PCM/CVSD系统中内插和抽取高效滤波算法

介绍了内插和抽取滤波器的应用背景(PCM和CVSD实时相互转换系统)。提出了一种基于半带滤波器的简单而有效的内插和抽取滤波器的设计思路,并在现有算法的基础上分别提出了内插滤波器和抽取滤波器的高效实现算法。从而提高了在以TMS320VC5416为硬件平台的PCM和CVSD实时相互转换系统的效率。     

基于Python的级联半带滤波器RTL自动生成方法

针对数字上下变频的滤波需求和自动化的发展趋势,提出了一种级联半带滤波器寄存器传输级（RTL）自动生成方法。该方法首先读取设计需求参数,创建Python字典对象,将设计参数存储于此字典对象中;然后读取字典对象中的设计参数,基于Python分析并生成并行流水线结构的级联半带滤波器RTL代码。试验结果表明,提出的方法能够自动生成任意N级的级联半带插值滤波器或者任意N级的级联半带抽取滤波器,生成的滤波器能够实现2N倍的插值或者抽取,提升了设计效率,缩短了开发周期。     

高阶重采样滤波器的多项式近似实现

可变载频带限信号的重采样,一般归结为按转换比P/Q(P为内插比,Q为抽取比)对原采样序列做内插和抽取。当P值很大时,需要多路内插滤波器,由于抗镜像的需要,滤波器系数矩阵非常庞大,使得高阶重采样难以实现。该文提出一种多项式近似滤波器的方法,用一组低阶多项式近似内插滤波器系数矩阵,简化了滤波器的结构,运算效率高,且内插延迟可任意改变。计算机仿真的结果表明:该结构适用于可变延迟的高阶带限内插滤波器。在一定条件下,误差在容许的范围之内。     

基带采样率内任意带通信号的正交采样

提出了一种在基带采样率内对任意带通信号的正交采样技术。载频位于ADC(模/数转换器)基带采样率内的任意带通信号,经1/2抽取和(-1)n调制,再由全通线性多相滤波器内插后,得到其复包络的调制输出XI(n)和XQ(n)。     

DVB-T中多速率DDC的设计与FPGA实现

针对DVB-T标准ETSI EN 300 744 V1.5.1,设计了可用于DVB-T接收整机的多速率DDC模块,并在FPGA中仿真实现.在复用数字振荡混频模块的基础上,根据输入信号的不同带宽(6M/8MHz)选择不同的抽取滤波器组完成抽取因子为3或4的多速率处理任务,利用两级半带滤波器(HBF)级联完成4倍抽取滤波,单级奈奎斯特滤波器完成3倍抽取滤波.     

4路多相信号半带抽取滤波器优化实现结构

高速FIR滤波器的4路多相实现结构工作的采样速率是单路串行实现结构的4倍,针对4路多相信号半带抽取滤波器直接实现结构计算复杂度大这一问题,提出了一种4路多相信号半带抽取滤波器的优化实现结构。推导得到4路多相信号FIR滤波器优化实现结构;在此基础上,分析不同输出组合的计算复杂度,给出4路多相信号半带抽取滤波器的优化实现结构。对于4路多相信号半带抽取滤波器,仿真结果表明,提出的优化实现结构的计算复杂度约为直接实现结构的75%,验证了其优越性。     

基于抽取滤波器多相分解的盲自适应符号同步算法

针对软件无线电系统中接收信号采样时钟独立于发送信号时钟的特点,提出了在抽取操作过程中实现符号同步,避免了在基带阶段进行内插操作,以得到多相位信息的做法．利用过抽样信号固有的多相位信息,通过最大似然相位误差捡测算法,调整延时机构,使信号传输延时匹配于延时分枝索引值,实现后续信号抽取输出为最佳采样值．利用级联积分梳状(CIC)滤波器和半带滤波器(HBF)作为抽取滤波器,应用滤波器多相分解算法,降低了系统实现的复杂性和运算量．仿真结果表明,当信号传输延时匹配于延时分枝索引值时,相位误差和延时分枝索引值将趋于稳定．     

