Canonical Filter Research Articles

Advanced Synthesis and Precise Design of Inline Fully Canonical Filters With High Selectivity

This article presents an advanced synthesis and novel designs of inline fully canonical filters with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> cascaded extracted pole (EP) sections, providing higher band-edge selectivity and suppressing interference from adjacent bands. The conventional fully canonical prototype is reconfigured to significantly facilitate filter design and fabrication. Moreover, to synthesize realizable circuits, a general node scaling method based on admittance matrix is implemented. All associated elements, especially the junction susceptances, can be scaled directly without the aid of inverters, avoiding intricate equivalent circuit transforms. Furthermore, the importance of precise control over the junction susceptances in filter design is discussed for the first time. Accurate junction susceptance enables the exact mapping from synthesized circuits to physical structures, preventing brute-force optimization. In particular, a new and compact EP structure in coaxial technique is proposed with flexibly tuned loading phase and accurately regulated junction susceptance. The good measurement results of a coaxial EP filter with four reflection zeros (RZs) and four transmission zeros (TZs) validate the effectiveness of the proposed theories. Also, this is the first reported coaxial fully canonical filter with only build-in capacitive couplings to avoid passive intermodulation (PIM).

Multi-Oriented Rotation-Equivariant Network for Object Detection on Remote Sensing Images

Object detection has attracted a lot of attention in the field of image automatic interpretation. Detectors based on convolution neural networks (CNNs) applied in natural scene images encode detection results with horizontal bounding boxes (HBBs), which can not accurately calibrate the position and shape of the arbitrary-orientation objects on remote sensing images (RSIs). To solve these issues, we propose an object detection framework named multi-oriented rotation-equivariant network (MORE-Net) in this letter. The MORE-Net consists of a multi-oriented rotating filter (MORF) and a multi-oriented ground objects detector. The MORF generates rotation-equivariant features by rotating canonical filter according to the predefined discrete directions. Equipped with MORF, the multi-oriented ground objects detector extracts rotation-invariant semantic representations for each decoded rotated region of interest (RRoI). We also propose area smooth (AS) L1 loss to impose tighter shape and position constraints to the RROIs. Extensive experiments and comprehensive evaluations on the large-scale DOTA dataset demonstrate the effectiveness of the proposed framework, which achieves a mean average precision (mAP) value of 81.27 on the DOTA-v1.0 dataset and a mAP value of 78.03 on the DOTA-v1.5 dataset.

Analytical Synthesis of Fully Canonical Cascaded-Doublet Prototype Filters

This letter presents an analytical procedure for the synthesis of a low-pass prototype filter constituted by the cascade of doublet blocks coupled by inverters or shunt reactances. The considered doublet is a second-order resonating structure composed of two resonant and two nonresonant nodes which allow the placement of two finite transmission zeros (TZs) in the frequency response. The considered configuration allows realizing compact fully canonical filters with the maximum number of TZs (the doublets are typically implemented through dual-mode cavities). The proposed procedure introduces for the first time an analytical procedure for synthesizing the low-pass prototype of the cascaded-block topology, overcoming the time consuming and inaccurate procedures currently based on numerical optimization. Two synthesis examples are presented to validate the novel synthesis procedure.

A General Synthesis Technique of Mixed-Topology Including Parallel-Connected Structures for Fully Canonical Ladder-Type Acoustic Filters

A synthesis method for mixed-topology fully canonical filters native oriented to acoustic wave (AW) technology is presented in this article. The method combines inline sections with parallel-connected structures to generate ladder-like networks. In regards to the classical ladder filters, this topology exhibits more flexibility in terms of transmission zeros (TZs) allocation above and below the passband by disassociating the TZs from the resonance frequency of the resonators. This degree of freedom can also be used as an advantage to accommodate the technology constraints of AW resonators. In addition to the general synthesis method, it is provided a set of considerations to be taken into account in designing a feasible filter for AW technology. To show the robustness and capabilities of the methodology proposed have been included two synthesis examples and experimental assessment by using an equivalent reactance Butterworth Van Dyke model for rapid prototyping.

