Random Noise Research Papers - Academia.edu (original) (raw)

7 Followers

Recent papers in Random Noise

The presence of ultradian rhythms in the mobility and behavior of rhesus monkeys was explored in eleven animals equipped with mobility FM transmitters and placed alone or in pairs in a laboratory cage situated in a sound proof,... more

The presence of ultradian rhythms in the mobility and behavior of rhesus monkeys was explored in eleven animals equipped with mobility FM transmitters and placed alone or in pairs in a laboratory cage situated in a sound proof, temperature controlled room, with food and water ad libitum. After an habituation period of 3 to 7 days, telemetric recordings of the mobility of the animals showed ultradian rhythms with a periodicity of 70 min during the daily 12-h light period. During the corresponding 12-h dark period, there were two dominant cycles of 103–144 min and 48 min respectively. Restriction of food and water to 1h/day did not change the duration of these cycles, and the presence of random noise also had little effect. Spontaneous behavior occurred in regular sequences within each cycle. Social relationships were found to affect rhythmicity of behavior, since pairs of monkeys synchronized their cycles. The lever press response recorded during auto-shaping showed the same rhythmicity found in mobility, demonstrating the influence of ultradian rhythms on learned behavior.

Factor analysis is a general purpose technique for dimension- ality reduction with applications in diverse areas including computer vision, collaborative filtering and computational bi- ology. Sparse factor analysis is a natural extension... more

Factor analysis is a general purpose technique for dimension- ality reduction with applications in diverse areas including computer vision, collaborative filtering and computational bi- ology. Sparse factor analysis is a natural extension that can be motivated by the observation that sparse features tend to generalize better, or justified based on a priori beliefs about the underlying generative model of the

Recent works have shown that an exponentially weighted moving average (EWMA) controller can be used on semiconductor processes to maintain process targets over extended periods for improved product quality and decreased machine downtime.... more

Recent works have shown that an exponentially weighted moving average (EWMA) controller can be used on semiconductor processes to maintain process targets over extended periods for improved product quality and decreased machine downtime. Proper choice of controller parameters (EWMA weights) is critical to the performance of this system. This work examines how different process factors affect the optimal controller parameters. We show that a function mapping from the disturbance state (magnitude of linear drift and random noise) of a given process to the corresponding optimal EWMA weights can be generated, and an artificial neural network (ANN) trained to learn the mapping. A self-tuning EWMA controller is proposed which dynamically updates its controller parameters by estimating the disturbance state and using the ANN function mapping to provide updates to the controller parameters. The result is an adaptive controller which eliminates the need for an experienced engineer to tune th...

The benefits of Hyperion hyperspectral data to agriculture have been studied at sites in the Coleambally Irrigation Area of Australia. Hyperion can provide effective measures of agricultural performance through the use of established... more

The benefits of Hyperion hyperspectral data to agriculture have been studied at sites in the Coleambally Irrigation Area of Australia. Hyperion can provide effective measures of agricultural performance through the use of established spectral indexes if systematic and random noise is managed. The noise management strategy includes recognition of "bad" pixels, reducing the effects of vertical striping, and compensation for

Eight different variable selection techniques for model-based and non-model-based clustering are evaluated across a wide range of cluster structures. It is shown that several methods have difficulties when non-informative variables (i.e.,... more

Eight different variable selection techniques for model-based and non-model-based clustering are evaluated across a wide range of cluster structures. It is shown that several methods have difficulties when non-informative variables (i.e., random noise) are included in the model. Furthermore, the distribution of the random noise greatly impacts the performance of nearly all of the variable selection procedures. Overall, a variable

This paper presents a method for tracking ground moving targets with a GMTI radar. To avoid detection by the GMTI radar, targets can deliberately stop for some time before moving again. The GMTI radar does not detect a target when the... more

This paper presents a method for tracking ground moving targets with a GMTI radar. To avoid detection by the GMTI radar, targets can deliberately stop for some time before moving again. The GMTI radar does not detect a target when the radial velocity (along the line-of-sight from the sensor) falls below a certain minimum detectable velocity (MDV). We develop a new approach by using state-dependent mode transition probabilities to track move-stop-move targets. Since in a real scenario, the maximum deceleration is always limited, a target cannot switch to the stopped-target model from a high speed. Therefore, with the use of the stopped-target model, the Markov chain of the mode switching has jump probabilities that depend on the target's kinematic state. A mode transition matrix with zero jump probabilities to the stopped-target mode is used when the speed is above a certain "stopping" limit (above which the target cannot stop in one sampling interval, designated as "fast stage") and another transition matrix with non-zero jump probabilities to the stopped-target mode is used when the speed is below this limit (designated as "slow stage"). The stage probabilities are calculated using the kinematic state statistics from the interacting multiple model (IMM) estimator and then used to combine the state-dependent mode transition probabilities (SDP) in the two different transition matrices. The experimental results show that the proposed algorithm outperforms previous methods.

