By Topic

Circuits and Systems, IEEE Transactions on

Issue 8 • Date August 1979

Filter Results

Displaying Results 1 - 9 of 9
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • Filters with variable order and virtually independent time delay

    Page(s): 657 - 659
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (376 KB)  

    An all-pole filter and a filter with a rational transfer function are considered, both of which are generalizations of the Bessel filter. It is shown that by properly selecting the generalizing parameters, the time delay in each case is, for low frequencies, very nearly independent of the order of the filter. It is possible, also, to select the parameters so that a nearly constant time delay is maintained for low frequencies while the filter order is varied to obtain a number of amplitude responses. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A class of multivariable positive real functions realizable by resistively terminated lossless ladder networks

    Page(s): 659 - 662
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (480 KB)  

    Necessary and sufficient conditions are obtained for the realizability of multivariable positive real functions which are of the first degree in all variables except one by resistively terminated ladder networks with all of their transmission zeros at the origin or at infinity. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Simple series for phase angle, delay, and group delay

    Page(s): 662 - 663
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    Simple power series in \omega for the phase angle, delay, and group delay due to a conjugate pair of complex poles or zeros are presented, and an example of their application to the design of phase equalizers is given. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The realizability boundary and the realization of nonsingular fourth-order conductance matrices by 6-terminal networks

    Page(s): 642 - 656
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1560 KB)  

    A new approach to the realization of n th-order conductance matrices by (n +p) -terminal, n -port networks is described. Results due to Cederbaum and Foster are employed. Cederbaum has given an example of a paramount matrix that is not realizable as either a resistance matrix conductance matrix. Foster has shown that any dominant conductance matrix is realizable. A conductance matrix with any fixed set of off-diagonal elements can be made dominant, by suitably increasing the diagonal elements. Thus we conclude that some sets of diagonal elements will permit a paramount conductance matrix to be realized, while others will not. Therefore, given a conductance matrix with a fixed set of off-diagonal elements, there exists a continuum of n -tuples of diagonal elements which are on the border line of realizability. We define this continuum as the realizability boundary. Networks which realize conductance matrices on the realizability boundary are called minimal networks and are characterized by not having shunt conductances at the ports. Next we apply these principles to the solution of the (n+2) -terminal 4-port case. Using a procedure derived from Guillemin, we find and catalog all possible minimal networks. For any given situation, a maximum of 22 networks is required to be tested. Synthesis procedures and examples are given. In addition, we show that certain ports of some minimal networks have constant driving-point conductances. We illustrate how to predict this effect, and suggest a practical application of this result. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Inductance, capacitance networks terminated in resistance

    Page(s): 603 - 641
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4376 KB)  

    This paper investigates the properties of the driving point impedance of general, passive, lossless two-ports without transformers which are terminated in resistance (LC-R networks). The culmination of the work consists of synthesis procedures in two areas. 1) A wide class of impedances Z(p) = [A (p) + B(p)]/[C(p) + D(p)] is characterized in terms of the locations of the zeros of (AC- BD) , where A,C and B,D are even and odd polynomials, respectively. Then a method of realization as an LC-R ladder network is given for members of this class. 2) The even part synthesis problem for an arbitrary, general impedance is solved. The realization takes the forms of a series connection as well as a parallel connection of at most four LC-R ladder networks. A number of examples give concrete form to the theoretical developments. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Synthesis of monotonic passband sharp cutoff filters with constant group delay response

    Page(s): 597 - 602
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (672 KB)  

    A new technique is developed for determining nonminimumphase filter functions that can meet stringent specifications for both the attenuation and group delay characteristics. The organization of a computer program for automatic computation of these transfer functions according to any prescribed specifications is described. As revealed by comparison with some other known approximation techniques, the method proposed is very effective, especially when the delay equalization to within a prescribed error is not required over the entire passband. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Three-point iterations derived from exponential curve fitting

    Page(s): 669 - 670
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (152 KB)  

    Three-point Iterative methods make use of an approximating function, p(x) of F(x) which functions have three x and y values in common. Some functions p(x) give an analytic expression in x when p(x) = O \cdot p(x) can be a hyperbolic function, a quadratic polynomial, or an exponential function. In this paper it is demonstrated that the hyperbolic function is a special case, derived from the exponential function. The condition for convergence and the rate of convergence are discussed. A comparison by means of some examples is made with the hyperbolic function and with the method of Jarratt and Nudds which is a modification of the hyperbolic approximation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.