Bandpass filters with N + 1 transmission zeros using λ/4 resonator and λ/2 resonator

AbstractThe present paper presents a novel method to cope with the difficulty that an N-order filter can introduce a maximum number of transmission zeros (TZs) – no more than N. The method is described by a coupling topology and realized by a regular fully canonical filter structure to increase one more TZ for an N-order filter without any filter size increase. Besides, the TZ shift property, as another advantage of the proposed topology, is investigated. The coupling topology consists of a group of fully canonical coupling nodes with the harmonic node (non-resonating node (NRN)) excited by a λ/4 resonator. Owing to the unique coupling topology and coupling node distribution as well as the inclusion of an NRN, N + 1 TZs are generated by an N-order bandpass filter, while all the TZs are close to the passband and far away from the harmonic passband resulting the highly selective passbands. For demonstration, two-, three-, and four-order filters are designed and measured with the experimental results that confirm the proposed technique.

A PCA–CCA network for RGB-D object recognition

Object recognition is one of the essential issues in computer vision and robotics. Recently, deep learning methods have achieved excellent performance in red-green-blue (RGB) object recognition. However, the introduction of depth information presents a new challenge: How can we exploit this RGB-D data to characterize an object more adequately? In this article, we propose a principal component analysis–canonical correlation analysis network for RGB-D object recognition. In this new method, two stages of cascaded filter layers are constructed and followed by binary hashing and block histograms. In the first layer, the network separately learns principal component analysis filters for RGB and depth. Then, in the second layer, canonical correlation analysis filters are learned jointly using the two modalities. In this way, the different characteristics of the RGB and depth modalities are considered by our network as well as the characteristics of the correlation between the two modalities. Experimental results on the most widely used RGB-D object data set show that the proposed method achieves an accuracy which is comparable to state-of-the-art methods. Moreover, our method has a simpler structure and is efficient even without graphics processing unit acceleration.

A Fully Canonical Bandpass Filter Design Using Microstrip Transversal Resonator Array Configuration

This paper proposes a new and simple microstrip bandpass filter structure for the design of a fully canonical transversal array filter. The filter is constructed by the parallel arrangement of microstrip even-and odd-mode half-wavelength resonators. In this filter, transmission zeros (TZs) are not produced by cross couplings used in conventional filter designs, but by an intrinsic negative coupling of the odd-mode resonators having open ends with respect to the even-mode resonators with shorted ends. Thus, the control of the resonant frequency and the external Q factor of each resonator makes it possible to form both a specified passband and TZs. As an example, a fully canonical bandpass filter with 2-GHz center frequency, 6% bandwidth, and four TZs is synthesized with a coupling-matrix optimization, and its structural parameters are designed. The designed filter achieves a rapid roll-off and low-loss passband response, which can be confirmed numerically and experimentally.

A spatial filter bank model of the Veselago–Pendry superlens

We present a canonical filter bank model that emulates the transmission of both propagating and evanescent electromagnetic fields through a Veselago–Pendry Superlens. The model consists of an array of coupled resonator pairs; each pair transmits one term of the spatial Fourier series of the object field. In addition to emulating the steady-state transfer function of the superlens, the model also approximates its dynamic response. Closed-form expressions for the values of the circuit elements are derived in terms of the geometry and the constitutive parameters of the lens at the frequency of operation in the steady state. Losses can be included by adding dissipative elements to the equivalent circuit. The concept of the Veselago–Pendry superlens as a filter in k-space provides physical insight into the resonant nature of image transmission and may suggest new ways to realize such a device.

