Time filter

Source Type

Cambridge, Canada

Yu M.,COM DEV Ltd. | Wang Y.,University of Ontario Institute of Technology
IEEE Microwave Magazine | Year: 2011

Several optimization techniques applied in different steps of the synthesis of filters and multiplexers are presented. A filter that is realized with resonators having low quality factors (Q) is less selective and exhibits more rounded band-edge performance compared to a filter with high Q resonators. An adaptive predistortion method applied adaptive correction factors to the transmission poles of the filter function polynomial leading to improvement in insertion loss and group delay equalization and permits a filter that is implemented in lower Q technology to emulate the performance of a higher Q counterpart. The adaptive predistortion can be used to shape the input multiplexer (IMUX) filter to complement and hence compensate the output multiplexer (OMUX) passband response, without compromising the channel performance constraints, especially isolation. Filters can be modeled using either distributed circuit models or by incorporating dispersion characteristics into circuit models such as the coupling matrix. Source

Kabir H.,Carleton University | Wang Y.,University of Ontario Institute of Technology | Yu M.,COM DEV Ltd. | Zhang Q.-J.,Carleton University
IEEE Transactions on Microwave Theory and Techniques | Year: 2010

Neural networks are useful for developing fast and accurate parametric model of electromagnetic (EM) structures. However, existing neural-network techniques are not suitable for developing models that have many input variables because data generation and model training become too expensive. In this paper, we propose an efficient neural-network method for EM behavior modeling of microwave filters that have many input variables. The decomposition approach is used to simplify the overall high-dimensional neural-network modeling problem into a set of low-dimensional sub-neural-network problems. By incorporating the knowledge of filter decomposition with neural-network decomposition, we formulate a set of neural-network submodels to learn filter subproblems. A new method to combine the submodels with a filter empirical/equivalent model is developed. An additional neural-network mapping model is formulated with the neural-network submodels and empirical/equivalent model to produce the final overall filter model. An H-plane waveguide filter model and a side-coupled circular waveguide dual-mode filter model are developed using the proposed method. The result shows that with a limited amount of data, the proposed method can produce a much more accurate high-dimensional model compared to the conventional neural-network method and the resulting model is much faster than an EM model. © 2006 IEEE. Source

Wang Y.,University of Ontario Institute of Technology | Yu M.,COM DEV Ltd.
IEEE MTT-S International Microwave Symposium Digest | Year: 2011

In this paper, the performance improvements and limitations of the enhanced microwave multiplexer are thoroughly studied for applications in communications satellites considering in-band flatness of gain and group delay, out-of-band rejection, temperature stability and insertion loss. The overall performance of the enhanced multiplexer with Mth order channel filters is similar to conventional design with M+1th order channel filters. It is shown that the difference in the coefficients of thermal expansion of the materials for the channel filters and the manifold has an effect on the multiplexer temperature stability. Impact of channel filter resonator Q factor and fractional bandwidth on the passband insertion loss is also demonstrated using different designs. © 2011 IEEE. Source

Langille J.A.,University of New Brunswick | Ward W.E.,University of New Brunswick | Scott A.,COM DEV Ltd. | Arsenault D.L.,University of New Brunswick
Applied Optics | Year: 2013

An implementation of the field widened Michelson concept has been applied to obtain high resolution two-dimensional (2D) images of low velocity (<50 m/s) Doppler wind fields in the lab. Procedures and techniques have been developed that allow Doppler wind and irradiance measurements to be determined on a bin by bin basis with an accuracy of less than 2.5 m/s from CCD images over the observed field of view. The interferometer scanning mirror position is controlled to subangstrom precision with subnanometer repeatability using the multi-application low-voltage piezoelectric instrument control electronics developed by COM DEV Ltd.; it is the first implementation of this system as a phase stepping Michelson. In this paper the calibration and characterization of the Doppler imaging system is described and the planned implementation of this new technique for imaging 2D wind and irradiance fields using the earth's airglow is introduced. Observations of Doppler winds produced by a rotating wheel are reported and shown to be of sufficient precision for buoyancy wave observations in airglow in the mesopause region of the terrestrial atmosphere. © 2013 Optical Society of America. Source

Kabir H.,Carleton University | Zhang L.,Freescale Semiconductor | Yu M.,COM DEV Ltd. | Aaen P.H.,Freescale Semiconductor | And 2 more authors.
IEEE Microwave Magazine | Year: 2010

An overview of neural network-based modeling techniques and their applications in microwave modeling and design is presented. The neural network represent RF/microwave components with the help of training data that are pairs of model input-output (IO) data generated from detailed microwave simulation or measurement. Neural networks have significant advantages over other techniques for multidimensional function approximation as they permit a compact representation of a multidimensional function, requiring minimal storage of coefficients and being very efficient to evaluate. The neural network can produce a parametric model by exploiting existing microwave knowledge in the form of empirical/analytical/equivalent model during neural network development. Neural network maps an existing model to match a new device with a technique called Neuro-Space Mapping (Neuro-SM). Source

Discover hidden collaborations