Subsurface radar signal deconvolution

We present two methods of signal deconvolution for systems whose impulse response (wavelet function) can be explicitly determined, and where the goal is to locate short impulses in the presence of strong, reverberation-like interferences.The first method, which we call algebraic deconvolution, differs from other known techniques in two ways: first of all, explicit use of the wavelet function provides more powerful a priori knowledge than the autocorrelation or the power spectrum. Secondly, this method permits to flexibly trade off noise versus resolution.In the second method presented here, we use an analytical model (synthetic wavelet) of the system impulse response to determine an inverse filter.These methods have been developed for video pulse radar signals, and encouraging early results have been obtained.     

Triangular lattice Potts models

In this paper we study the 3-state Potts model on the triangular lattice which has two- and three-site interactions. Using a Peierls argument we obtain a rigorous bound on the transition temperature, thereby disproving a conjecture on the location of its critical point. Low-temperature series are generated and analyzed for three particular choices of the coupling constants; a phase diagram is then drawn on the basis of these considerations. Our analysis indicates that the antiferromagnetic transition and the transition along the coexistence line are of first order, implying the existence of a multicritical point in the ferromagnetic region. Relation of the triangularq-state Potts model with other lattice-statistical problems is also discussed. In particular, an Ashkin-Teller model and the hard-hexagon lattice gas solved by Baxter emerge as special cases in appropriate limits.Supported in part by NSF grant No. DMR 78-18808.     

The global Markov property for lattice systems

We prove the global Markov property for lattice systems of classical statistical mechanics, with bounded spins and finite range interactions. The method uses the one developed by two of us to prove the global Markov property of Euclidean generalized random fields. The result shows that the systems considered have a transition matrix, which together with a distribution on a hyperplane, describes completely the system.     

Phase transitions in lattice gas models for water

Water-like lattice gases on the triangular and body-centered cubic lattices are investigated. Molecules may reside on the lattice sites in either of two possible orientations, a hydrogen bond being formed between molecules on neighboring sites if they have the proper orientation with respect to one another. For a range of chemical potential at sufficiently low temperatures, the models are shown to have an ordered phase consisting of an open, hydrogen-bonded, icelike structure. The models are shown to be transitionfree at sufficiently high temperature, indicating the existence of a critical point.Research supported in part by the U.S. National Science Foundation, Grant CHE-7726177, and by The Robert A. Welch Foundation, Grant P-446.     

Global solutions of the Boltzmann equation on a lattice

The nonlinear Boltzmann equation with a discretized spatial variable is studied in a Banach space of absolutely integrable functions of the velocity variables. Conservation laws and positivity are utilized to extend weak local solutions to a global solution. This is shown to be a strong solution by analytic semigroup techniques.Supported by National Science Foundation Grant ENG-7515882.     

一种对噪声混沌信号的滤波方法

本文研究噪声混沌信号的滤波问题。组合信号建模技术,提出了一种基于无先导变换(Unscented Transform,UT)的滤波方法。仿真结果表明,不管混沌系统的参数如何变化,这种算法能都能有效地抑制噪声对混沌系统的影响。与基于扩展卡尔曼滤波器(Extended Kalman filter,EKF)算法相比,在低信噪比的情况下,基于UT的滤波方法都有较好的滤波性能;而在高信噪比的情况下,它有与扩展卡尔曼滤波器基本相同的性能。     

小抽取倍数的FIR高效实现

对采样信号进行抽取是软件无线电中的一个重要环节,小抽取率时通常使用有限冲击响应(FIR)滤波器.当FIR滤波器的阶数较高时,将会对信号处理的速度产生较大的影响.针对级联-积分-梳状(CIC)滤波器系数少且全为"1"的特点,提出了在小抽取率时采用单级CIC滤波器和FIR滤波器级联的方式来减少FIR滤波器的阶数,从理论上给出了FIR滤波器减少的阶数和CIC滤波器的阶数之间的关系.在此基础上,借助多相滤波结构还可进一步降低FIR滤波器阶数.计算机模拟验证了方法的有效性.     

Design and performance of millimeter-wave end-coupled bandpass filters

The design and performance of millimeter-wave microstrip/suspended-stripline end-coupled bandpass filters are presented. Filters in Ka-band (26.5 to 40 GHz) and W-band (75 to 110 GHz) using suspended-stripline and microstrip have been designed. Good agreement between the experimental results and those predicted theoretically was observed. Measured passband insertion losses of less than 1 and 0.5 dB have been achieved in W- and Ka-band, respectively. An extremely wideband waveguide-to-microstrip transition with an insertion loss of 0.25 dB over the full W-band is also presented.     

Versatility of the dipolar filter selection: from 1H nuclear spin diffusion experiment to the measurement of nuclear Overhauser effect in homopolymer melts

Dipolar filters select ¹H magnetization according to local dipolar dephasing, which corresponds to site mobility in systems with heterogeneous molecular mobility. Combined with a conventional exchange experiment, it is usually applied to polymeric samples exhibiting structures on the nanometer length scale associated with a strong dynamic contrast. There, the resulting ¹H nuclear spin diffusion experiment yields the size of the structure. When the same experiment is applied to homopolymer melts exhibiting a weak dynamic contrast and dynamic heterogeneities on significant shorter length scales, the recorded magnetization decay is in agreement with decays expected from a heterogeneous nanostructure. However, dipolar filters actually can also select mobile parts of the repeat unit, e.g. the end of the alkyl side chains and the subsequent magnetization transfer then can occur via cross relaxation due to non coherent zero-quantum transitions (nuclear Overhauser effect, NOE). The difficulties of distinguishing these two cases are examined and it is demonstrated that NOE experiments exploiting magnetization selection via the dipolar filter allow quantifying the local dynamics of the side chains. This opens new possibilities for measurements of local dynamics in non isotopically labeled homopolymer melts.     

用于投影显示的分色合色膜系的消偏振设计

由于斜入射时,光学薄膜存在一定的偏振效应,将产生S-和P-偏振光的光谱分离.在Philips棱镜系统中,分色合色膜系通常分色时对S-光应用,合色时则对P-光应用.从提高光能利用率、减少杂散光和增加系统对比度等因素综合考虑,要求分色合色膜系的S-和P-偏振光的分离尽可能小.新设计方法采用宽带法布里－珀罗薄膜干涉滤光片中心波长两侧的干涉带作为长波通或短波通截止滤光片,可实现S-和P-偏振分量的分离几乎为零.