Tight wavelet frames in low dimensions with canonical filters

This paper is to construct tight wavelet frame systems containing a set of canonical filters by applying the unitary extension principle of Ron and Shen (1997). A set of filters are canonical if the filters in this set are generated by flipping, adding a conjugation with a proper sign adjusting from one filter. The simplest way to construct wavelets of s-variables is to use the 2s−1 canonical filters generated by the refinement mask of a box spline. However, almost all wavelets (except Haar or the tensor product of Haar) defined by the canonical filters associated with box splines do not form a tight wavelet frame system. We consider how to build a filter bank by adding filters to a canonical filter set generated from the refinement mask of a box spline in low dimension, so that the wavelet system generated by this filter bank forms a tight frame system. We first prove that for a given low dimension box spline of s-variables, one needs at least additional 2s filters to be added to the canonical filters from the refinement mask (that leads to the total number of highpass filters in the filter bank to be 2s+1−1) to have a tight wavelet frame system. We then provide several methods with many interesting examples of constructing tight wavelet systems with the minimal number of framelets that contain canonical filters generated by the refinement masks of box splines. The supports of the resulting framelets are not bigger than that of the corresponding box spline whose refinement mask is used to generate the first 2s−1 canonical filters in the filter bank. In many of our examples, the tight frame filter bank has the double-canonical property, meaning it is generated by adding another set of canonical filters generated from a highpass filter to the canonical filters generated by the refinement mask to make a tight frame system.

Improving the sensitivity of sample clustering by leveraging gene co-expression networks in variable selection.

BackgroundMany variable selection techniques have been proposed for the clustering of gene expression data. While these methods tend to filter out irrelevant genes and identify informative genes that contribute to a clustering solution, they are based on criteria that do not consider the potential interactive influence among individual genes. Motivated by ensemble clustering, there is a strong interest in leveraging the structure of gene networks for gene selection, so that the relationship information between genes can be effectively utilized, while the selected genes are expected to preserve all the possible clustering structures in the data.ResultsWe present a new filter method that uses the gene connectivity in the gene co-expression network as the evaluation criteria for variable selection. The gene connectivity measures the importance of the genes in term of their expression similarity with others in the co-expression network. The hard threshold and soft threshold transformations are employed to construct the gene co-expression networks. Both simulation studies and real data analysis have shown that the network based on soft thresholding is more effective in selecting relevant variables and provides better clustering results compared to the hard thresholding transformation and two other canonical filter methods for variable selection. Furthermore, a new module analysis approach is proposed to reveal the higher order organization of the gene space, where the genes of a module share significant topological similarity and are associated with a consensus partition of the sample space. We demonstrate that the identified modules can lead to biologically meaningful sample partitions that might be missed by other methods.ConclusionsBy leveraging the structure of gene co-expression network, first we propose a variable selection method that selects individual genes with top connectivity. Both simulation studies and real data application have demonstrated that our method has better performance in terms of the reliability of the selected genes and sample clustering results. In addition, we propose a module recovery method that can help discover novel sample partitions that might be hidden when performing clustering analyses using all available genes. The source code of our program is available at http://nba.uth.tmc.edu/homepage/liu/netVar/.

Silicon Micromachined Canonical ${\hbox{E}}$-Plane and ${\hbox{H}}$-Plane Bandpass Filters at the Terahertz Band

In this letter, several bandpass filters operating in the WR-1.5 band (500 to 750 GHz) are presented. The deep reactive ion etching (DRIE) silicon micromachining process is used for the fabrication of the filters. Two canonical filter topologies based on E- and H-plane are implemented. The work presented here has two specific objectives: a) to get important fabrication process parameters, such as tolerances, vertical angles, surface roughness, and repeatability and b) to validate the proper working of the waveguide filters in the terahertz band. These filters do not have any tuning element. Experimental results show better than 10 dB return loss and approximately 1 and 2.5 dB insertion loss (for 6% fractional bandwidth) for the E- and H-plane topology, respectively. The obtained results are in agreement with fabrication tolerances of 2 μm and vertical angles deviations up to 3°.

Design and Diagnosis of Wideband Coupled-Resonator Bandpass Filters

A method for circuit elements parameter extraction of wideband filters from their electromagnetic (EM) simulated response is presented. A combline-like resonator for filter realization is introduced. Several possible electric and magnetic coupling structures are investigated. The equivalent lumped element circuit has electric and magnetic couplings modeled by capacitors and inductors instead of frequency-independent reactance or impedance inverters. The parameter extraction method of direct-coupled-resonator filters is developed and is applied to symmetric canonical filters. For filters with nonadjacent spurious couplings, optimization is used to extract the lumped element parameters to reproduce the EM simulated response. Excellent agreement between the EM simulation, lumped elements extracted responses and measurements is obtained. Filter synthesis is carried out by an optimization process. Several filter designs are presented to illustrate the design procedure and validate the circuit modeling. Measured results of an eight-pole filter verified the design process.

