1.

論文

論文
Hirano, Akihiro ; Nakayama, Kenji
出版情報: 第26回信号処理シンポジウム講演論文集 = Proc. of 25th SIP Symposium.  pp.477-481,  2011-01-01.  電子情報通信学会 = The Institute of Electronics, Information and Communication Engineers
URL: http://hdl.handle.net/2297/35266
概要: This paper presents efficient implementa- tion of RLS-based adaptive filters with a large number of taps on nVIDIA GeForce graphics processing unit (GPU) and CUDA software development environment. Modification of the order and the combination of calcu- lations reduces the number of accesses to slow off-chip memory. Assigning tasks into multiple threads also takes memory access order into account. For a 4096-tap case, a GPU program is almost three times faster than a CPU program. 続きを見る
2.

論文

論文
Hirano, Akihiro ; Nakayama, Kenji
出版情報: 第27回信号処理シンポジウム講演論文集 = Proc. of 27th SIP Symposium.  pp.241-245,  2012-01-01.  電子情報通信学会 = The Institute of Electronics, Information and Communication Engineers
URL: http://hdl.handle.net/2297/35267
概要: This paper presents efficient implementa- tion of RLS-based adaptive filters with a large number of taps on nVIDIA GeForce graphics processing unit (GPU) and CUDA software development environment. Modification of the order and the combination of calcu- lations reduces the number of accesses to slow off-chip memory. Assigning tasks into multiple threads also takes memory access order into account. Multiple shader pro- cessor arrays are used to handle a large matrix. For a 8192-tap case, a GPU program is almost 30-times faster than a CPU program. Real-time processing is possible for an 8kHz-sampling and 512-tap case by us- ing 32 shader processors, which is only 25% of GeForce 8800GTS. 続きを見る
3.

論文

論文
Horita, Akihide ; Nakayama, Kenji ; Hirano, Akihiro
出版情報: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).  4668 LNCS - I  pp.169-179,  2007-01-01.  Springer Verlag (Germany)
URL: http://hdl.handle.net/2297/9503
概要: 金沢大学大学院自然科学研究科情報システム<br />Feed-Forward (FF-) and Feed-Back (FB-) structures have been proposed for Blind Source Separati on (BSS). The FF-BSS systems have some degrees of freedom in the solution space, and signal distortion is likely to occur in convolutive mixtures. On the other hand, the FB-BSS structure does not cause signal distortion. However, it requires a condition on the propagation delays in the mixing process. In this paper, source separation performance in the FB-BSS is theoretically analyzed taking the propagation delays into account. Simulation is carried out by using white signals and speech signals as the signal sources. The FF-BSS system and the FB-BSS system are compared. Even though the FB-BSS can provide good separation performance, there exits some limitation on location of the signal sources and the sensors. © Springer-Verlag Berlin Heidelberg 2007. 続きを見る
4.

論文

論文
Hirano, Akihiro ; Nakayama, Kenji ; Arai, Shinya ; Deguchi, Masayuki
出版情報: IEICE transactions on fundamentals of electronics, communications and computer sciences.  E84-A  pp.414-421,  2001-02-01. 
URL: http://hdl.handle.net/2297/5653
概要: 金沢大学大学院自然科学研究科知能情報・数理
5.

論文

論文
Nakayama, Kenji ; Hirano, Akihiro ; Katoh, Shinya ; Yamamoto, Tadashi ; Nakanishi, Kenichi ; Sawada, Manabu
出版情報: Proceedings of the International Joint Conference on Neural Networks.  2  pp.1373-1378,  2002-05-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6806
概要: In training neural networks, it is important to reduce input variables for saving memory, reducing network size, and ach ieving fast training. This paper proposes two kinds of selecting methods for useful input variables. One of them is to use information of connection weights after training. If a sum of absolute value of the connection weights related to the input node is large, then this input variable is selected. In some case, only positive connection weights are taken into account. The other method is based on correlation coefficients among the input variables. If a time series of the input variable can be obtained by amplifying and shifting that of another input variable, then the former can be absorbed in the latter. These analysis methods are applied to predicting cutting error caused by thermal expansion and compression in machine tools. The input variables are reduced from 32 points to 16 points, while maintaining good prediction within 6 ホシm, which can be applicable to real machine tools. 続きを見る
6.

