Two‐dimensional recursive digital filters with nearly circular‐symmetric magnitude response and approximately linear phase

This paper presents a general structure using 1‐D and two‐dimensional (2‐D) recursive digital all‐pass filters (DAFs) for the design of 2‐D recursive circularly symmetric digital low‐pass filters (CS‐DLFs). The general structure is a cascade of two stages composed of all‐pass building blocks. The first stage is a parallel connection of a 2‐D recursive DAF with a symmetric half‐plane (SHP) support for its filter coefficients and a 2‐D pure delay block. The second stage composed of a parallel connection of a 1‐D recursive DAF and a 1‐D pure delay block is used for eliminating the unwanted pass‐band induced by the first stage. As a result, the design of a 2‐D CS‐DLF in either the least‐squares or the minimax sense can be formulated in a simple linear optimization problem in terms of the weighted‐phase response error for each DAF. Design results with nearly circularly symmetric magnitude response and approximately linear phase are also provided for illustration and comparison. Copyright © 2010 John Wiley & Sons, Ltd.     

Design of digital allpass‐based two‐channel quincunx QMF banks

This paper presents a novel structure for constructing two‐channel quincunx quadrature mirror filter (QQMF) banks. The analysis and synthesis filters of the QQMF bank are composed of two‐dimensional (2‐D) recursive digital allpass filters with symmetric half‐plane support regions. Using the 2‐D doubly complementary half‐band property possessed by the analysis and synthesis filters, we facilitate the design of the digital allpass‐based two‐channel QQMF banks. For finding the coefficients of the 2‐D recursive symmetric half‐plane digital allpass filters, the design problem is appropriately formulated to result in an optimization problem that can be solved by using a weighted least‐squares algorithm in the minimax (L_∞) optimal sense. The novelty of the proposed design method is that the designed 2‐D QQMF bank achieves perfect magnitude response and possesses satisfactory phase response without requiring extra phase equalizer. Simulation results are also provided for illustration and comparison. Copyright © 2012 John Wiley & Sons, Ltd.     

Accurate design‐oriented simulation‐driven modeling of miniaturized microwave structures

One of the important prerequisites for efficient design optimization of microwave structures is availability of fast yet reliable replacement models (surrogates) so that multiple evaluations of the structure at hand can be executed in reasonable timeframe. Direct utilization of full‐wave electromagnetic (EM) simulations for handling optimization‐related tasks is often prohibitive. A popular approach to construction of fast surrogates is data‐driven modeling. Unfortunately, it normally requires a large number of training samples, and it is virtually infeasible for structures that exhibit highly nonlinear responses (e.g. filters or couplers). In this work, a design‐oriented modeling technique is proposed where good accuracy is achieved by careful non‐uniform design space sampling that accounts for nonlinear relationship between the operating frequency of the structure and its geometry parameters, as well as carrying out the modeling process only for selected characteristic points of the structure responses (those that determine satisfaction/violation of given design specifications). Our approach is demonstrated using a miniaturized microstrip rat‐race coupler modeled in a wide range of geometry parameters and compared to conventional data‐driven modeling using kriging interpolation. Design optimization examples are also provided. Copyright © 2016 John Wiley & Sons, Ltd.     

Development of 3‐D equivalent‐circuit modelling with decoupled L‐ILU factorization in semiconductor‐device simulation

In this paper, we develop a three‐dimensional (3‐D) device simulator, which combines a simplified, decoupled Gummel‐like method equivalent‐circuit model (DM) with levelized incomplete LU (L‐ILU) factorization. These complementary techniques are successfully combined to yield an efficient and robust method for semiconductor‐device simulation. The memory requirements are reduced significantly compared to the conventionally used Newton‐like method. Furthermore, the complex voltage‐controlled current source (VCCS) is simplified as a nonlinear resistor. Hence, the programming and debugging for the nonlinear resistor model is much easier than that for the VCCS model. Further, we create a connection‐table to arrange the scattered non‐zero fill‐ins in sparse matrix to increase the efficiency of L‐ILU factorization. Low memory requirements may pave the way for the widespread application in 3‐D semiconductor‐device simulation. We use the body‐tied silicon‐on‐insulator MOSFET structure to illustrate the capability and the efficiency of the 3‐D DM equivalent‐circuit model with L‐ILU factorization. Copyright © 2007 John Wiley & Sons, Ltd.     