A New Approach to Canonical Dual-Mode Cavity Filter Design

This paper presents a new approach to the design of canonical dual-mode cavity filters within representation theory. Instead of the folded canonical coupling matrix, which results from using the resonances of the empty cavities as a basis, a representation based on the resonances of the cavities with all perturbations present is used. The resulting coupling matrix takes the form of a cul-de-sac configuration. For canonical dual-mode filters with a symmetric response, it allows direct design of the entire structure, except of the input cavity that requires optimization. A sixth-order filter with two symmetrically located transmission zeros is presented to document the validity of the approach. A sixth-order filter with two transmission zeros above the passband is also presented.

Inline and canonical stripline filters with ridge coupling sections for LTCC applications

Inline and canonical filters employing stripline resonators coupled by evanescent-mode ridge waveguide sections are presented. The proposed configurations are suitable choices for broadband chip fi...

Inline and canonical stripline filters with ridge coupling sections for LTCC applications

Inline and canonical filters employing stripline resonators coupled by evanescent-mode ridge waveguide sections are presented. The proposed configurations are suitable choices for broadband chip filters integrated in printed circuit boards, especially for those with constraints on size and loss. Low temperature co-fired ceramic (LTCC) technology can be used to manufacture such filters. To validate the idea, prototypes of Chebyshev and elliptic function filters are designed following a systematic procedure. Stepped impedance stripline resonators are used in the Chebyshev filter to improve the spurious response. A novel canonical structure is presented for achieving high selectivity elliptic function filter responses. Rigorous mode matching method is applied to analyze and optimize all the filters, whose results are validated by commercial software. © 2008 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2009.

Derivation of Second Order Canonic All-Pass Digital Filter Realizations

This paper deals with the derivation of two canonic all-pass digital filter realizations, first proposed by Mitra and Hirano. In contrast to their derivation, which uses a three-pair approach, our derivation is much simpler because we use a two-pair approach, in which only four, instead of nine parameters have to be chosen.

Synthesis of canonical topology microwave filters using Householder's transformations

The general even-mode coupling matrix, which contains all possible couplings of a microwave filter can be synthesised with well established filter synthesis technique. Presented is a simple but efficient method to annihilate unwanted couplings to realise canonical topology microwave filters using Householder's transformations. Furthermore, this method can be easily adopted into filter design software to synthesise canonical topology microwave filters.

A dual-passband filter of canonical structure for satellite applications

Due to the complex arrangement of frequency plans and spatial coverages in modern satellite communication systems, channels that are noncontiguous in frequency may be amplified by a single power amplifier and transmitted to the ground through one beam. In this letter, a dual-mode canonical filter with dual-passband is presented. The filter adopts dual-mode technique for mass and volume reduction. Canonical structure is adopted for maximum zero realization. To validate the design technique, a six-pole dual-mode dual-passband filter of canonical structure for Ka-band (30/20GHz) satellite transponder is realized. The measured frequency response of the filter shows good agreement with the computed one.

Coupling matrix of a 10th‐order dual‐mode asymmetric canonical filter

AbstractThis paper deals with a simple method to extract the coupling matrix of a 10th‐order dual‐mode asymmetric canonical filter. The coupling matrix of an asymmetric canonical filter is obtained by applying a plane rotation technique to the coupling matrix of a symmetric canonical filter. This paper gives a list of pivots and rotation angles to obtain the coupling matrix of an asymmetric canonical‐structure filter. © 2003 Wiley Periodicals, Inc. Microwave Opt Technol Lett 38: 64–68, 2003

Canonical Biquadratic All-pass and Notch Filters Employing Differential Difference Current Conveyor

Canonical Biquadratic All-pass and Notch Filters Employing Differential Difference Current Conveyor