Magnetostatic wave convolvers

The present paper reviews recent theoretical results, and reports initial experimental results, on the convolution of contra-propagating magnetostatic forward volume waves (MSFVWs), in the form of cw signals or time-limited cw pulses, in an epitaxial yttrium iron garnet (YIG) film. Computations of the convolver bilinearity factorF _int indicate an efficient convolution process over a wide bandwidth, with values ofF _int that are of the same order as, or better than, the reported experimental results for MSW convolution in a YIG cylindrical or plate geometry. The values of F_int determined experimentally are in excellent agreement with theory. These results are of interest to microwave system developers particularly if bandwidths of 1 GHz or larger can be realized in practice. A limiting feature of magnetostatic wave (MSW) convolvers is that the maximum delay time of a delay line that is realizable without excessive insertion loss is in the order of 0.5s. The advantage of MSW convolvers, of course, lies in their ability to perform signal processing directly at microwave frequencies, and in applications such as electronic warfare the advantageously large bandwidths would mitigate the limitations in delay time.This work was supported in part by a contract from the AIL Division of the Eaton Corporation.     

Design criteria for surface-acoustic-wave convolvers

The effect of acoustic dispersion and attenuation on the performance of surface acoustic wave convolvers is shown to be equivalent to an ideal convolver in cascade with a linear time-invariant filter. Using this result calculations of the implementation loss due to acoustic dispersion and attenuation, transducer bandpass response, and input/output matching networks are given showing the effects of phase and amplitude responses which are representative of typical convolver designs.     

Ambiguity-function generator using acoustic surface-wave convolvers

It is shown that ambisuity function of a time-limited signal can be obtained using ultrasonic convolvers. Experimental results are presented using a separate media space charge coupled acoustic surface-wave convolver.     

Spatial uniformity measurement of SAW convolvers

For correlation of coded waveforms, the nonlinear interaction process in a SAW convolver should ideally give an amplitude and phase independent of position. A new experimental method of measuring this spatial uniformity used CW test waveforms and gave a resolution of 25 ns?a considerably higher resolution than previously attained.     

A polyphase architecture for serial-input convolvers

It is shown that a cascade of carry-free serial-parallel multipliers, fed by bit-serial samples with no separation between successive samples (i.e., with maximum sampling rate for a given bit rate), produces one convolution value everyp withp>1. This circuit is called aphase-convolver, and it is shown that a full convolver can be obtained by associatingp phase convolvers (polyphase convolver). Under some assumptions, a polyphase convolver can be implemented as a stack of bit-slices. It appears possible to program, by means of variables stored in a register, a slices-pack in order to obtain a convolver for a prescribed number of convolution terms and of weights and samples lengths. It is also possible to by-pass a faulty slice (or a group of slices) by means of variables stored in a second register.     

Three-channel variable filter-bank for digital hearing aids

This study proposes a three-channel (3-channel) variable filter-bank (VFB) that consists of variable lowpass, variable bandpass and variable highpass digital filters. The three variable digital filters are obtained from a normalised analog prototype Chebyshev type-I lowpass filter using analog frequency transformations along with a modified bilinear transformation. Since both the magnitudes (gains) and band edge frequencies of the three variable digital filters are independently adjustable, the 3-channel VFB is considerably flexible, and can be successfully applied for compensating various hearing loss patterns in digital hearing aids. Various audiograms have been used to verify that high-accuracy fittings can be achieved with low order variable filters. Moreover, the authors reveal and theoretically prove the numerator coefficient-symmetries of the variable lowpass, variable bandpass and variable highpass filters, and show that each variable filter requires only one multiplication for its numerator filtering operations, so the total number of multiplications can be significantly reduced. More specifically, only 11 multiplications and 14 additions are required in the whole 3-channel VFB. Therefore the 3-channel VFB has extremely simple structure and high tuning flexibility for hearing aids.     

