WO2005015737A1 - システム推定方法及びプログラム及び記録媒体、システム推定装置 - Google Patents
システム推定方法及びプログラム及び記録媒体、システム推定装置 Download PDFInfo
- Publication number
- WO2005015737A1 WO2005015737A1 PCT/JP2004/011568 JP2004011568W WO2005015737A1 WO 2005015737 A1 WO2005015737 A1 WO 2005015737A1 JP 2004011568 W JP2004011568 W JP 2004011568W WO 2005015737 A1 WO2005015737 A1 WO 2005015737A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- filter
- processing unit
- matrix
- value
- state
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B3/00—Line transmission systems
- H04B3/02—Details
- H04B3/20—Reducing echo effects or singing; Opening or closing transmitting path; Conditioning for transmission in one direction or the other
- H04B3/23—Reducing echo effects or singing; Opening or closing transmitting path; Conditioning for transmission in one direction or the other using a replica of transmitted signal in the time domain, e.g. echo cancellers
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
- G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H21/0043—Adaptive algorithms
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H21/0043—Adaptive algorithms
- H03H2021/0049—Recursive least squares algorithm
- H03H2021/005—Recursive least squares algorithm with forgetting factor
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S367/00—Communications, electrical: acoustic wave systems and devices
- Y10S367/901—Noise or unwanted signal reduction in nonseismic receiving system
Definitions
- the present invention is a system estimation method and program, and a recording medium, relates to a system estimation device, in particular, using a high-speed H ⁇ filter ring algorithm of the hyper H ⁇ filter has been developed based on H ⁇ evaluation criterion, robust state estimates TECHNICAL FIELD
- the present invention relates to a system estimation method, a program, a recording medium, and a system estimation device that simultaneously realize optimization and a forgetting coefficient.
- system estimation refers to estimating parameters of a mathematical model (transfer function or impulse response) of an input / output relationship of a system based on input / output data.
- Typical applications are echo cancellers in international communications, automatic equalizers in data communications, echo cancellers in acoustic systems, sound field reproduction, and active noise control in automobiles.
- Figure 8 shows an example of a configuration diagram for the system estimation (unknown system Well be represented by IIR (Infinite Impulse Response) Fireta ⁇ ) 0
- This system includes an unknown system 1 and an adaptive filter 2.
- the adaptive filter 2 has an FIR digital filter 3 and an adaptive algorithm 4.
- a linear system represented by a state-space model such as
- x k state vector or simply state; unknown, which is the object of estimation.
- y k observation signal; input to the filter and is known.
- H k observation matrix
- v k observation noise; unknown.
- P forgetting factor; generally determined by trial and error.
- K k filter gain; obtained from the matrix ⁇ I k .
- ⁇ ⁇ +1] k Corresponds to the covariance matrix of the error of xi
- 0 Corresponds to the initial state covariance matrix; unknown, but ⁇ 0 ⁇ ⁇ is used for convenience.
- the present inventor has already proposed a system identification algorithm by high Eta ⁇ filter (see Patent Document 1).
- the fast ⁇ ⁇ filtering algorithm is capable of tracking time-varying systems that change rapidly with the amount of computation per unit time step ⁇ ( ⁇ ). At the limit of the upper limit, it is perfectly matched with the fast Kalman filtering algorithm.
- Such system identification enables high-speed real-time identification and estimation of time-invariant and time-varying systems.
- Non-Patent Documents 2 and 3 As a method generally known in the field of system estimation, see, for example, Non-Patent Documents 2 and 3.
- FIG. 9 shows an illustration of the communication system and echo.
- a hybrid transformer is installed at the connection between the two-wire and four-wire circuits as shown in the figure, and impedance matching is performed. If this impedance matching is perfect, the signal (voice) from speaker ⁇ reaches only speaker ⁇ . However, it is generally difficult to achieve perfect matching, and a phenomenon occurs in which a part of the received signal leaks to the four-wire circuit, is amplified, and then returns to the receiver (talker A) again. This is an echo.
