【24h】

DCT/IDCT PROCESSOR DESIGN FOR HDTV APPLICATIONS

机译:HDTV应用的DCT / IDCT处理器设计

获取原文
获取原文并翻译 | 示例
获取外文期刊封面目录资料

摘要

Transform coders, in particular the Discrete Cosine Transform (DCT), have been widely used in the implementation of low-rate codecs for video compression. The DCT has become an integral part of several standards such as MPEG [1], and CCITT Recommendations H. 261 and H.263 [2]. For real time implementation of the DCT at MPEG rates, general purpose DSP's [3] have been proposed. However, with the imminent arrival of High Definition Television and the considerably higher sample rates (75 MHz sampling frequencies) and higher throughput requirements, processors specific to the DCT (ASIC's) with highly parallel and/or pipelined architectures have been recently reported [4]. The use of such processors can be justified only if they result in a cost effective implementation. In this paper we will describe a simple but cost effective design of a combined DCT/IDCT processor. The implementation uses a parallel architecture that exploits symmetry properties in the DCT/IDCT algorithm and requires only simple hardwired multipliers. Clocking at 100 MHz, the processor is capable of handling real-time HDTV signals. The core area of the processor has been minimized while keeping the I/O requirements simple. Finite wordlength and accuracy studies have been carried out to ensure that the resulting implementation conforms to existing standards (H.261)[2].
机译:变换编码器,尤其是离散余弦变换(DCT),已广泛用于实现视频压缩的低速率编解码器。 DCT已成为诸如MPEG [1]和CCITT建议H. 261和H.263 [2]之类的几个标准的组成部分。为了以MPEG速率实时实现DCT,已经提出了通用DSP [3]。然而,随着高清电视的到来以及更高的采样率(75 MHz采样频率)和更高的吞吐率要求,最近已经报道了具有高度并行和/或流水线架构的DCT(ASIC)专用处理器[4]。 。仅当此类处理器可实现具有成本效益的实施方式时,才有理由使用它们。在本文中,我们将描述一种简单但经济高效的组合DCT / IDCT处理器设计。该实现使用并行架构,该架构利用DCT / IDCT算法中的对称属性,并且仅需要简单的硬连线乘法器。该处理器的时钟频率为100 MHz,能够处理实时HDTV信号。在保持I / O要求简单的同时,处理器的核心区域已被最小化。已经进行了有限的字长和准确性研究,以确保最终的实现符合现有标准(H.261)[2]。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号