Data processing method and device

Abstract

The embodiment of the invention discloses a data processing method and device. The method comprises the steps: storing at least two parts of original data, and enabling each part of original data to be stored in a plurality of RAMs (i.e., RAM groups) in a distributed manner, thereby enabling different parts of data covered by a sliding window to be read from the plurality of RAM groups at the sametime when the data covered by the sliding window occupies a plurality of address units; therefore, the parallel reading of the sliding window data in one read clock period is realized, so that the data processing efficiency of the embedded system is improved.

Description

technical field [0001] The present invention relates to the technical field of data processing, and more specifically, to a data processing method and device. Background technique [0002] When an embedded system performs data processing, the general process is to read data from the memory, then send the read data to the computing unit to complete specific operations, and finally output the calculation results. When the current embedded system reads data, it reads in sequence, that is, it can only read the data in one address in each clock cycle. This processing method makes the data reading speed slower, especially if there is slippage The data processing algorithm required by the window extraction processing may cause the data in a sliding window to be read within multiple clock cycles, and the data in the sliding window cannot be efficiently obtained, so the real-time performance of data processing cannot meet the needs of high-speed processing. [0003] Therefore, how t...

AI Technical Summary

Problems solved by technology

When the current embedded system reads data, it reads in sequence, that is, it can only read the data in one address in each clock cycle. This processing method makes the data reading speed slower, especially if there is slippage The data processing algorithm required by the window extraction processing may cause the data in a sliding window to be read within multiple clock cycles, and the data in the sliding window cannot be efficiently obtained, so the real-time performance of data processing cannot meet the needs of high-speed processing

Images