Convergence Property of Tri-Quantized-x NLMS Algorithm
スポンサーリンク
概要
- 論文の詳細を見る
The signed regressor algorithm, a variation of the least mean square(LMS) algorithm, is characterized by the estimation way of using the clipped reference signals, namely, its sign (±). This clipping, equivalent to quantizing the reference signal to ±1, only increases the estimation error by about 2 dB. This paper proposes to increase the number of the quantization steps to three, namely, ±1 and 0, and shows that the 'tri-quantized-x' normalized least mean square(NLMS) algorithm with three quantization steps improves the convergence property.
- 社団法人電子情報通信学会の論文
- 2000-12-25
著者
-
FUJII Kensaku
Wireless Communication Systems Lab. (L41), FUJlTSU LABORATORIES LTD.
-
TANAKA Yoshinori
Wireless Communication Systems Lab. (L41), FUJlTSU LABORATORIES LTD.
-
Fujii K
Wireless Communication Systems Lab.(l41) Fujitsu Laboratories Ltd.
-
Tanaka Y
Fujitsu Lab. Ltd. Kawasaki‐shi Jpn
-
Tanaka Yoshinori
Wireless Communication Systems Lab.(l41) Fujitsu Laboratories Ltd.
関連論文
- Analysis on the Convergence Property of the Sub-RLS Algorithm
- Normalized Least Mean EE' algorithm and Its Convergence Condition(Special Section on Acoustic Signal Processing)
- Hybrid Active Noise Control Systems Based on the Simultaneous Equations Method (Special Section on Noise Cancellation and Reduction Techniques)
- Convergence Property of Tri-Quantized-x NLMS Algorithm
- Oxidative Depolymerization of Chitosan by Hydroxyl Radical