptive compression mode selection for memory buffers such as those used in or with a portable computing device (PCD) are presented. During operation of the PCD a first compression mode is selected for a buffer and the buffer is formatted to the first compression mode. Any access to the buffer by a component of the PCD, core of the PCD or software application running on the PCD is monitored. Based on the amount and/or type of access to the buffer, a second compression mode for the buffer is selected. The buffer is formatted to the second compression mode, providing a cost effective ability to adaptively format buffers based on the component(s), cores(s), and/or software application(s) accessing the buffers, and allowing for improving or optimizing bandwidth, memory footprint, resource conflict, power consumption, latency, and/or performance of component(s), core(s), or software application(s) accessing buffers as desired.

Some embodiments allow data conversions and meta-data to be specified using the page tables. As a result, individual memory resources can use their own compression methods or other conversions, and meta-data need only be specified for memory resources as necessary. Hardware engines and application code can access the memory resources as if they are stored in memory in the post-conversion, uncompressed form.