- the echo increases as the transmission distance increases (the delay time increases The effect is large and the quality of the call is significantly degraded (in the case of pulse transmission, the deterioration of the call quality due to the echo is significant even at a short distance).
- FIG. 10 shows a principle diagram of the echo canceller.
- an echo canceller is introduced, the impulse response of the echo path is sequentially estimated using the directly observable received signal and the echo, and a pseudo echo obtained using the echo is estimated from the actual echo.
- the subtraction cancels the echo and attempts to remove it.
- Estimation of the impulse response of the echo path is performed so that the mean square error of the residual echo e k is minimized.
- the factors that interfere with the estimation of the echo path are the line noise and the signal (voice) from speaker A.
- the estimation of the impulse response is interrupted. Since the impulse response length of the hybrid transformer is about 50 [ms], if the sampling period is set to 125 [ ⁇ s], the order of the impulse response of the echo path is actually about 400.
- Patent Document 1
- the error covariance matrix used in the Kalman filter originally has a quadratic form with any non-zero vector that is always positive (hereinafter referred to as “positive definite”). It is known that the quadratic form becomes negative (hereinafter referred to as “negative definite”), and becomes numerically unstable. Also, since the computational complexity is 0 (N 2 ) (or 0 (N 3 )), if the dimension N of the state vector x k is large, the number of operations per step increases rapidly, and Not suitable.
- an object of the present invention is to establish an estimation method capable of theoretically optimally determining a forgetting coefficient, and to develop a numerically stable estimation algorithm and a high-speed algorithm thereof.
- Another object of the present invention is to provide a system estimation method that can be applied to an echo canceller, sound field reproduction, noise control, or the like in a communication system or an acoustic system.
- the present invention in order to solve the above problems, forgetting ⁇ number to derive the optimal determinable state estimation algorithm using the H ⁇ optimization method devised anew.
- the processing unit reads an initial value or a value including the observation matrix H k at a certain time from the storage unit, and executes a hyper H ⁇ filter represented by the following equation using the forgetting coefficient; 0,
- Processing unit storing a value obtained regarding the hyper H ⁇ filter in the storage unit, processing unit, the observation matrix a filter gain K s,; by, the upper limit value r f and the forgetting coefficient) 0 calculating the existence condition based, the processing section, the step of performing said hyper H ⁇ filter will reduce the upper limit value r f Repeating, setting the upper limit value smaller in a range where the existence condition is satisfied at each time, and storing the value in a storage unit;
- a processing unit that executes an estimation algorithm
- a storage unit that is read and / or written by the processing unit and stores observation values, set values, and estimated values related to the state space model
- the processing unit, the upper limit value f, the input is the observed signal y k of the filter, entering a value including the observation matrix H k from the storage unit or the input unit, the processing unit, in accordance with the upper limit value r f, the state Determining the forgetting factor P associated with the spatial model,
- the processing unit reads an initial value or a value including the observation matrix H k at a certain time from the storage unit, and executes a hyper H ⁇ filter represented by the following equation using the forgetting factor; 0,
- the processing unit stores a value obtained for the hyper H ⁇ filter in a storage unit.
- the processing unit the observation matrix a filter gain K s,; by, calculating the existence condition based on the upper limit value r f and the forgetting ⁇ number), wherein the processing unit gradually reduces the upper limit value r f
- the upper limit value is set to a small value within a range where the existence condition is satisfied at each time, and the value is stored in the storage unit.
- the system estimation device provided with:
- the estimation method of the present invention can optimally determine the forgetting factor and can operate stably even with a single precision, so that high performance can be realized at low cost.
- ordinary private telecommunications equipment often performs calculations with single precision in terms of cost and speed. For this reason, the present invention will bring its effect to various industrial fields as a practical state estimation algorithm.
- FIG. 1 is a configuration diagram of hardware according to the present embodiment.
- FIG. 3 is a flowchart of the algorithm of the H ⁇ filter (S105) in FIG.
- FIG. 4 is an explanatory diagram of the square root array algorithm of Theorem 2.
