Scaling trends of power noise in 3-D ICs

Power supply noise in three-dimensional integrated circuits (3-D ICs) considering scaled CMOS and through silicon via (TSV) technologies is the focus of this paper. A TSV and inductance aware cell-based 3-D power network model is proposed and evaluated. Constant TSV aspect ratio and constant TSV area penalty scaling, as two scenarios of TSV technology scaling, are discussed. A comparison of power noise among via-first, via-middle, and via-last TSV technologies with CMOS scaling is also presented. When the TSV technology is a primary bottleneck in high performance 3-D ICs, an increasing TSV area penalty should be adopted to produce lower power noise. As a promising TSV technology, via-middle TSVs are shown to produce the lowest power noise with CMOS technology scaling.     

基于噪声估计的自适应块匹配和三维滤波降噪算法

针对经典的块匹配和三维滤波(BM3D)降噪算法中最 为核心的噪声水平(方差)参数在使用中需要 人工手动设置极大影响了降噪效果并限制了它的应用,提出了一种新的基于自然场景 统计(NSS)的噪声水平特征矢量和支持向量回归(SVR)技术的快速噪声水平估计算法并应用于 经典BM3D算法 中,使之转变为自适应降噪算法(Adaptive BM3D)。本文算法首先利用小波变换对图像进行 不 同尺度和不同方向的分解,提取各子带滤波系数并用通用高斯分布模型(GGD)建模,以模型 参数构成反映噪 声图像噪声水平的特征矢量;然后用SVR方法在大量噪声图像样本上进行训练获得图像噪声 水平预测模型。 实验表明:改进后的ABM3D算法实际图像降噪效果比BM3D算法获得进一步提升,并且仍然 保持了非常高的执行效率,相对于当前各主流算法具有明显的竞争力。     

RFID-based 3D human pose tracking: A subject generalization approach

Three-dimensional (3D) human pose tracking has recently attracted more and more attention in the computer vision field. Real-time pose tracking is highly useful in various domains such as video surveillance, somatosensory games, and human-computer interaction. However, vision-based pose tracking techniques usually raise privacy concerns, making human pose tracking without vision data usage an important problem. Thus, we propose using Radio Frequency Identification (RFID) as a pose tracking technique via a low-cost wearable sensing device. Although our prior work illustrated how deep learning could transfer RFID data into real-time human poses, generalization for different subjects remains challenging. This paper proposes a subject-adaptive technique to address this generalization problem. In the proposed system, termed Cycle-Pose, we leverage a cross-skeleton learning structure to improve the adaptability of the deep learning model to different human skeletons. Moreover, our novel cycle kinematic network is proposed for unpaired RFID and labeled pose data from different subjects. The Cycle-Pose system is implemented and evaluated by comparing its prototype with a traditional RFID pose tracking system. The experimental results demonstrate that Cycle-Pose can achieve lower estimation error and better subject generalization than the traditional system.     

一种对三维芯片中硅通孔进行完整建模与寄生参数提取的有效方法

针对三维芯片中硅通孔（through-silicon via, TSV）的准确电学建模问题,本文提出了一种电阻电容（RC）电路模型以及相应的有效参数提取技术。该电路模型同时考虑了半导体效应与静电场影响,适合于低频与中频的电路信号范围。该方法采用一种基于悬浮随机行走（floating random walk, FRW）算法的静电场电容提取技术,然后将它与刻画半导体效应的MOS电容结合,形成等效电路模型。与Synopsys公司软件Sdevice所采用的对静电场/半导体效应进行完整仿真的方法相比,本文方法计算效率更高,并且也能处理一般的TSV电路版图。对多个含TSV的结构进行了计算实验,结果验证了本文方法在从10KHz到1GHz频率范围内的建模准确性,也显示出它相比Sdevice方法最多有47倍的加速比。