An optimized transmultiplexer using combinational window functions

Abstract

This paper proposes an efficient approach for the design of M-channel maximally decimated near-perfect reconstruction (NPR) type transmultiplexer. Cosine modulation is used to design the synthesis and analysis sections of the transmultiplexer. The prototype filter is designed by using high sidelobe falloff rate (SLFOR) combinational window functions. A bisection-type optimization algorithm has been applied to minimize the interference parameters like inter-channel interference (ICI) and inter-symbol interference (ISI). The proposed algorithm has certain advantages than earlier reported work. The algorithm is of generalized nature, independent of the window function used in the design of the prototype filter. Second, it is fast and computationally efficient as only a single parameter is used as a variable which provides almost uniform interference level in all subchannels. Design examples are included for comparison with earlier reported work. Very small values of ICI and ISI have been obtained by using variable and combinational window functions.

A prototype filter design for cosine modulated transmultiplexer using weighted constrained least squares technique

“…For the given RF = 1.6, A p = 1.63 × 10 −3 , A s = 40 dB, cutoff frequency is calculated from Eq. (15), which is ω c = 0.05, stop band edge frequency (ω s ) = 0.1 using ω s = RF/2M and in this proposed method, ω p is kept slightly smaller than ω c , because in near ideal filter, ω p and ω c must be the same (ω p = ω c − 0.001). The stop band and pass band ripples are calculated using Eqs.…”

“…In general, N is computed using N = (A s − 7.95)/14.36 f, where f = (ω s − ω p )/2 . However, in this work, for comparing the performance of the proposed method, same value of N is taken as in earlier published results [3,15,31]. The prototype filter has been designed with the given specifications, and the resulting parameters are ICI = −252.78 dB, ISI = −116.58 dB, SICI = 116.86 dB, SISI = −19.33 dB, and SI = −19.33 dB.…”

“…They were commenced to drop off the system cost in addition to increase the processing rates. In the past, a number of TMUX systems have been devised to minimize the system cost, overall computational complexity needed to design a transmultiplexer system [10][11][12][13][14][15][16]. In early stage of research, transmultiplexer systems were designed using either non-DFT based and or DFT based techniques [17].…”

“…Based on the cosine modulation technique, several algorithms [21][22][23][24][25][26][27][28] were developed for designing transmultiplexer systems. A detailed discussion on CM TMUXs, and their applications are given in [1,2,[11][12][13][14][15][16] and the references therein.…”

An Optimized Near Perfect Reconstruction Trans-multiplexer

“…Referring to Fig. 1, the relation between M-input signals and output of l th sub-channel in z-domain can be expressed [6] where C(z) represents the transmission characteristics of the transmission channel. The eqn.…”