- Figure 7 shows the result of estimating the impulse response using the fast numerically stable algorithm of Theorem 3.
- FIG. 8 is a configuration diagram for system estimation.
- FIG. 9 is an explanatory diagram of a communication system and an echo.
- FIG. 10 is a principle diagram of the echo canceller.
- x k state vector or simply state; unknown, which is the subject of estimation.
- v k Observation noise; unknown.
- y k observation signal; input to the filter and is known.
- G k driving matrix; known at the time of execution.
- H k observation matrix
- ⁇ ! 0 Corresponds to the initial state covariance matrix; unknown, but ⁇ for convenience. 1 is used.
- K s , k Filter gain; obtained from matrix ⁇ I.
- the system estimation method or the system estimation device ′ system includes a system estimation program for causing a computer to execute each procedure, a computer-readable recording medium recording the system estimation program, and a computer internal memory including the system estimation program. It can be provided by a program product that can be loaded into a computer, a computer such as a server including the program, or the like.
- FIG. 1 is a configuration diagram of hardware according to the present embodiment.
- This hardware includes a processing unit 101, which is a central processing unit (CPU), an input unit 102, an output unit 15103, a display unit 104, and a storage unit 105. Further, the processing unit 101, the input unit 102, the output unit 103, the display unit 104, and the storage unit 105 are connected by an appropriate connection means such as a star or a bus.
- the storage unit 105 stores the known data indicated in “1. Explanation of symbols” estimated by the system as necessary. Also, unknown "known data Ya ⁇ been hyper -H ⁇ filter on the data 'other data by the processing unit 101, is written inclusive and Z or read as necessary.
- Equation (11) represents the filter equation
- equation (12) represents the filter gain
- equation (13) represents the Recatch equation.
- the driving matrix G K is generated as follows.
- the upper limit value r f is set to be smaller as possible so as to satisfy the following existence condition.
- (r f ) is a monotone decay function of r f that satisfies;
- the feature of Theorem 1 is that robustness of state estimation and optimization of forgetting factor p are performed simultaneously.
- FIG. 2 shows a flowchart of the robustness of the H ⁇ filter and the optimization of the forgetting factor yO.
- Block “EXC> 0” H ⁇ filter existence condition
- the processing unit 101 reads or inputs the upper limit value r f from the storage unit 105 or the input unit 102 (S101). In this example, r f >> 1 is given.
- the processing unit 101 determines the forgetting factor p according to equation (15) (S103).
- the processing unit 101 executes the hyper H ⁇ filter of Expressions (10) to (13) based on the forgetting factor yO (S105).
- the processing unit 101 calculates the right side of equation (17) (or equation (18) described later) (this is EXC) (S107), and if its existence condition is satisfied at all times (sio9), to reduce the r f ⁇ r just repeating the same processing (si 11).
- the information is stored in the unit 105 (S113).
- ⁇ r is added, but other preset values may be added.
- This optimization process is called r-iteration.c
- the processing unit 101 needs the appropriate intermediate value and final value obtained in each step such as the H ⁇ filter calculation step S105 and the existence condition calculation step S107.
- the information may be stored in the storage unit 105 as appropriate and read from the storage unit 105.
- FIG. 3 shows a flowchart of the algorithm of the (hyper) H ⁇ filter (S105) in FIG.
- the hyper H ⁇ filtering algorithm can be summarized as follows.
- Step S201 The processing unit 101 reads the initial condition of the recursive expression from the storage unit 105, or inputs the initial condition from the input unit 102 and determines it as illustrated. Note that L indicates a predetermined maximum number of data.
- Step S203 The processing unit 101 compares the time k with the maximum data number L. If the time k is larger than the maximum number of data, the processing section 101 ends the processing, and if it is less than the maximum data number, the processing section 101 proceeds to the next step. (If not necessary, the conditional statement can be removed. Or, if necessary, restarting may be performed.)
- Step S205 The processing unit 101 calculates the filter gain K s , k using Expression (12).
- Step S207 The processing unit 101 updates the filter equation of the hyper H ⁇ filter of Expression (11).