A statistical approach to defect detection and disc rimination has been applied to the case of hot roll ed steel. The probability distribution of pixel intensities has been estimated from a small set of images without de fects, and this... more

A statistical approach to defect detection and disc rimination has been applied to the case of hot roll ed steel. The probability distribution of pixel intensities has been estimated from a small set of images without de fects, and this distribution is used to select pixels with unlikely values as candidates for defects. Discriminat ion of true defects from random noise pixels is achieved by a d ynamical thresholding procedure, which tracks the b ehaviour of clusters of selected pixels for varying threshol d level. Boundary levels of the dynamic threshold range are determined from the estimated probability distribut ion of the pixel intensities.

The benefits of EO-1 data, and especially Hyperion hyperspectral data, are being studied at sites in the Coleambally Irrigation Area of Australia where a seasonal time series has been developed. Hyperion can provide effective measures of... more

The benefits of EO-1 data, and especially Hyperion hyperspectral data, are being studied at sites in the Coleambally Irrigation Area of Australia where a seasonal time series has been developed. Hyperion can provide effective measures of agricultural performance through the use of spectral indices if systematic and random noise is managed and such noise management methods have been established for

Current processes validation methods rely on di- verse input states and exponential applications of state tomog- raphy. Through generalization of classical test theory exceptions to this rule are found. Instead of expanding a complete... more

Current processes validation methods rely on di- verse input states and exponential applications of state tomog- raphy. Through generalization of classical test theory exceptions to this rule are found. Instead of expanding a complete operator basis to validate a process, the objective is to utilize quan tum effects making each gate realized in the process act on a complete set of characteristic states and next extract functional in forma- tion. Random noise, systematic errors, initialization inaccuracies and measurement faults must also be detected. This concept is applied to the switching class comprising the search oracle. In a first approach, the test set cardinality is held constant to six; both testability and added depth complexity of an additional "design-for-test" circuit are related to the function real ized in the oracle. Oracles realizing affine functions are shown to gene rate no net entanglement and are thus the easiest to test, where oracles realizing bent functions are the most difficult to te st. A second approach replaces extraction complexity with a linear growth in experiment count. An interesting corollary of this study is the success found when addressing the classical test problem quantum mechanically. The validation of all classical degrees of freedom in a quantum switching network were found to necessitate exponentially fewer averaged observables than the number of tests in the classical lower bound.

We make use of recent results from random matrix theory to identify a derived threshold, for isolating noise from image features. The procedure assumes the existence of a set of noisy images, where denoising can be carried out on... more

We make use of recent results from random matrix theory to identify a derived threshold, for isolating noise from image features. The procedure assumes the existence of a set of noisy images, where denoising can be carried out on individual rows or columns independently. The fact that these are guaranteed to be correlated makes the correlation matrix an ideal tool for isolating noise. The random matrix result provides lowest and highest eigenvalues for the Gaussian random noise for which case, the eigenvalue distribution function is analytically known. This provides an ideal threshold for removing Gaussian random noise and thereby separating the universal noisy features from the non-universal components belonging to the specific image under consideration.

A methodological problem in applied clustering involves the decision of whether or not to standardize the input variables prior to the computation of a Euclidean distance dissimilarity measure. Existing results have been mixed with some... more

A methodological problem in applied clustering involves the decision of whether or not to standardize the input variables prior to the computation of a Euclidean distance dissimilarity measure. Existing results have been mixed with some studies recommending standardization and others suggesting that it may not be desirable. The existence of numerous approaches to standardization complicates the decision process. The present simulation study examined the standardization problem. A variety of data structures were generated which varied the intercluster spacing and the scales for the variables. The data sets were examined in four different types of error environments. These involved error free data, error perturbed distances, inclusion of outliers, and the addition of random noise dimensions. Recovery of true cluster structure as found by four clustering methods was measured at the correct partition level and at reduced levels of coverage. Results for eight standardization strategies are presented. It was found that those approaches which standardize by division by the range of the variable gave consistently superior recovery of the underlying cluster structure. The result held over different error conditions, separation distances, clustering methods, and coverage levels. The traditionalz-score transformation was found to be less effective in several situations.

ABSTRACT The study presents the analysis of detrended fluctuations (DFA) in inter spike intervals (ISI) of neuronal ensemble from cortex of awake behaving macaque monkeys. The original DFA method was applied to analyze fluctuation of... more

ABSTRACT The study presents the analysis of detrended fluctuations (DFA) in inter spike intervals (ISI) of neuronal ensemble from cortex of awake behaving macaque monkeys. The original DFA method was applied to analyze fluctuation of variances from fitted trends of different order. The spectrum of local scale exponent was calculated to investigate the presence of different scaling regions. It was observed that the single scaling exponent is insufficient to describe the firing pattern dynamics, the better fit is achieved using both short-term a1 and long-term α2 scaling coefficients. The validation procedure using phase randomized surrogates provided more reliable local scale exponents' estimates. Generalized DFA analysis revealed the presence of multifractality in ISI time series. Results indicate that multifractality is partly due to the broad probability distribution function and partly due to the presence of long-range correlations. Isodistributional surrogate data were used to test the significance of generalized Hurst exponent spectrum and origin of multifractal behavior. In conclusion, both DFA and its multifractal expansion reveal the presence of long-range correlation in ISI time series indicating the presence of memory in the neuronal firing pattern.

Log In