Electronic Filters pp 185-207 | Cite as

# Linear Phase Low-Pass Filters with Improved Selectivity

## Abstract

Synthesis of constant group delay low-pass polynomial filters in a maximally flat manner (having all derivatives in the origin equal to zero) is straightforward. One uses the Bessel polynomials with the complex variable introduced instead of the usual real *x*. That is described in this chapter and tables of coefficients and poles of the first ten Bessel polynomials are given. A design example is given too. The next issue visited is approximation of constant group delay by a polynomial filter in equi-ripple manner. An iterative algorithm is proposed allowing for this problem to be solved for arbitrary values of the order of the filter and the maximum group delay error assigned in percentage of the nominal one. Influence of the approximation error to the properties of the filter is studied. Both, maximally flat and equi-ripple group delay approximants exhibit poor selectivity. To improve that an algorithm is described and an example is given as to how transmission zeros at the ω-axis may be introduced. Comparisons are made (in the frequency and time domain) between the two linear phase polynomial solutions being extended with transmission zeros. Finally, advice is given as to how to synthesize low-pass all-pass functions.

## References

- 1.Thomson WE (1949) Delay networks having maximally flat frequency characteristics. Proc Inst Electr Eng Part III 96(44):487–490Google Scholar
- 2.Starch L (1954) Synthesis of constant-time-delay ladder networks using Bessel poly-nomials. Proc IRE 42:1666–1675CrossRefGoogle Scholar
- 3.Dimopoulos H (2012) Analog electronic filters theory design and synthesis. Springer, New YorkCrossRefGoogle Scholar
- 4.Bond C (2003) Bessel filters polynomials, poles and circuit elements. http://www.crbond.com/papers/bsf2.pdf. Last visited May 2019
- 5.Litovski VB (1979) Synthesis of monotonic passband sharp cutoff filters with constant group delay response. IEEE Trans Circuits Syst CAS 26(8):597–602CrossRefGoogle Scholar
- 6.Remez EY (1934) Sur la détermination des polynômes d’approximation de degré donnée. Comm Soc Math Kharkov 10(41):41–63Google Scholar
- 7.Humpherys DV (1963) Rational function approximation of polynomials with equiripple error. Report R-159, Coordinated Science Laboratory University of Illinois, Urbana, Contract Da-36-039-Sc-85122 D/A Sub-Task 3-99-01-002Google Scholar
- 8.Skwirzinski J, Zdunek J (1964) Design of networks with prescribed delay and amplitude characteristics. Marconi Rev Fourth Quart 1964:162–187Google Scholar
- 9.Carlin H, Wu J (1976) Amplitude selectivity versus constant delay in minimum-phase lossless filters. IEEE Trans Circuits Syst 23(7):447–455CrossRefGoogle Scholar
- 10.Gregorian R, Temes GC (1978) Design techniques for digital and analog all-pass circuits. IEEE Trans Circuits Syst CAS 25(12):981–988CrossRefGoogle Scholar
- 11.Hibino M, Ishizaki Y, Watanabe H (1968) Design of chebyshev filters with flat group-delay characteristics. IEEE Trans Circuit Theory 15(4):316–325CrossRefGoogle Scholar