- Step S209 The processing section 101, section corresponds to the covariance matrix of the error sigma
- the processing unit 101 stores in an appropriate storage unit "! 05 as needed H ⁇ filter intermediate value calculated meth appropriate in each step, such as computation steps S205 ⁇ S209 and final value, the value or the like of the existence condition Alternatively, the information may be read from the storage unit 105.
- the existence of the hyper H ⁇ filter can be determined with the complexity O (N).
- K si is the filter gain determined by equation (12) ⁇ (proof)
- the above hyper H ⁇ filter is The amount of calculation per unit time step is 0 (N 2 ), that is, an arithmetic operation proportional to N 2 is required.
- N is the dimension of the state vector x k . Therefore, as the dimension of x k increases, the computation time required to execute this filter increases rapidly.
- the error covariance matrix must always be positive definite due to its properties, but it may be numerically negative definite. In particular, when calculation is performed with single precision, this tendency becomes remarkable. At this time, it is known that the filter becomes unstable. Therefore, in order to make the algorithm practical and cost-effective, it is desirable to develop a state estimation algorithm that can operate with single precision (eg, 32 bits).
- FIG. 4 shows an explanatory diagram of the square root array algorithm of Theorem 2. This calculation algorithm can be used in the calculation of the H ⁇ filter (S105) in the flowchart of Theorem 1 shown in FIG.
- the factor matrix ⁇ ⁇ (square root matrix of ⁇ ⁇ ) is obtained using an update formula based on the j-unitary transformation.
- the filter gain k is obtained as shown in the figure from the 1-1 block matrix and the 2-1 block matrix generated at this time. Therefore, ⁇ 1/2 1/2 K s k k-k k -1 k k-> ⁇ , and the positive definiteness of is guaranteed, and can be numerically stabilized. Note that the amount of calculation per unit step of the ⁇ filter of Theorem 2 remains 0 ( ⁇ 2 ).
- the processing unit 101 reads a term included in each element of the determinant on the left side of the equation (22) from the storage unit 105 or obtains the term from an internal memory or the like, and executes a J-unitary transformation (S301).
- the processing unit 101 calculates the system gains K k , K s , from the elements of the determinant on the right side of the obtained equation (22). The calculation is performed based on the equation (21) (S303, S305).
- the processing unit 101 calculates the state estimated value x
- FIG. 5 shows a flowchart of the numerically stable high-speed algorithm of Theorem 3.
- This high-speed algorithm is incorporated into the! "! ⁇ filter calculation step (S1 05) 2 is optimized by T first plate Reshiyon.
- T first plate Reshiyon T first plate Reshiyon.
- the H ⁇ filtering algorithm can be summarized as follows:
- Step S401 The processing unit 101 determines initial conditions of a recursive expression as illustrated.
- L indicates the maximum number of data.
- Step S403 The processing unit 101 compares the time k with the maximum data number L. If the time k is larger than the maximum number of data, the processing unit 101 ends the processing, and if it is less than the maximum number of data, proceeds to the next step. (If it is not necessary, the conditional statement can be removed or restarted.) [Step S405] The processing unit 101 replaces the term K k + 1 corresponding to the filter gain with the equations (27) and (3 1). And recursively calculate.
- Step S406 The processing unit 101 recursively calculates R e , k + 1 using Expression (29).
- Step S407 The processing unit 101 further calculates K s , k using equations (26) and (31).
- Step S409 The processing unit 101 determines the existence condition EXC> 0 here, and proceeds to step S411 if the existence condition is satisfied.
- Step S41 3 how, processor 1 01 increases the required path r f satisfy the presence conditions in step S409, the flow returns to step S401.
- Step S41 1 The processing unit 101 updates the filter equation of the 1-! ⁇ filter of Expression (25).
- Step S415 The processing unit 101 recursively calculates R r .k + 1 using Expression (30). Ma In addition, the processing unit 101 recursively calculates k + 1 using Expressions (28) and (31).