論文

論文
Nakayama, Kenji ; Hirano, Akihiro ; Dejima, Yasuhiro
出版情報: IEEE Proc. The 47th International Midwest Symposium on Circuits and Systems (MWSCAS2004), Hiroshima, Japan.  3  pp.III_207-III_210,  2004-09-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6816
概要: This paper analyzes signal separation and distortion in a feedforward blind source separation (BSS) applied to convolutive mixture environment. In the BSS learning, a separation block is adjusted so as to make the output signals statistically independent. One direction for this adjustment is to extract only one signal source at a single output. The other direction is to make the output signal spectra white. The second direction causes signal distortion and conflicts with the first direction. This confliction prevent signal separation. These relations are also dependent on a length of the impulse responses in the mixture and the filter length in the separation block. These properties are theoretically analyzed and are investigated through simulations using the white signal sources and the speech signal sources. 続きを見る
7.

論文

論文
Tokui, N. ; Nakayama, Kenji ; Hirano, Akihiro
出版情報: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings.  6  pp.349-352,  2003-01-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6812
概要: In order to achieve fast convergence and less computation for adaptive filters, a joint method combining a whitening process and the NLMS algorithm is a hopeful approach. However, updating the filter coefficients is not synchronized with the reflection coefficient updating resulting in unstable behavior. We analyzed effects of this, and proposed the Synchronized Learning Algorithm to solve this problem. Asynchronous error between them is removed, and fast convergence and small residual error were obtained. This algorithm, however, requires O(ML) computations, where M is an adaptive filter length, and L is a lattice predictor length. It is still large compared with the NLMS algorithm. In order to achieve less computation while the fast convergence is maintained, a block implementation method is proposed. The reflection coefficients are updated at some period, and are fixed during this interval. The proposed block implementation can be effectively applied to parallel form adaptive filters, such as sub-band adaptive filters. Simulation using speech signal shows that a learning curve of the proposed block implementation a little slower than the our original algorithm, but can save the computational complexity. 続きを見る
8.

論文

論文
Hirano, Akihiro ; Nakayama, Kenji ; Ushimaru, S.
出版情報: Proc. 2003 IEEE International Symposium on Intellogent Signal Processing and Communication Systems (ISPACS2003), Awajishima, Japan.  pp.367-370,  2003-12-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6802
9.

論文

論文
Nakayama, Kenji ; Katou, Haruo ; Hirano, Akihiro
出版情報: Proc.IEEE, ISPACS2006, Yonago, Japan.  pp.459-462,  2006-12-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6849
10.

論文

論文
Horita, Akihide ; Nakayama, Kenji ; Hirano, Akihiro ; Dejima, Yasuhiro
出版情報: Proceedings of the International Joint Conference on Neural Networks.  2  pp.1257-1262,  2005-08-01.  IEEE(Institute of Electrical and Electronics Engineers)
URL: http://hdl.handle.net/2297/6815
概要: Source separation and signal distortion in three kinds of BSSs with convolutive mixture are analyzed. They include a fee dforward BSS, trained in the time domain and in the frequency domain, and a feedback BSS, trained in the time domain. First, an evaluation measure of signal distortion is discussed. Second, conditions for source separation and distortion free are derived. Based on these conditions, source separation and signal distortion are analyzed. The feedforward BSS has some degree of freedom, and the output spectrum can be changed. The feedforward BSS, trained in the frequency domain, has weighting effect, which can suppress signal distortion. This weighting is, however, effective only when the source spectra are similar to each other. Since, the feedforward BSS, trained in the time domain, does not have any constraints on signal distortion free, its output signals can be easily distorted. A new learning algorithm with a distortion free constraint is proposed. On the other hand, the feedback BSS can satisfy both source separation and distortion free conditions simultaneously. Simulation results support the theoretical analysis. © 2005 IEEE. 続きを見る