论文部分内容阅读
为了达到实时视频编码的低功耗、低带宽、省资源3个要求,文中基于改进的归一化部分失真搜索算法,提出一种新颖的运动估计硬件结构.新结构同时支持归一化部分失真搜索和自适应搜索区域调整.前者可降低运动估计的计算复杂度,从而实现低功耗省资源两个要求,后者能避免不必要的外存访问,从而降低数据带宽.在UMC 90 nm CMOS工艺下实现结果表明,相比于传统结构的最好结果,文中结构以6.2%的吞吐率损失,换取面积效率和功耗效率分别提高425.5%和397.5%.
In order to meet the requirements of low power consumption, low bandwidth and resource saving of real-time video coding, a novel hardware structure of motion estimation is proposed based on improved normalized partial distortion search algorithm. The new structure also supports normalized partial distortion Search and adaptive search area adjustment.The former can reduce the computational complexity of motion estimation in order to achieve two requirements of low power consumption resources, which can avoid unnecessary memory access, thereby reducing data bandwidth.In the UMC 90 nm CMOS The results of the process show that compared with the best results of the traditional structure, the structure of the paper is lost at the throughput rate of 6.2%, and the area efficiency and power consumption efficiency increase by 425.5% and 397.5% respectively.