【摘要】 海洋潜标通过搭载各种观测设备实现海面以下的海洋环境要素的长期观测，是海洋环境监测中最可靠、最准确的手段之一。本文针对潜标传感器观测数据的特点，提出了一种数据预处理算法，该算法通过数据筛选、数据差分和数据去重，完成对海洋传感器数据的预处理。数据预处理算法的运算简便、消耗的临时空间资源少。Matlab平台仿真实验结果表明：对CTDDO传感器观测数据进行预处理后，动态Huffman算法的压缩效率提高了约86%、LZW算法的压缩效率提高了约72%、Mini-LZO算法的压缩效率最多提高约25%，并且数据预处理算法具备较强的通用性。数据预处理算法不仅可以大幅提高无损压缩算法对传感器观测数据的压缩效率，有利于有效节省数据存储空间、提高通信网络在单位时间内的吞吐传输能力；而且能够提高海洋潜标嵌入式系统的计算速度，有利于降低系统功耗，进而延长潜标的水下自主工作时间，具有广阔的应用前景。更多还原

【Abstract】 Marine submerged buoy is one of the most reliable and accurate means for monitoring the marine environment by carrying vari-ous observation equipment to achieve long-term observation of marine environmental elements below the surface. Aiming at the characteris-tics of sensor observation data, a data pre-processing algorithm is proposed, which completes the preprocessing of marine sensor data through data screening algorithm, differential compression algorithm and data deduplication algorithm. The data pre-processing algorithm is simple and consumes less temporary space resources. The simulation experiment results of Matlab show that with the pre-processing of CTD-DO sensor observation data, the compression efficiency of dynamic Huffman algorithm, LZW algorithm and Mini-LZO algorithm can be improved by 86%, 72% and up to 25% respectively; and the data pre-processing algorithm has strong versatility. The data pre-processing algorithm can not only greatly improve the compression efficiency of lossless compression algorithm for sensor observation data, which is beneficial to effectively save data storage space and improve the throughput and transmission capacity of communication network in unit time. Besides, it can improve the calculation speed of the embedded system of marine submerged buoy, which can reduce the power con-sumption of the system, and prolong the autonomous working time. In a word, the data pre-processing algorithm has broad application prospects.更多还原