- the processing unit 101 stores the appropriate intermediate value and final value obtained in each step such as the H ⁇ filter calculation steps S405 to S415 and the existence condition calculation step S409 in the appropriate storage unit 105 as necessary, Alternatively, the information may be read from the storage unit 105.
- the (time-varying) impulse response ⁇ hjk] ⁇ of the echo path gives the observed value ⁇ y k ⁇ of the echo ⁇ d k ⁇ . Is expressed by the following equation.
- Vk d k + v k + v k , 2 0, 1,2, ... (33)
- a pseudo echo is generated as follows using the obtained value.
- the problem can be reduced to the problem of successively estimating the impulse response ⁇ hi [k] ⁇ of the echo path from the directly observable received signal iu k ⁇ and echo iy k l.
- Nashi must represent a state space model is first made equation (32) from the state equation and an observation equation. Therefore, since the problem is to estimate the impulse response ⁇ hi [k] ⁇ , let ihi [k] ⁇ be the state variable x k and allow about w k fluctuations, the following for the echo path:
- a state space model can be established.
- the received signal ⁇ uj is approximated by a second-order AR model as follows.
- Figure 7 shows the results of estimating the impulse response using the fast numerically stable algorithm of Theorem 3.
- the vertical axis in Fig. 7 (b) is
- J ( ⁇ ⁇ S) -norm, X, ⁇ , ⁇ on the left side can be determined as follows.
- S is a diagonal matrix having a diagonal component of 1 or 1.
- the observation matrix H k has a shift characteristic
- the present invention will bring its effect to various industrial fields as a practical state estimation algorithm. Further, the present invention can be applied to an echo canceller, a sound field reproduction, a noise control, or the like in a communication system or an acoustic system.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Filters That Use Time-Delay Elements (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Complex Calculations (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/567,514 US7480595B2 (en) | 2003-08-11 | 2004-05-08 | System estimation method and program, recording medium, and system estimation device |
JP2005513012A JP4444919B2 (ja) | 2003-08-11 | 2004-08-05 | システム推定方法及びプログラム及び記録媒体、システム推定装置 |
CN200480022991.8A CN1836372B (zh) | 2003-08-11 | 2004-08-05 | ***估计方法及***估计装置 |
EP04771550.3A EP1657818B1 (en) | 2003-08-11 | 2004-08-05 | System estimation method, program, recording medium, system estimation device |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2003291614 | 2003-08-11 | ||
JP2003-291614 | 2003-08-11 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2005015737A1 true WO2005015737A1 (ja) | 2005-02-17 |
Family
ID=34131662
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2004/011568 WO2005015737A1 (ja) | 2003-08-11 | 2004-08-05 | システム推定方法及びプログラム及び記録媒体、システム推定装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US7480595B2 (ja) |
EP (2) | EP1657818B1 (ja) |
JP (1) | JP4444919B2 (ja) |
CN (1) | CN1836372B (ja) |
WO (1) | WO2005015737A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2007119766A1 (ja) | 2006-04-14 | 2007-10-25 | Incorporated National University Iwate University | システム同定方法及びプログラム及び記憶媒体、システム同定装置 |
JP2008236270A (ja) * | 2007-03-19 | 2008-10-02 | Tokyo Univ Of Science | 雑音抑圧装置および雑音抑圧方法 |
