A 2-bit Adaptive Delta Modulation System with Improved Performance

  • PDF / 496,188 Bytes
  • 5 Pages / 600.03 x 792 pts Page_size
  • 11 Downloads / 181 Views

DOWNLOAD

REPORT


Research Article A 2-bit Adaptive Delta Modulation System with Improved Performance E. A. Prosalentis and G. S. Tombras Laboratory of Electronics, Department of Physics, University of Athens, Panepistimiopolis, Athens 157 84, Greece Received 10 May 2006; Revised 24 August 2006; Accepted 26 August 2006 Recommended by Douglas O’shaughnessy A 2-bit adaptive delta modulation system with improved performance is proposed in this paper. Its main characteristic is a new adaptation algorithm that incorporates both memory and look-ahead instantaneous step-size estimation and leads the modulator into generating a 2-bit output codeword. As shown by computer simulation results, the proposed system offers reduced overshoot and fast response to signal variations in comparison to other similar systems. Copyright © 2007 E. A. Prosalentis and G. S. Tombras. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1.

INTRODUCTION

Adaptive delta modulation (ADM) is a common alternative to fixed step-size delta modulation (DM) offering increased dynamic range and reduced slope-overload noise at the expense of some added complexity. This is achieved by varying the step size of the basic 1-bit quantizer according to a decided rule with respect to input signal variations. Among the many adaptive schemes that have been described in the past, the widely known ADM with 1-bit memory or first-order constant factor delta modulation (CFDM), [1, 2], was the first system to introduce a “memory” function for step-size estimation at each sampling instant. Since then, various modifications and extensions of that basic instantaneously adaptive scheme have been proposed in the literature including 2- or 3-bit ADM and 2-digit ADM [2– 8]. These multidigit systems provide for a variable rate in step changes between adjacent sampling instants incorporating various forms of “memory” and/or “look-ahead” stepsize estimation, that is, feedback and/or feedforward adaptation, respectively, and offer enhanced overall performance in normalized comparison to single-bit adaptive DM [6–8]. Moreover, although they produce multidigit output codewords, they are considered to maintain the basic property of DM in that the quantized output signal value at each sampling instance is obtained from the predicted signal sample by adding or subtracting the corresponding stepsize.

In this paper, we consider the adaptation algorithms of two multidigit ADM schemes, the 2-digit adaptive system by Tombras [7] and Tombras and Karybakas [8] and the 2-bit adaptive system by Aldajani and Sayed [9, 10]. Both systems, briefly described in Section 2, offer an exponentially variable rate in step-size changes and the corresponding quantizers generate output codewords with information about both the sign and the relative magnitude of the step-size to the receiver decoder. Following this approach, in Section 3, we describe a modified adapt