Browse Prior Art Database

A successive minimum-mean-square-error OFDM-data-symbol cancellation scheme for channel estimation in the Pseudo-Random-Postfix OFDM context

IP.com Disclosure Number: IPCOM000031047D
Original Publication Date: 2004-Sep-08
Included in the Prior Art Database: 2004-Sep-08
Document File: 5 page(s) / 106K

Publishing Venue

Motorola

Related People

Markus Muck: AUTHOR [+4]

Abstract

This contribution proposes an iterative channel impulse response (CIR) estimation scheme in the context of the Pseudo-Random-Postfix OFDM (PRP-OFDM) modulation. While conventional techniques eliminate interference on the CIR estimates by simple mean value calculation, the new proposal uses soft decoder outputs in order to perform iterative interference cancellation. In a typical example for QAM-64 constellations, the mean-square-error (MSE) of the CIR estimates is improved by approx. 12dB after three iterations. Based on PRP-OFDM postfixes only, it is thus possible to perform channel estimation for higher order constellations (QAM-64 and higher) with an initial CIR estimation over a small observation window (for QAM-64 typically 30 to 40 OFDM symbols). Any loss in throughput due to additional redundancy for CIR estimation purposes, e.g. pilot tones, learning symbols, etc. is avoided at the cost of an increase in decoding complexity.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 40% of the total text.

A successive minimum-mean-square-error OFDM-data-symbol cancellation scheme for channel estimation in the Pseudo-Random-Postfix OFDM context

Authors: Markus Muck, Xavier Miet, Marc de Courville, Pierre Duhamel

Abstract

This contribution proposes an iterative channel impulse response (CIR) estimation scheme in the context of the Pseudo-Random-Postfix OFDM (PRP-OFDM) modulation. While conventional techniques eliminate interference on the CIR estimates by simple mean value calculation, the new proposal uses soft decoder outputs in order to perform iterative interference cancellation. In a typical example for QAM-64 constellations, the mean-square-error (MSE) of the CIR estimates is improved by approx. 12dB after three iterations. Based on PRP-OFDM postfixes only, it is thus possible to perform channel estimation for higher order constellations (QAM-64 and higher) with an initial CIR estimation over a small observation window (for QAM-64 typically 30 to 40 OFDM symbols). Any loss in throughput due to additional redundancy for CIR estimation purposes, e.g. pilot tones, learning symbols, etc. is avoided at the cost of an increase in decoding complexity. 

1. Introduction

Nowadays, Orthogonal Frequency Division Multiplexing (OFDM) seems the preferred modulation scheme for modern broadband communication systems [1-4].

All these systems are based on a traditional Cyclic Prefix OFDM (CP-OFDM) modulation scheme. Typically, coherent modulation schemes are applied which require channel estimates for the decoding in the receiver. The problem of CP-OFDM is that this estimation step usually requires learning symbols and pilot tones which reduce the system throughput. In order to avoid this drawback, a new modulation scheme has been recently proposed: the Pseudo Random Postfix OFDM (PRP-OFDM) modulation [5-6] uses instead of a cyclic prefix extension a known vector weighted by a pseudo random scalar sequence [7]. This way, unlike for the former OFDM modulators, the receiver can exploit an additional information: the prior knowledge of a part of the transmitted block. [5] illustrates several efficient equalization and decoding schemes assuming perfect synchronization. [8] compares standard CP-OFDM with CIR estimation based on rotating pilot schemes to low-complexity PRP-OFDM CIR estimation: If the interference of OFDM data symbols on the CIR estimated is simply eliminated by mean value calculation over a large observation window, in a typical example PRP-OFDM based CIR estimation leads to superior results in terms of mean-square-error (MSE) up to an SNR of approx. 15dB.

This contribution proposes an iterative CIR estimation scheme that refines first rough CIR estimates and makes PRP-OFDM suitable for higher constellation types, i.e. QAM-64 and higher. The new scheme is complex if used for CIR estimation only, but in combination with iterative forward-error-correction (FEC) decoding schemes the system complexity only increases slightly.

2. Channel estimation

In...