CN107831650A (zh) * | 2016-09-16 | 2018-03-23 | 霍尼韦尔有限公司 | 用于工业上基于模型的过程控制器的闭环模型参数识别技术 |
CN110069870A (zh) * | 2019-04-28 | 2019-07-30 | 河海大学 | 一种基于自适应无迹h∞滤波的发电机动态状态估计方法 |
JP2019205049A (ja) * | 2018-05-23 | 2019-11-28 | 国立大学法人岩手大学 | システム同定装置及び方法及びプログラム及び記憶媒体 |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8924337B2 (en) * | 2011-05-09 | 2014-12-30 | Nokia Corporation | Recursive Bayesian controllers for non-linear acoustic echo cancellation and suppression systems |
IL218047A (en) | 2012-02-12 | 2017-09-28 | Elta Systems Ltd | Devices and methods are added, for spatial suppression of interference in wireless networks |
WO2015136626A1 (ja) * | 2014-03-11 | 2015-09-17 | 株式会社明電舎 | ドライブトレインの試験システム |
US10014906B2 (en) | 2015-09-25 | 2018-07-03 | Microsemi Semiconductor (U.S.) Inc. | Acoustic echo path change detection apparatus and method |
US10122863B2 (en) | 2016-09-13 | 2018-11-06 | Microsemi Semiconductor (U.S.) Inc. | Full duplex voice communication system and method |
CN116679026B (zh) * | 2023-06-27 | 2024-07-12 | 江南大学 | 自适应无偏有限脉冲响应滤波的污水溶解氧浓度估计方法 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07185625A (ja) * | 1993-12-27 | 1995-07-25 | Nippon Steel Corp | 帯状鋼板の最低板厚保障のための制御方法 |
JP2002135171A (ja) * | 2000-10-24 | 2002-05-10 | Japan Science & Technology Corp | システム同定方法 |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS61200713A (ja) | 1985-03-04 | 1986-09-05 | Oki Electric Ind Co Ltd | デイジタルフイルタ |
US5394322A (en) * | 1990-07-16 | 1995-02-28 | The Foxboro Company | Self-tuning controller that extracts process model characteristics |
JP2872547B2 (ja) | 1993-10-13 | 1999-03-17 | シャープ株式会社 | 格子型フィルタを用いた能動制御方法および装置 |
JPH07158625A (ja) * | 1993-12-03 | 1995-06-20 | 英樹 ▲濱▼野 | 吊り下げ部材の取付け用金具 |
SE516835C2 (sv) * | 1995-02-15 | 2002-03-12 | Ericsson Telefon Ab L M | Ekosläckningsförfarande |
US5734715A (en) * | 1995-09-13 | 1998-03-31 | France Telecom | Process and device for adaptive identification and adaptive echo canceller relating thereto |
US5987444A (en) * | 1997-09-23 | 1999-11-16 | Lo; James Ting-Ho | Robust neutral systems |
US6175602B1 (en) * | 1998-05-27 | 2001-01-16 | Telefonaktiebolaget Lm Ericsson (Publ) | Signal noise reduction by spectral subtraction using linear convolution and casual filtering |
US6487257B1 (en) * | 1999-04-12 | 2002-11-26 | Telefonaktiebolaget L M Ericsson | Signal noise reduction by time-domain spectral subtraction using fixed filters |
US6711598B1 (en) * | 1999-11-11 | 2004-03-23 | Tokyo Electron Limited | Method and system for design and implementation of fixed-point filters for control and signal processing |
US6801881B1 (en) * | 2000-03-16 | 2004-10-05 | Tokyo Electron Limited | Method for utilizing waveform relaxation in computer-based simulation models |
-
2004
- 2004-05-08 US US10/567,514 patent/US7480595B2/en active Active
- 2004-08-05 CN CN200480022991.8A patent/CN1836372B/zh active Active
- 2004-08-05 WO PCT/JP2004/011568 patent/WO2005015737A1/ja active Application Filing
- 2004-08-05 EP EP04771550.3A patent/EP1657818B1/en active Active
- 2004-08-05 JP JP2005513012A patent/JP4444919B2/ja active Active
- 2004-08-05 EP EP12007720.1A patent/EP2560281B1/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07185625A (ja) * | 1993-12-27 | 1995-07-25 | Nippon Steel Corp | 帯状鋼板の最低板厚保障のための制御方法 |
JP2002135171A (ja) * | 2000-10-24 | 2002-05-10 | Japan Science & Technology Corp | システム同定方法 |
