The Experts below are selected from a list of 315 Experts worldwide ranked by ideXlab platform
S. C. Chan - One of the best experts on this subject based on the ideXlab platform.
-
A New Variable Forgetting Factor-Based Bias-Compensated RLS Algorithm for Identification of FIR Systems With Input Noise and Its Hardware Implementation
IEEE Transactions on Circuits and Systems I: Regular Papers, 2020Co-Authors: S. C. ChanAbstract:This paper proposes a new variable forgetting factor QRD-based recursive least squares algorithm with bias compensation (VFF-QRRLS-BC) for system identification under Input Noise. A new variable forgetting factor scheme is proposed to improve its convergence speed and steady-state mean squares error. A new method for recursive estimation of the additive Noise variance is also proposed for reliable bias compensation. The mean and mean-square asymptotic behaviors of the algorithm are analyzed and a self-calibration scheme is further proposed to improve the steady-state mean squares error (MSE) due to finite sample effect. Simulations show that the proposed VFF approach offers improved tracking and steady-state MSE performance over the conventional recursive least squares method and its fixed FF counterpart. A linear array architecture is proposed for the realization of this algorithm and several hardware efficient techniques are introduced to avoid the expensive cubic root and division operations required. The proposed algorithm is validated on Xilinx Zynq®-7000 AP SoC ZC702 Field Programmable Gate Array (FPGA). For a 10-tap finite impulse response (FIR) system, the implementation requires only about 11.5k slice look-up table (LUT)s, 4.5k slice registers and 50 DSP48s and it can work up to about 0.58 MHz sample rate with a 200 MHz system clock. The hardware resources are considerably lower than traditional techniques using divider and cubic root realization. The linear array architecture also serves as an attractive alternative to the systolic array in medium to low rate applications due to its reduced hardware usages.
-
A variable forgetting factor QRD-based RLS algorithm with bias compensation for system identification with Input Noise
2016 IEEE International Symposium on Circuits and Systems (ISCAS), 2016Co-Authors: S. C. Chan, L. ZhangAbstract:This paper proposes a variable forgetting factor QRD-based recursive least squares algorithm with bias compensation (VFF-QR-RLS-BC) for system identification with Input Noise. The new algorithm is based on the least square estimation with bias compensation framework and it employs a variable forgetting factor to improve the tracking speed and a QRD-based implementation for recursively solving the LS problem with bias compensation. Simulation results show that the proposed method can obtain improved convergence rate in sudden system change environment and satisfactory performance under stationary environment.
-
ISCAS - A variable forgetting factor QRD-based RLS algorithm with bias compensation for system identification with Input Noise
2016 IEEE International Symposium on Circuits and Systems (ISCAS), 2016Co-Authors: S. C. Chan, L. ZhangAbstract:This paper proposes a variable forgetting factor QRD-based recursive least squares algorithm with bias compensation (VFF-QR-RLS-BC) for system identification with Input Noise. The new algorithm is based on the least square estimation with bias compensation framework and it employs a variable forgetting factor to improve the tracking speed and a QRD-based implementation for recursively solving the LS problem with bias compensation. Simulation results show that the proposed method can obtain improved convergence rate in sudden system change environment and satisfactory performance under stationary environment.
L. Zhang - One of the best experts on this subject based on the ideXlab platform.
-
A variable forgetting factor QRD-based RLS algorithm with bias compensation for system identification with Input Noise
2016 IEEE International Symposium on Circuits and Systems (ISCAS), 2016Co-Authors: S. C. Chan, L. ZhangAbstract:This paper proposes a variable forgetting factor QRD-based recursive least squares algorithm with bias compensation (VFF-QR-RLS-BC) for system identification with Input Noise. The new algorithm is based on the least square estimation with bias compensation framework and it employs a variable forgetting factor to improve the tracking speed and a QRD-based implementation for recursively solving the LS problem with bias compensation. Simulation results show that the proposed method can obtain improved convergence rate in sudden system change environment and satisfactory performance under stationary environment.
-
ISCAS - A variable forgetting factor QRD-based RLS algorithm with bias compensation for system identification with Input Noise
2016 IEEE International Symposium on Circuits and Systems (ISCAS), 2016Co-Authors: S. C. Chan, L. ZhangAbstract:This paper proposes a variable forgetting factor QRD-based recursive least squares algorithm with bias compensation (VFF-QR-RLS-BC) for system identification with Input Noise. The new algorithm is based on the least square estimation with bias compensation framework and it employs a variable forgetting factor to improve the tracking speed and a QRD-based implementation for recursively solving the LS problem with bias compensation. Simulation results show that the proposed method can obtain improved convergence rate in sudden system change environment and satisfactory performance under stationary environment.
Michael D. Todd - One of the best experts on this subject based on the ideXlab platform.
