34results about How to "Improve Bandwidth Utilization Efficiency" patented technology

Bandwidth utilization within a cellular network is increase by reducing the amount of traffic to be transmitted, in particularly in traffic that has already been compressed. An encoded signal that comprises an umber of frame signals (or simply frames) is received and each frame is classified in accordance with one or more pre-defined characterization criterion. The characterization can include identifying the frames as speech type signals, such as voice signals or noise signal. Voice signal frames may be further characterized as a stationary frames (where the voice signal is essentially at a constant level), as a transition frame between phonemes, and the like. A noise type of frame may be further characterized as a silence frame, a background noise and the like. Video frame types can be characterized as a frame with a rapid / slow change in respect to the preceding frame, a frame with a rapid / slow change in respect to pixels in that frame, and the like. Depending on the classification of each frame, the encoded signal may be formatted and be replaced by a corresponding representation signal, wherein the number of bits comprised in a plurality of the formatted signals, is less than the number of bits comprised in the received encoded signal. Consequently, a formatted frame signal may be represented by a selected corresponding representation signal, or by a number of selected a corresponding representation signals which correspond to the sub-frame signals, and when taken together, carry the information required for the regeneration or reconstruction of the entire encoded frame signal.

The present application relates to an image compression method. A computer device divides an image to be compressed into multiple compression units; each compression unit includes at least two sub-blocks; and then, based on a preset compressed data format, the compressed data of each compression unit is respectively obtained ; Wherein, the compressed data includes the reference pixel value of the compression unit, the first pixel associated data, the second pixel associated data, the sub-block position where the reference pixel value of the compression unit is located, and the pixel position where the reference pixel value of each sub-block is located; The first pixel correlation data is used to represent the pixel relationship between the compression unit and each sub-block in the compression unit, and the second pixel correlation data is used to represent the pixel relationship in each sub-block. By adopting the above method, the compression ratio of the image to be compressed can be improved, the bandwidth utilization efficiency can be improved, and the time delay for reading compressed data can be reduced; information distortion caused by the compression of the image to be compressed can be avoided.

The invention discloses a supercomputer-based optimization method for a fluid machinery simulation program. According to the architecture and programming characteristics of Shenwei-Taihu Light supercomputing system and the characteristics of fluid mechanical simulation program, a feasible optimization scheme of the system is proposed, which includes block-based multi-core parallel optimization, DMA transmission optimization, data layout optimization, double buffer optimization, SIMD vectorization optimization and register communication optimization in turn. This method provides a general optimization method for the developer who develops, transplants or optimizes the fluid mechanical simulation program for the Shenwei-Taihu Lake Light supercomputing platform, realizes the full utilizationof Shenwei-Taihu Light computing resources, improves the program performance, and shortens the simulation time.