Optimal analog active filter design using craziness‐based particle swarm optimization algorithm

Because of the manufacturing constraints, the optimal selection of passive component values for the design of analog active filter is very critical. As the search on possible combinations in preferred values for capacitors and resistors is an exhaustive process, it has to be automated with high accuracy within short computation time. Evolutionary computation may be an attractive alternative for automatic selection of optimal discrete component values such as resistors and capacitors for analog active filter design. This paper presents an efficient evolutionary optimization approach for optimal analog filter design considering different topologies and manufacturing series by selecting their component values. The evolutionary optimization technique employed is craziness‐based particle swarm optimization (CRPSO). PSO is very simple in concept, easy to implement and computationally efficient algorithm with two main advantages: fast convergence and only a few control parameters. However, the performance of PSO depends on its control parameters and may be influenced by premature convergence and stagnation problem. To overcome these problems, the PSO algorithm has been modified to CRPSO and is used for the selection of optimal passive component values of fourth‐order Butterworth low‐pass analog active filter and second‐order state variable low‐pass filter, respectively. CRPSO performs the dual task of efficiently selecting the component values as well as minimizing the total design errors of low‐pass active filters. The component values of the filters are selected in such a way so that they become E12/E24/E96 series compatible. The simulation results prove that CRPSO efficiently minimizes the total design error with respect to previously used optimization techniques. Copyright © 2014 John Wiley & Sons, Ltd.     

A power‐area‐efficient, 3‐band, 2‐RX MIMO,TD‐LTE receiver with direct‐coupled ADC

In this work, a power‐area‐efficient, 3‐band, 2‐RX MIMO, and TD‐LTE (backward compatible with the HSPA+, HSUPA, HSDPA, and TD‐SCDMA) CMOS receiver is presented and implemented in 0.13‐μm CMOS technology. The continuous‐time delta‐sigma A/D converters (CT ?Σ ADCs) are directly coupled to the outputs of the transimpedance amplifiers, eliminating the need of analog anti‐aliasing filters between RX front‐end and ADCs in conventional structures. The strong adjacent channel interference without low‐pass filter attenuation is handled by proper gain control. A low‐power small‐area solution for excess loop delay compensation is implemented in the CT ?Σ ADC. At 20 MHz bandwidth, the CT ?Σ ADC achieves 66 dB dynamic range and 3.5 dB RX chip noise figure is measured. A maximum of 2.4 dB signal‐to‐noise ratio degradation is measured in all the adjacent channel selectivity (ACS) and blocking tests, demonstrating the effectiveness of the strategy against the low‐pass filter removal from the conventional architecture. The receiver dissipates a maximum of 171 mW at 2‐RX MIMO mode. To our best knowledge, it is the first research paper on the design of fully integrated commercial TD‐LTE receiver. Copyright © 2014 John Wiley & Sons, Ltd.     

Passivity–preserving interpolation‐based parameterized model order reduction of PEEC models based on scattered grids

The increasing operating frequencies and decreasing IC feature size call for 3‐D electromagnetic (EM) methods, such as the Partial Element Equivalent Circuit (PEEC) method, as necessary tools for the analysis and design of high‐speed systems. Very large systems of equations are often generated by 3‐D EM methods and model order reduction (MOR) techniques are commonly used to reduce such a high model complexity. A typical design process includes optimization and design space exploration, and hence requires multiple simulations for different design parameter values. Traditional MOR techniques perform model reduction only with respect to frequency and such design activities call for parameterized MOR (PMOR) methods that can reduce large systems of equations with respect to frequency and other design parameters of the circuit, such as geometrical layout or substrate characteristics. We present a novel PMOR technique applicable to the PEEC method that provides parametric reduced order models, stable and passive by construction, over a user defined design space. We treat the construction of parametric reduced order models on scattered design space grids. Pertinent numerical examples validate the proposed approach. Copyright © 2010 John Wiley & Sons, Ltd.     

0.5‐V fractional‐order companding filters

Novel configurations of fractional‐order filter topologies, realized through the employment of the concept of companding filtering, are introduced in this paper. As a first step, the design procedure is presented in a systematic algorithmic way, while in the next step, the basic building blocks of sinh‐domain and log‐domain integrators are presented. Because of the employment of metal–oxide–semiconductor (MOS) transistors operated in the subthreshold region, the derived filter structures offer the capability for operation in an ultra‐low‐voltage environment. In addition, because of the offered resistorless realizations, the proposed topologies are reconfigurable, in the sense that the order of the filter could be chosen through appropriate bias current sources. The performance of the derived fractional‐order filters has been evaluated through simulation and comparison results using the Analog Design Environment of the Cadence software and MOS transistor parameters provided by the Taiwan Semiconductor Manufacturing Company (TSMC) 180‐nm complementary MOS (CMOS) process. Copyright © 2014 John Wiley & Sons, Ltd.     

