大数据背景下的信号处理

宽带信号、高维信号、高分辨信号以及多传感器组网技术的快速发展，使得信号采集数据增长率高于数据存储增长率和信号处理速度增长率，信号处理进入了大数据时代．本文指出了大数据背景下的信号处理的关键问题．对于多传感器组网具有的多样性和复杂性的海量信号数据，必须进行信息融合．本文介绍了信号融合的主要模型方法，分析了信息融合技术的发展趋势．智能传感网技术能降低对信号处理和通信容量的要求，有效地在大数据中提取有价值数据．本文给出了智能传感器的基本结构，阐述了智能传感器的计算方法．结合大数据容量信号对高速实时处理的要求，介绍了高速数字信号处理芯片以及高性能硬件平台发展现状，展望了高速信号处理核心技术的发展动向．

Signal processing in the context of big data

The rapid developments of wideband signal, high-dimensional signal, high resolution image, and multi-sensor network have made tile growth rate of data acquisition amount higher than the growth rate of data storage amount and the growth rate of signal processing speed. Signal processing steps into tile big data era. This paper points out the key issues of signaI processing with big data. Considering the diversity and complexity of the big data produced by nmlti-sensor networks, it is necessary to conduct information fusion. The main information fusion models are described, and the trends of information fusion technology are analyzed. Intelligent sensor network technology can reduce tile capacity of signal processing and communication, and efficiently extract valuable information from big data. The basic architecture of a smart sensor is given, and calculation methods for intelligent sensor networks are described. From viewpoint of high-speed and reaLtime signal processing requirements of signals with big data, we also introduce developments of high-speed digital signal processing chips and high-performance hardware platforms. The vision of core technology in high-speed signal processing is presented.