General analysis of bilinear SAW convolvers

A general analysis for bilinear devices is derived, showing that such devices can be characterised by a two-dimensional frequency response or by a real two-dimensional impulse response. The forms of these functions for an ideal SAW convolver are given. It is shown that in general the output signal/noise ratio of a convolver depends on the timing of the input signal.     

Mismatch losses in multiply tapped SAW convolvers

An analysis procedure is given for calculating the output mismatch loss in a multiply tapped SAW convolver. The analysis takes into account the lossy transmission-line properties of convolvers and includes realistic values for the parasitics due to bonding pads and wires. Results illustrating trade-offs between bandwidth and insertion loss for a convolver with six taps are given, which provide guidelines for the design of broadband output matching networks.     

Narrow-aperture chirp transducers for SAW convolvers

Narrow-aperture SAW beams can be generated efficiently using chirp transducers, thus avoiding the need for beam compression in convolvers. The transducers can be designed such as to compensate for phase errors due to dispersive SAW propagation. Experimental confirmation was obtained using convolvers with 120 MHz bandwidth and 16 ?s interaction length, and a bilinearity factor of ?65 dBm was achieved.     

Pipelines logical convolvers for binary picture processing

Pipelines logical convolvers (PLCs) are well suited to document facsimile processing as they can simply be inserted into the video data stream of scanners, etc. Their introduction has no effect on throughput and they require only a partial picture memory. A method is described for designing and implementing PLCs, using a structured approach that encourages experimentation and produces a cost efficient circuit. An example PLC is presented which performs image smoothing.<>     

SAW convolvers using the transverse-horizontal bilinear field

Conventional SAW convolvers rely on the transverse-vertical polarisation field induced by crystal nonlinearity in the presence of two contraflow input waves. The letter presents results for single-track SAW convolvers fabricated on LiNbO3 using the transverse-horizontal (in-plane) electric polarisation. The new electrode arrangement involved reduces the possibility of unwanted 'fold-over convolution from fabrication defects and increases design flexibility.     

Orthonormal and biorthonormal filter banks as convolvers, andconvolutional coding gain

Convolution theorems for filter bank transformers are introduced. Both uniform and nonuniform decimation ratios are considered, and orthonormal as well as biorthonormal cases are addressed. All the theorems are such that the original convolution reduces to a sum of shorter, decoupled convolutions in the subbands. That is, there is no need to have cross convolution between subbands. For the orthonormal case, expressions for optimal bit allocation and the optimized coding gain are derived. The contribution to coding gain comes partly from the nonuniformity of the signal spectrum and partly from nonuniformity of the filter spectrum. With one of the convolved sequences taken to be the unit pulse function,,e coding gain expressions reduce to those for traditional subband and transform coding. The filter-bank convolver has about the same computational complexity as a traditional convolver, if the analysis bank has small complexity compared to the convolution itself     

One- and Two-Stage Tunable Receivers

In this paper, we propose and assess a CFAR detector that can adjust its “directivity” through a real scalar parameter. It relies on the usual assumption that a set of homogeneous training data is available and encompasses as special cases the well-known Kelly's GLRT and the recently introduced W-ABORT detector. More important, it can be tuned in order to control the level to which sidelobe signals are rejected. Such functionality is particularly important to contain the number of false alarms in presence of mismatched signals. We also consider a parametric detector which resorts to a diagonally loaded sample covariance matrix commonly adopted to take advantage of the presence of strong interferers. The performance assessment of such detector has shown that it can significantly outperform Kelly's GLRT in terms of prediction probabilities for matched signals and in terms of selectivity, but it is not strictly CFAR. We also propose to use the CFAR parametric detector as second stage of a two-stage tunable detector and show that such a two-stage detector can outperform already proposed tunable receivers in terms of selectivity. The analysis of the detectors is conducted assuming a homogeneous Gaussian environment; with reference to this scenario and to the CFAR detectors we derive analytical expressions for the probability of false alarm and the probability of detection for both matched and mismatched signals.      