-
On the Probability Structure of Output Noise From a Digital Phase Demodulation System Subject to Biased Intensity-Based Input Noise
Journal of Lightwave Technology, 2008Co-Authors: Michael D. ToddAbstract:Interferometry is a common technique used in fiber sensing that requires a demodulation algorithm to extract the signal of interest. Of great interest for sensor characterization is the performance of the demodulation scheme under the influence of Input Noise. Here we consider correlated, biased intensity Noise corrupting an interferometer. We analytically compute a probability density function of this output Noise only and use this to compute low-order statistical moments of the output Noise. We compare the analytical formulations with simulated data from a representative demodulation scheme used in a currently existing fiber Bragg grating sensor system and find excellent agreement within the example of Gaussian Input Noise.
-
Output-Noise Statistical Characterization for Digital-Phase-Demodulation Systems With Intensity-Based Input Noise
Journal of Lightwave Technology, 2007Co-Authors: Michael D. ToddAbstract:A large fraction of fiber-optic-sensor systems make use of interferometry in the measurement process, implying the need for demodulation to obtain the encoded signal of interest. Of great interest for sensor characterization is the performance of the demodulation scheme under the influence of Noise. In this paper, we construct a very general transfer function that relates intensity-induced Input Noise to its phase Noise realization in the output for a wide class of digital-demodulation techniques. We proceed to compute analytically a probability density function of this output Noise and use this to compute low-order statistical moments of the output Noise. We compare the analytical formulations with simulated data from a representative demodulation scheme used in a currently existing fiber Bragg grating (FBG) sensor system and find excellent agreement
Carl Edward Rasmussen - One of the best experts on this subject based on the ideXlab platform.
-
gaussian process training with Input Noise
Neural Information Processing Systems, 2011Co-Authors: Andrew Mchutchon, Carl Edward RasmussenAbstract:In standard Gaussian Process regression Input locations are assumed to be Noise free. We present a simple yet effective GP model for training on Input points corrupted by i.i.d. Gaussian Noise. To make computations tractable we use a local linear expansion about each Input point. This allows the Input Noise to be recast as output Noise proportional to the squared gradient of the GP posterior mean. The Input Noise variances are inferred from the data as extra hyperparameters. They are trained alongside other hyperparameters by the usual method of maximisation of the marginal likelihood. Training uses an iterative scheme, which alternates between optimising the hyperparameters and calculating the posterior gradient. Analytic predictive moments can then be found for Gaussian distributed test points. We compare our model to others over a range of different regression problems and show that it improves over current methods.
-
NIPS - Gaussian Process Training with Input Noise
2011Co-Authors: Andrew Mchutchon, Carl Edward RasmussenAbstract:In standard Gaussian Process regression Input locations are assumed to be Noise free. We present a simple yet effective GP model for training on Input points corrupted by i.i.d. Gaussian Noise. To make computations tractable we use a local linear expansion about each Input point. This allows the Input Noise to be recast as output Noise proportional to the squared gradient of the GP posterior mean. The Input Noise variances are inferred from the data as extra hyperparameters. They are trained alongside other hyperparameters by the usual method of maximisation of the marginal likelihood. Training uses an iterative scheme, which alternates between optimising the hyperparameters and calculating the posterior gradient. Analytic predictive moments can then be found for Gaussian distributed test points. We compare our model to others over a range of different regression problems and show that it improves over current methods.
A.t.k. Tang - One of the best experts on this subject based on the ideXlab platform.
-
a 3 spl mu v offset operational amplifier with 20 nv spl radic hz Input Noise psd at dc employing both chopping and autozeroing
International Solid-State Circuits Conference, 2002Co-Authors: A.t.k. TangAbstract:A 3 /spl mu/V offset op-amp has been designed using both autozeroing and chopping to give 20 nV//spl radic/Hz Input Noise at DC with low energy at the chopping frequency. The design includes additional circuitry for reduced switching transients. Power consumption is 4 mW from a 5 V supply. Die area is 0.6/spl times/1.12 mm using a 0.6 /spl mu/m double-poly double-metal CMOS process.
-
A 3 /spl mu/V-offset operational amplifier with 20 nV//spl radic/Hz Input Noise PSD at DC employing both chopping and autozeroing
2002 IEEE International Solid-State Circuits Conference. Digest of Technical Papers (Cat. No.02CH37315), 2002Co-Authors: A.t.k. TangAbstract:A 3 /spl mu/V offset op-amp has been designed using both autozeroing and chopping to give 20 nV//spl radic/Hz Input Noise at DC with low energy at the chopping frequency. The design includes additional circuitry for reduced switching transients. Power consumption is 4 mW from a 5 V supply. Die area is 0.6/spl times/1.12 mm using a 0.6 /spl mu/m double-poly double-metal CMOS process.