一种基于高频噪声定量约束函数的地震高分辨率处理方法

系统分析了现有地震数据高分辨率处理方法的适用性及不足,探索了一种基于高频噪声定量约束函数的高分辨率处理方法,有效且保真地提高地震资料信噪比与分辨率。通过分析叠后地震数据有效信号带宽,设计了压制高频噪声的最大概率准则和定量约束函数,通过多次迭代压制噪声,提高全频带信号的信噪比,逐步分级拓宽地震有效频带宽度,循环迭代求取反射系数,将提取的地震子波与反射系数褶积得到高分辨率地震数据体。测试校验对比其它高分辨率地震处理结果,处理后的数据体具有垂向分辨率高、主频高且频带宽的特点,有效视主频由50 Hz提高至100 Hz左右,地质目标视分辨率相应提高近1倍。基于该提频数据与波阻抗反演实现了2~5 m地质目标的定性与定量预测,钻井验证符合率90%以上,有效指导井位部署。该方法弥补了在传统“有效带宽”内提高地震高频信号的不足,实现了全频带内重构高、低频有效信号的目的。

多保真度数据学习算法的定量噪声评价

多保真度数据是当前材料领域数据的主要存在形式。在数据生产端,不同量化方法在材料同种属性的计算上存在较大差距。对于数据消费端的机器学习算法,研究人员为最大化提取数据中知识设计了各种方法。采用定量噪声添加的方法,评价不同噪声强度、类型对不同多保真度数据学习方法的影响,通过迭代降噪验证数据修正方法的适用场景。结果表明:多保真度数据的利用方式至关重要,需对各子数据集中数据量及含噪情况进行综合考量。在使用不同噪声类型与强度构造出的多种数据集上,得益于数据间的协同效应,逐步删除低保真度数据的"Onion"训练方式明显优于按数据集所含噪声减小方向逐个进行的训练方式。在多保真度数据训练中,无论何种噪声强度及训练方式,线性噪声对模型的影响更小。对于采样噪声来说,在各环节更好地模拟了真实多保真度数据,建议被后续研究采用。此外,复杂噪声难以让少量真值数据发挥"纠偏"作用,更适合进行迭代降噪处理。

DISCRETE BIDIRECTIONAL ASSOCIATIVE MEMORY WITH LEARNING FUNCTION

In this paper we propose a new discrete bidirectional associative memory (DBAM) which is derived from our previous continuous linear bidirectional associative memory (LBAM). The DBAM performs bidirectionally the optimal associative mapping proposed by Kohonen. Like LBAM and NBAM proposed by one of the present authors,the present BAM ensures the guaranteed recall of all stored patterns,and possesses far higher capacity compared with other existing BAMs,and like NBAM, has the strong ability to suppress the noise occurring in the output patterns and therefore reduce largely the spurious patterns. The derivation of DBAM is given and the stability of DBAM is proved. We also derive a learning algorithm for DBAM,which has iterative form and make the network learn new patterns easily. Compared with NBAM the present BAM can be easily implemented by software.

Pixel and Column Fixed Pattern Noise Suppression Mechanism in CMOS Image Sensor

A double sampling circuit to eliminating fixed pattern noise(FPN) in CMOS image sensor (CIS) is presented. Double sampling is implemented by column switch capacitor amplifier directly, and offset compensation is added to the amplifier to suppress column FPN. The amplifier is embedded in a 64×64 CIS and successfully fabricated with chartered 0.35 μm process. Theory analysis and circuit simulation indicate that FPN can be suppressed from millivolt to microvolt. Test results show that FPN is smaller than one least-significant bit of 8 bit ADC. FPN is reduced to an acceptable level with double sampling technique implemented with switch capacitor amplifier.

A continuous differentiable wavelet threshold function for speech enhancement

Enhanced speech based on the traditional wavelet threshold function had auditory oscillation distortion and the low signal-to-noise ratio (SNR). In order to solve these problems, a new continuous differentiable threshold function for speech enhancement was presented. Firstly, the function adopted narrow threshold areas, preserved the smaller signal speech, and improved the speech quality; secondly, based on the properties of the continuous differentiable and non-fixed deviation, each area function was attained gradually by using the method of mathematical derivation. It ensured that enhanced speech was continuous and smooth; it removed the auditory oscillation distortion; finally, combined with the Bark wavelet packets, it further improved human auditory perception. Experimental results show that the segmental SNR and PESQ (perceptual evaluation of speech quality) of the enhanced speech using this method increase effectively, compared with the existing speech enhancement algorithms based on wavelet threshold.

Performance of cumulant-based rank reduction estimator in presence of unexpected modeling errors

Compared with the rank reduction estimator(RARE) based on second-order statistics(called SOS-RARE), the RARE based on fourth-order cumulants(referred to as FOC-RARE) can handle more sources and restrain the negative impacts of the Gaussian colored noise. However, the unexpected modeling errors appearing in practice are known to significantly degrade the performance of the RARE. Therefore, the direction-of-arrival(DOA) estimation performance of the FOC-RARE is quantitatively derived. The explicit expression for direction-finding(DF) error is derived via the first-order perturbation analysis, and then the theoretical formula for the mean square error(MSE) is given. Simulation results demonstrate the validation of the theoretical analysis and reveal that the FOC-RARE is more robust to the unexpected modeling errors than the SOS-RARE.

Classical information theory in quantum notations

Shannon's noisy channel coding theorem is a cornerstone in the history of classical information theory. A proof of Shannon's noisy channel coding theorem using quantum notations is presented, which sets up a bridge between classical information theory and the developing quantum information theory.