Low‐power,low‐noise,active‐RC band‐pass filters using a ‘lossy’ LP–BP transformation

A new and straightforward design procedure for simple canonical topologies of allpole, active‐RC, low‐selectivity band‐pass (BP) filters, with low sensitivity to component tolerances is presented. The procedure is primarily intended for discrete‐component, low‐power filter applications using just one amplifier for relatively high‐order filters. The design procedure starts out with an ‘optimized’ low‐pass (LP) prototype filter, yielding an ‘optimized’ BP filter, whereby the wealth of ‘optimized’ single‐amplifier LP filter designs can be exploited. Using a so‐called ‘lossy’ LP–BP transformation, closed‐form design equations for the design of second‐ to eighth‐order, single‐amplifier BP filters are presented. The low sensitivity, low power consumption, and low noise features of the resulting circuits, as well as the influence of the finite gain‐bandwidth product and component spread, are demonstrated for the case of a fourth‐order filter example. The optimized single‐opamp fourth‐order filter is compared with other designs, such as the cascade of optimized Biquads. Using PSpice with a TL081 opamp model, the filter performance is simulated and the results compared and verified with measurements of a discrete‐component breadboard filter using 1% resistors, 1% capacitors, and a TL081 opamp. Copyright © 2011 John Wiley & Sons, Ltd.     

Modelling of lossy curved surfaces in the 3‐D frequency‐domain finite‐difference methods

A conformal first‐order or Leontovic surface‐impedance boundary condition (SIBC) for the modelling of fully three‐dimensional (3‐D) lossy curved surfaces in a Cartesian grid is presented for the frequency‐domain finite‐difference (FD) methods. The impedance boundary condition is applied to auxiliary tangential electric and magnetic field components defined at the curved surface. The auxiliary components are subsequently eliminated from the formulation resulting in a modification of the local permeability value at boundary cells, allowing the curved 3‐D surface to be described in terms of Cartesian grid components. The proposed formulation can be applied to model skin‐effect loss in time‐harmonic driven problems. In addition, the impedance matrix can be used as a post‐processor for the eigenmode solver to calculate the wall loss. The validity of the proposed model is evaluated by investigating the quality factors of cylindrical and spherical cavity resonators. The results are compared with analytic solutions and numerical reference data calculated with the commercial software package CST Microwave Studio™ (MWS). The convergence rate of the results is shown to be of second‐order for smooth curved metal surfaces. The overall accuracy of the approach is comparable to that of CST MWS™. Copyright © 2006 John Wiley & Sons, Ltd.     

Optimal design of third‐order microstrip bandpass filters by direct synthesis technique (DST)

To design microstrip filters is not easy for the sake of their distributed‐element effect. Undoubtedly, to understand their physical mechanism is very important to their design. In this paper, one effective approach to design some third‐order microstrip bandpass filters with each of 2 transmission zeros at each side of the passband is discussed. Lumped‐element equivalent circuits are used to represent these microstrip filters. Then, these lumped‐element equivalent circuits can be synthesized by direct synthesis technique we recently proposed, so that it is likely to calculate initial structural parameters of these microstrip filters and then facilitate their design. Verified by the measured results of the filter designed through the approach in this paper, the performance of the filters is close to ideal frequency responses. Furthermore, another third‐order microstrip bandpass filter is presented, in which open‐circuited stubs at input/output ports are introduced to suppress one specified harmonic to improve out‐of‐band attenuation.     

A simulation study of high‐order CMOS hyperbolic‐sine filters

This paper advances the field of externally linear–internally nonlinear (ELIN) filters by introducing a synthesis method that enables the design of high‐order class‐AB sinh filters by means of complementary metal–oxide semiconductor (CMOS) weak‐inversion sinh integrators comprising only one type of devices in their translinear loops. The proposed transistor‐level synthesis approach is demonstrated through the examples of (1) a biquadratic and (2) a fifth‐order filter, and their simulated performance is studied. The biquadratic filter achieves a dynamic range of 94 dB and has a tunable quality factor Q up to the value of 8, whereas its natural frequency can be tuned for four orders of magnitude. Its static power consumption amounts to 6.2 μW for Q = 1 and f_o = 2 kHz. The fifth‐order Chebyshev sinh CMOS filter with a cut‐off frequency of 100 Hz, a pass band ripple of 1 dB, and a power consumption of ~300 nW is compared head‐to‐head with its pseudo‐differential class‐AB CMOS log domain counterpart. The sinh filter achieves similar or better signal‐to‐noise ratio (SNR) and signal‐to‐noise‐plus‐distortion ratio (SNDR) performances with half the capacitor area but at the expense of higher power consumption from the same power supply level. All three presented filter topologies are novel. Cadence design framework simulations have been performed using the commercially available 0.35 µm AMS (austriamicrosystems) process parameters. Copyright © 2013 John Wiley & Sons, Ltd.     

