The Lt-norm method is one of the widely used matching filters for adaptive multiple subtraction. When the primaries and multiples are mixed together, the L1-norm method might damage the primaries, leading to poor late...The Lt-norm method is one of the widely used matching filters for adaptive multiple subtraction. When the primaries and multiples are mixed together, the L1-norm method might damage the primaries, leading to poor lateral continuity. In this paper, we propose a constrained L1-norm method for adaptive multiple subtraction by introducing the lateral continuity constraint for the estimated primaries. We measure the lateral continuity using prediction-error filters （PEF）. We illustrate our method with the synthetic Pluto dataset. The results show that the constrained L1-norm method can simultaneously attenuate the multiples and preserve the primaries.展开更多
文摘Y99-61803-55 2002075用于助听器反馈对消的密集、超低功率、可编程连续滤波器组=Compact,Ultra low power,programmablecontinuoustime filter banks for feedback cancellation inhearing aid[会,英]/Nair,K.& Harjani,R.//Proceed-ings of the 12th International Conference of VLSI De-sign.—55～60(EZ)描述了一种小型超低功率连续可编程滤波器的设计。它具有广泛用途。其特殊用途是助听器中的反馈对消滤波器,它与开环传递函数相匹配,用于抑制声音振荡。为实现集成对消滤波器,设计了超低功率大线性范围跨导单元。在3V 供电时跨导单元功耗为0.8μm。采用2μCMOS 技术对滤波器进行了设计与制造。
基金This work is sponsored by National Natural Science Foundation of China （No. 40874056）, Important National Science ＆ Technology Specific Projects 2008ZX05023-005-004, and the NCET Fund.Acknowledgements The authors are grateful to Liu Yang, and Zhu Sheng-wang for their constructive remarks on this manuscript.
文摘The Lt-norm method is one of the widely used matching filters for adaptive multiple subtraction. When the primaries and multiples are mixed together, the L1-norm method might damage the primaries, leading to poor lateral continuity. In this paper, we propose a constrained L1-norm method for adaptive multiple subtraction by introducing the lateral continuity constraint for the estimated primaries. We measure the lateral continuity using prediction-error filters （PEF）. We illustrate our method with the synthetic Pluto dataset. The results show that the constrained L1-norm method can simultaneously attenuate the multiples and preserve the primaries.