基于FPGA的高速抽取过程的实现

下变频后的高速基带信号必须经过抽取后才能减轻后续基带处理的负担。多级抽取滤波有利于滤波器的实现。梳状滤波器用作多级抽取的首级，半带滤波器可以实现2倍的抽取，剩余的抽取滤波器用升幅FIR实现。针对各级滤波器系数的特点，在滤波器的硬件实现上采用了一系列简化方法。计算机仿真结果表明，用FPGA实现高速抽取系统是十分灵活有效的。     

一种时延设计方法与DSP实现

多径时延是信道模拟中一个重要的研究技术指标.介绍一种采用内插和抽取技术实现高精度的延时处理算法.采用D倍内插,经过内插滤波器将输入信号的采样率提高到原来的D倍后,提高了采样精度.结合样点延迟时间进行内插后信号的D倍抽取操作,通过只对未被抽取信号才进行DSP运算的优化算法,使得DSP运行周期减少并达到实验要求.结果证明了该方法的可行性.     

多通道雷达数字接收机技术

针对多通道雷达数字接收机在现代战争中的应用需求,以四通道为例,提出了一种基于FPGA的数字接收机方案,采用多类滤波器级联技术,对系统进行硬件和软件设计,通过控制积分梳状(CIC)滤波器和半带(HB)滤波器,实现覆盖范围为2~8192倍的抽取滤波,最后对系统进行硬件调试及现场测试,并对测试结果进行分析,验证了方案的正确性和可行性。     

An unwindowed multichannel lattice filter with orthogonal channels

An unwindowed (or covariance) multichannel lattice filter for recursive least-squares estimation is derived. The channels are orthogonalized to eliminate the need for matrix inversion. The channel-orthogonalization process leads to forward-propagating and backward-propagating blocks in both the lattice filter and the model-parameter construction algorithm. These blocks are particularly suitable for array processing, as illustrated by arrays presented in the paper     

Generalized raised-cosine filters

Data transmission over bandlimited channels requires pulse shaping to eliminate or control intersymbol interference (ISI). Nyquist filters provide ISI-free transmission. Here we introduce a phase compensation technique to design Nyquist filters. Phase compensation can be applied to the square-root of any zero-phase bandlimited Nyquist filter with normalized excess bandwidth less than or equal to one. The resulting phase compensated square-root filter is also a Nyquist filter. In the case of a raised-cosine spectrum, the phase compensator has a simple piecewise-linear form. Such a technique is particularly useful to accommodate two different structures for the receiver, one with a filter matched to the transmitting filter and one without a matched filter. We also use the phase compensation technique to characterize a more general family of Nyquist filters which subsumes raised-cosine spectra. These generalized raised-cosine filters offer more flexibility in filter design. For instance, the rate of asymptotic decay of the filter impulse response may be increased, or the residual ISI, introduced by truncation of the impulse response, may be minimized. Design examples are provided to illustrate these choices     

Efficient recursive realizations of FIR filters

A class of efficient filter structures is proposed which uses a recursive realization of an FIR filter. The structures are in some sense a generalization of the frequency sampling structure, but they are more versatile and arise from a time-domain rather than frequency-domain argument. The new structure has a tap out of every block delay, and the length of the block delay is the length of each piecewise section of the time-domain approximation. The number of taps, filter coefficients, and the amount of arithmetic are proportional to the number of piecewise sections, not to the actual filter length or order. This filter is particularly efficient when a long-length filter can be approximated by a few piecewise sections, which is the case for many practical filters.This research was supported by NSF grant ECS 81-00453.     

A Compact Channel-Dropping Filter for Stripline and Microwave Integrated Circuits

A compact channel-dropping filter geometry is described that is particularly suited for stripline and microwave integrated circuits (MIC). The channel-dropping filter is comprised of a hairpin bandpass filter and a new bandstop filter, referred to herein as a folded-line bandstop filter. Design parameters for the latter filter for a range of bandwidths are tabulated. Analytical and experimental data for a trial channel-dropping filter are presented and discussed.     