Even‐order passive filters: Pascal versus Chebyshev

The recently reported Pascal approximation with non‐equiripple magnitude response leads to transfer functions of order equal or comparable to that of the Chebyshev approximation, offering an alternative to the equiripple Chebyshev approximation. Both approximations can be used in passive filter design and have similar design limitations when the order turns out to be even. In this paper, these design issues are thoroughly addressed, and the exact conditions are set under which even‐order passive Chebyshev and Pascal filters cannot be directly synthesised and either the order has to be increased to the next odd integer value or the modified filter must be designed. Pascal approximation is used for the first time for the design of doubly resistively terminated passive LC filters, and it is shown that the even‐order design issue is much less restricting making Pascal superior to Chebyshev filters in that even‐order filters can be directly designed to meet specifications that cannot be met by Chebyshev filters of the same even order. The theoretical results are confirmed through several design examples. Copyright © 2012 John Wiley & Sons, Ltd.     

Ladder tuning‐block partitioned filters for simplified tuning

In this paper, we present a procedure for the design of low‐sensitivity, active‐RC filters that permits efficient functional tuning during the manufacturing process. Filters with finite zeros, such as elliptic (Chebyshev–Cauer) low‐pass filters are primarily considered, although the method can be applied to the design of other filters, e.g. allpole filters, as well. We show how to partition a given ladder filter into two parts. The first is a ladder filter of reduced order compared to the original; the second is a second‐ or third‐order active‐RC filter section, the ‘tuning block’, which, alone is used to tune the overall filter. The ladder, the components of which are fixed, provides most of the selectivity, while the cascaded tuning block determines the band‐edge characteristics and can be tuned relatively easily. A detailed design procedure for the filter partitioning is given. By obtaining a doubly terminated ladder filter, which is cascaded with a tuning block, both the inherent low sensitivity of the ladder and the tunability of the tuning block, are maintained. A Monte Carlo analysis of the partitioned filter demonstrates that the low sensitivity with respect to component tolerances, achievable by maintaining a doubly terminated ladder structure for the larger partitioned part of the filter, is preserved. Copyright © 2012 John Wiley & Sons, Ltd.     

Explicit design formulas for current‐mode leap‐frog OTA‐C filters and 300 MHz CMOS seventh‐order linear phase filter

The leap‐frog (LF) configuration is an important structure in analogue filter design. Voltage‐mode LF OTA‐C filters have recently been studied in the literature; however, general explicit formulas do not exist for current‐mode LF OTA‐C filters and there is also need for current‐mode LF‐based OTA‐C structures for realization of arbitrary transmission zeros. Three current‐mode OTA‐C structures are presented, including the basic LF structure and LF filters with an input distributor or an output summer. They can realize all‐pole characteristics and functions with arbitrary transmission zeros. Explicit design formulas are derived directly from these structures for the synthesis of, respectively, all‐pole and arbitrary zero filter characteristics of up to the sixth order. The filter structures are regular and the design formulas are straightforward to use. As an illustrative example, a 300 MHz seventh‐order linear phase low‐pass filter with zeros is presented. The filter is implemented using a fully differential linear operational transconductance amplifier (OTA) based on a source degeneration topology. Simulations in a standard TSMC 0.18µm CMOS process with 2.5 V power supply have shown that the cutoff frequency of the filter ranges from 260 to 320 MHz, group delay ripple is about 4.5% over the whole tuning range, noise of the filter is 420nA/√Hz, dynamic range is 66 dB and power consumption is 200 mW. Copyright © 2008 John Wiley & Sons, Ltd.     