Non-Patent Citations (3)
Title |
---|
NISHIYAMA, K, ET AL: "H-learning of layered neural networks", IEEE TRANSACTIONS ON NEURAL NETWORKS, USA, vol. 12, 6 November 2001 (2001-11-06), pages 1265 - 1277, XP002904275 * |
NISHIYAMA, K.: "Robust Estimation of a Single Complex Sinusoid in White Noise-H Filtering Approach", IEEE TRANSACTIONS ON SIGNAL PROCESSING, USA, vol. 47, 1999, pages 2853 - 2856, XP002904274 * |
See also references of EP1657818A4 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2007119766A1 (ja) | 2006-04-14 | 2007-10-25 | Incorporated National University Iwate University | システム同定方法及びプログラム及び記憶媒体、システム同定装置 |
JP2008236270A (ja) * | 2007-03-19 | 2008-10-02 | Tokyo Univ Of Science | 雑音抑圧装置および雑音抑圧方法 |
CN107831650A (zh) * | 2016-09-16 | 2018-03-23 | 霍尼韦尔有限公司 | 用于工业上基于模型的过程控制器的闭环模型参数识别技术 |
CN107831650B (zh) * | 2016-09-16 | 2023-01-31 | 霍尼韦尔有限公司 | 用于工业上基于模型的过程控制器的闭环模型参数识别的方法和装置 |
JP2019205049A (ja) * | 2018-05-23 | 2019-11-28 | 国立大学法人岩手大学 | システム同定装置及び方法及びプログラム及び記憶媒体 |
CN110069870A (zh) * | 2019-04-28 | 2019-07-30 | 河海大学 | 一种基于自适应无迹h∞滤波的发电机动态状态估计方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2005015737A1 (ja) | 2006-10-12 |
EP2560281A2 (en) | 2013-02-20 |
EP1657818A1 (en) | 2006-05-17 |
EP2560281A3 (en) | 2013-03-13 |
EP1657818B1 (en) | 2015-04-15 |
US20070185693A1 (en) | 2007-08-09 |
JP4444919B2 (ja) | 2010-03-31 |
CN1836372A (zh) | 2006-09-20 |
CN1836372B (zh) | 2010-04-28 |
US7480595B2 (en) | 2009-01-20 |
EP2560281B1 (en) | 2015-04-15 |
EP1657818A4 (en) | 2012-09-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2327156B1 (en) | Method for determining updated filter coefficients of an adaptive filter adapted by an lms algorithm with pre-whitening | |
EP2012430B1 (en) | System identifying method and program, storage medium, and system identifying device | |
JP2004514317A (ja) | 適応フィルタ | |
WO2005015737A1 (ja) | システム推定方法及びプログラム及び記録媒体、システム推定装置 | |
JP4067269B2 (ja) | システム同定方法 | |
CN108141202B (zh) | 包括自适应模块和校正模块的分区块频域自适应滤波器设备 | |
Diniz et al. | Improved set-membership partial-update affine projection algorithm | |
JP6894402B2 (ja) | システム同定装置及び方法及びプログラム及び記憶媒体 | |
US11610598B2 (en) | Voice enhancement in presence of noise | |
CN113873090B (zh) | 一种稳健估计仿射投影样条自适应回声消除方法 | |
Cioffi | An unwindowed RLS adaptive lattice algorithm | |
JP2002533970A (ja) | 安定適応フィルタおよびその方法 | |
JPH1168518A (ja) | 係数更新回路 | |
Carlemalm et al. | On detection of double talk and changes in the echo path using a Markov modulated channel model | |
Gabrea | An adaptive Kalman filter for the enhancement of speech signals in colored noise | |
JPH08125587A (ja) | 適応ディジタル信号処理装置 | |
JP3303898B2 (ja) | 適応的伝達関数推定方法及びそれを使った推定装置 | |
Lippuner et al. | The Kalman filter in the context of adaptive filter theory | |
JP2001077730A (ja) | 適応フィルタの係数推定装置 | |
JPH07312535A (ja) | 適応的未知系出力推定方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200480022991.8 Country of ref document: CN |
|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2005513012 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2004771550 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2004771550 Country of ref document: EP |
|
DPEN | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 10567514 Country of ref document: US Ref document number: 2007185693 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 10567514 Country of ref document: US |