High frequency ZnO-SiO2-Si surface acoustic wave convolvers

The separate comb grating transducer configuration is used in the fabrication of high frequency convolvers in the metal-ZnO-SiO2-Si surface acoustic wave device structure. A Rayleigh convolver with an operating frequency of 285 MHz and a Sezawa convolver with an operating frequency of 355 MHz are reported.     

CAD of precise SAW convolvers with high relative bandwidth

A CAD procedure for optimisation of SAW convolvers with narrow-aperture focusing chirp transducers including fully planar matching networks is demonstrated. A new finger arrangement to suppress internal reflections in broadband chirp transducers has been devised. As an example, 160 MHz bandwidth centred at 350MHz with 17.5?s delay line length was realised with an efficiency of ?63dBm, flat to ±0.8dB and phase-linear to ±6°.     

Reconfigurable pipelined 2-D convolvers for fast digital signalprocessing

In order to make software applications simpler to write and easier to maintain, a software digital signal-processing library that performs essential signal- and image-processing functions is an important part of every digital signal processor (DSP) developer's toolset. In general, such a library provides high-level interface and mechanisms, therefore, developers only need to know how to use algorithms, not the details of how they work. Complex signal transformations then become function calls, e.g., C-callable functions. Considering the two-dimensional (2-D) convolver function as an example of great significance for DSP's, this paper proposes to replace this software function by an emulation on a field-programmable gate array (FPGA) initially configured by software programming. Therefore, the exploration of the 2-D convolver's design space will provide guidelines for the development of a library of DSP-oriented hardware configurations intended to significantly speed up the performance of general DSP processors. Based on the specific convolver, and considering operators supported in the library as hardware accelerators, a series of tradeoffs for efficiently exploiting the bandwidth between the general-purpose DSP and accelerators are proposed. In terms of implementation, this paper explores the performance and architectural tradeoffs involved in the design of an FPGA-based 2-D convolution coprocessor for the TMS320C40 DSP microprocessor available from Texas Instruments Incorporated. However, the proposed concept is not limited to a particular processor     

Area-efficient 2-D shift-variant convolvers for FPGA-based digital image processing

Two-dimensional convolutions are local by nature; hence every pixel in the output image is computed using surrounding information, i.e., a moving window of pixels. Although the operation is simple, the hardware is conditioned by the fact that due to bandwidth efficiency full raster rows must be read from the external memory, and that a row-major image scan should be performed to support shift-variant convolutions. When extending the architectures developed in prior-art to support shift-variant convolutions, we realize that they require large amounts of on-chip memory. While this fact may not have a large cost increase in ASIC implementations, it makes field-programmable gate arrays (FPGA) implementations expensive or not feasible. In this paper, we propose several novel FPGA-efficient architectures for generating a moving window over a row-wise print path. Because the proposed concepts have different throughput and resource utilization, we provide a criteria to choose the optimum one for any design point.     

利用基于滤波器组的重叠变换抑制扩频通信系统中的时变干扰

本文提出了基于滤波器的重叠变换方法,并对滤波器组结构进行了一些改进,使之在提供频率信息的同时保留了信号的时域信息,因此物品处理时变干扰,仿真结果证明,采用基于滤波器的重叠变换并结合本文所介绍的阈值处理算法可以有效地抑制扩频通信系统中的时变干扰。     

Controlled One- and Multidimensional Modulations Using Chaotic Maps

Recently, several approaches for communications using chaos have been presented, often showing less than acceptable performance. In this paper, a short introduction to the topic is given, and it is shown that such methods can be efficient—if the information production related to the chaos in the transmitter is controlled and used for the payload of the communication. The influence of minimum distance aspects, a previously ignored key point, to the design of appropriate controlled schemes are studied. In terms of applicable theory, a link between schemes based on one-dimensional and multidimensional chaotic systems (as presented in this paper) and convolutional codes will be established. In order to support our argument, we give simulation results for some a priori examples and results of search procedure for more powerful modulations.