A design methodology for power‐efficient reconfigurable SC ΔΣ modulators

This paper presents a methodology to design reconfigurable switched‐capacitor delta‐sigma modulators (ΔΣMs) capable of keeping their corresponding power efficiency figures constant and optimal for a set of resolutions and signal bandwidths. This method is especially suitable for low‐bandwidth, medium‐to‐high‐resolution specifications, which are common in biomedical application range. The presented methodology is based on an analytic model of all different contributions to the power dissipation of the ΔΣM. In particular, a novel way to predict the static power dissipated by integrators based on class A and class AB operational transconductance amplifier is presented. The power‐optimal solution is found in terms of filter order, quantizer resolution, oversampling ratio, and capacitor dimensions for a targeted resolution and bandwidth. As the size of the sampling capacitors is crucial to determine power consumption, three approaches to achieve reconfigurability are compared: sizing the sampling capacitors to achieve the highest resolution and keep them constant, change only the first sampling capacitor according to the targeted resolution, or program all sampling capacitors to the required resolution. The second approach results in the best trade‐off between power efficiency and simplicity. A reconfigurable ΔΣM for biomedical applications is designed at transistor level in a 0.18‐µm complementary metal–oxide–semiconductor process following the methodology discussed. A comparison between the power estimated by the proposed analytic model and the transistor implementation shows a maximum difference of 17%, validating thus the proposed approach. Copyright © 2014 John Wiley & Sons, Ltd.     

Computer‐aided design and simulation of optical microring resonators

This paper contains results of the three‐dimensional full‐vectorial finite element (3D FEM) frequency‐domain simulations of microring silicon resonators intended for the add/drop multiplexing, switching, and sensor applications. The second‐order edge elements are used, and the computational domain is closed using the standard perfectly matched layer. The calculations are performed for add‐drop and all‐pass configurations in a wide spectral range covering the first and the second telecommunication windows. Although computationally very demanding, the three‐dimensional full‐vectorial finite element offers designers a feasible and efficient way to design and simulate complex integrated optical components containing microring resonators, especially in the case when the radii of rings are too small to approximate that theories were valid. Copyright © 2013 John Wiley & Sons, Ltd.     

Monotonic,critical monotonic,and nearly monotonic low‐pass filters designed by using the parity relation for Jacobi polynomials

A new class of continuous‐time low‐pass filter using a set of Jacobi polynomials, with all transmission zeros at infinity, is described. The Jacobi polynomial has been adapted by using the parity relation for Jacobi polynomials in order to be used as a filter approximating function. The resulting class of polynomials is referred to as a pseudo Jacobi polynomials, because they are not orthogonal. The obtained magnitude response of these filters is more general than the magnitude response of the classical ultraspherical filter, because of one additional degree of freedom available in pseudo Jacobi polynomials. This additional parameter may be used to obtain a magnitude response having either smaller passband ripples or sharper cutoff slope. Monotonic, critical monotonic, or nearly monotonic passband filter approximating functions can be also generated. It is shown that proposed pseudo Jacobi polynomial filter approximation also includes the Chebyshev filter of the first kind, the Chebyshev filter of the second kind, the Legendre filter, and many transitional filter approximations, as its special cases. Several examples are presented, and detailed formulas including the practical suggestions for their efficient implementation are also provided. The proposed nearly monotonic filter is compared with the least‐square‐monotonic filters, designed as critical monotonic, in details. The advantages of the new filters are discussed. Copyright © 2017 John Wiley & Sons, Ltd.     

Constructing numerically stable Kalman filter‐based algorithms for gradient‐based adaptive filtering

This paper addresses the numerical aspects of adaptive filtering (AF) techniques for simultaneous state and parameters estimation arising in the design of dynamic positioning systems in many areas of research. The AF schemes consist of a recursive optimization procedure to identify the uncertain system parameters by minimizing an appropriate defined performance index and the application of the Kalman filter (KF) for dynamic positioning purpose. The use of gradient‐based optimization methods in the AF computational schemes yields to a set of the filter sensitivity equations and a set of matrix Riccati‐type sensitivity equations. The filter sensitivities evaluation is usually carried out by the conventional KF, which is known to be numerically unstable, and its derivatives with respect to unknown system parameters. Recently, a novel square‐root approach for the gradient‐based AF by the method of the maximum likelihood has been proposed. In this paper, we show that various square‐root AF schemes can be derived from only two main theoretical results. This elegant and simple computational technique replaces the standard methodology based on direct differentiation of the conventional KF equations (with their inherent numerical instability) by advanced square‐root filters (and its derivatives as well). As a result, it improves the robustness of the computations against round off errors and leads to accurate variants of the gradient‐based AFs. Additionally, such methods are ideal for simultaneous state estimation and parameter identification because all values are computed in parallel. The numerical experiments are given. Copyright © 2015 John Wiley & Sons, Ltd.