Multilayer elliptic filter

A compact configuration for a four pole elliptic filter is described. This filter stacks quarter wave resonators in a novel manner that is particularly suitable for multilayer ceramic packaging. A design procedure is discussed. Simulations and measurements of a prototype filter are presented     

A Comb-Line Elliptic Filter

The design of a new type of elliptic filter, particularly suitable for narrow-band low-loss applications at VHF and UHF, is presented. The filter is derived from a lumped-element bandpass prototype by replacing the lumped inductors, which are normally the main contributory factors to the loss, by a comparatively low-loss distributed network. The latter consists of an n-wire digital line short-circuited at one end, the length of which is i/8 or less. An experimental elliptic filter of fifth order was constructed at 136.6 MHz with a pass bandwidth of 5 MHz, having 60-dB points at /spl plusmn/ 7 MHz from midband. The measured insertion loss of 1.1 dB is lower than that of a comparable lumped-element filter by a factor of at least 3,5:1. It compares favorably also with a comb-line filter, both in terms of loss and physical size.     

Band-Stop Filters For High-Power Applications

There are several advantages to the use of band-stop filters, rather than band-pass filters, in many systems. This is shown to be particularly true when signals at high-power levels must be transmitted or rejected. A formula has been derived which expresses the external Q of each resonator in a band-stop filter in terms of the element values of the normalized low-pass prototype and the parameters of the frequency transformation. The peak power capacity of iris-coupled waveguide cavity filters and TEM filters using capacitively coupled inductive stubs is then determined in terms of the external Q of the first resonator and the dimensions of the resonator. Experimental results given for a waveguide band-stop filter show good agreement with theory.     

An Analytic Approach to Random Phase Error and Its Impact on the Performance and Design of Arrayed-Waveguide Gratings

Random phase error due to fabrication process causes the filter response of arrayed-waveguide grating (AWG) to degrade, especially in terms of crosstalk. In the side-lobe region, which is critical to the channel crosstalk performance, each instantiation of the random phase error can yield a significantly different filter transmission than that of the average for that level of phase error. In this report, the statistical behavior of the AWG filter transmission in the side-lobe region is studied analytically. Both the distribution of random side-lobe level at a given wavelength and an upper bound of the outage probability for side-lobe maxima are given in a simple closed form. Accordingly, a crosstalk margin needs to be allocated to ensure a given fabrication yield and this is shown to depend on the fractional bandwidth of the AWG filter. For filter shapes that are close to Gaussian, this crosstalk margin can be 8 dB or more above the average crosstalk level, for small fractional bandwidth of about 1% and fabrication yields of 80% or higher. These relations should be useful to AWG designers particularly when the underlying fabrication process is susceptible to nonnegligible random phase errors.     

Vision-Based Target Tracking and Surveillance With Robust Set-Valued State Estimation

Tracking a target from a video stream (or a sequence of image frames) involves nonlinear measurements in Cartesian coordinates. However, the target dynamics, modeled in Cartesian coordinates, result in a linear system. We present a robust linear filter based on an analytical nonlinear to linear measurement conversion algorithm. Using ideas from robust control theory, a rigorous theoretical analysis is given which guarantees that the state estimation error for the filter is bounded, i.e., a measure against filter divergence is obtained. In fact, an ellipsoidal set-valued estimate is obtained which is guaranteed to contain the true target location with an arbitrarily high probability. The algorithm is particularly suited to visual surveillance and tracking applications involving targets moving on a plane.      

All-fiber spectral filters with nonperiodic bandpasscharacteristics and high extinction ratios in the wavelength range 0.8μm<λ<1.6 μm

A simple modification is proposed to the dual-core spectral filter design which extends the range of single wavelength transmission considerably, making the filter characteristic nonperiodic in the wavelength range 0.8 μm<λ<1.6 μm. Also, a spectral filter based on collinear triple-core fiber is proposed which provides a substantially lower sidelobe level than the dual-core filter. Transmission characteristics of the proposed filters are calculated and compared with those of the unmodified dual-core filter. The advantages of the proposed filters, particularly for application in wavelength division multiplexed systems are addressed