Configurable hardware implementation of a pipelined DNLMS adaptive filter

The delayed normalized least-mean-square (DNLMS) adaptive filtering algorithm is suitable for implementing pipelined architectures. Though previous literature has provided such architectures for DNLMS adaptive filters, none have given a detailed implementation. This paper presents the configurable h...

Full description

Saved in:
Bibliographic Details
Published in:2014 26th International Conference on Microelectronics (ICM) pp. 9 - 12
Main Authors: Lee, Raymond, Khalid, Mohammed A. S., Abdel-Raheem, Esam
Format: Conference Proceeding
Language:English
Published: IEEE 01-12-2014
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The delayed normalized least-mean-square (DNLMS) adaptive filtering algorithm is suitable for implementing pipelined architectures. Though previous literature has provided such architectures for DNLMS adaptive filters, none have given a detailed implementation. This paper presents the configurable hardware implementation of a pipelined, modular, low-latency, portable DNLMS adaptive filter which is tested for echo cancellation. The design is implemented onto the Altera Stratix FPGA and has a maximum operating frequency of 32.27 MHz. The design methodology consists of architectural derivation, fixed-point and RTL simulations, physical synthesis, and real-time hardware.
ISSN:2159-1660
DOI:10.1109/ICM.2014.7071793