Input Variance Research Articles

Speech Signal Coding Using Forward Adaptive Quantization and Simple Transform Coding

The paper proposes a novel speech signal coding scheme that implements a simple transform coding and forward adaptive quantization. The proposed scheme is adapted to the input signal variance, providing highly efficient bandwidth usage, whereas implemented transform coding provides sub-sequences with more predictable signal characteristics, so that more suitable signal processing can be performed. The aforementioned transform coding precedes adaptive quantization, providing additional compression. The objective quality measure used for system performance estimation is SQNR (signal-to-quantization-noise ratio), which represents a standard measure for lossy coding types. The influence of transform coding is discussed by comparing the obtained results with the corresponding one achieved by applying only the same adaptive quantization. Furthermore, the comparison with system performance of PCM (pulse-code modulation) coding system confirms that the proposed coding scheme has a lot of potential for further implementation, since that the proposed system ensures SQNR gain up to 4.0983 [dB] for various values of system parameters. DOI: http://dx.doi.org/10.5755/j01.eie.22.3.15318

Adaptive estimation algorithm of boost-phase trajectory using binary asynchronous observation

A kind of adaptive filter algorithm based on the estimation of the unknown input is proposed for studying the adaptive adjustment of process noise variance of boost phase trajectory. Polynomial model is used as the motion model of the boost trajectory, truncation error is regarded as an equivalent to the process noise and the unknown input and process noise variance matrix is constructed from the estimation value of unknown input according to the quantitative relationship among the unknown input, the state estimation error, and optimal process noise variance. The simulation results show that in the absence of prior information, the unknown input is estimated effectively in terms of magnitude, a positive definite matrix of process noise covariance which is close to the optimal value is constructed real-timely, and the state estimation error approximates the error lower bound of the optimal estimation. The estimation accuracy of the proposed algorithm is similar to that of the current statistical model algorithm using accurate prior information.

Efficient variance reduction approach based on the variation of the input importance

ABSTRACTTo reduce the output variance, the variance-based importance analysis can provide an efficient way by reducing the variance of the ‘important’ inputs. But with the reduction of the variance of those ‘important’ inputs, the input importance will change and it is no longer the most efficient way to reduce the variance of those ‘important’ inputs alone. Thus, analyst needs to consider reducing the variance of other inputs to obtain a more efficient way. This work provides a graphical solution for analyst to decide how to reduce the input variance to achieve the targeted reduction of the output variance efficiently. Furthermore, by the importance sampling-based approach, the graphical solution can be obtained with only a single group of samples, which can decrease the computational cost greatly.

Nonlinear System Identification with a Real–Coded Genetic Algorithm (RCGA)

Abstract This paper is devoted to the blind identification problem of a special class of nonlinear systems, namely, Volterra models, using a real-coded genetic algorithm (RCGA). The model input is assumed to be a stationary Gaussian sequence or an independent identically distributed (i.i.d.) process. The order of the Volterra series is assumed to be known. The fitness function is defined as the difference between the calculated cumulant values and analytical equations in which the kernels and the input variances are considered. Simulation results and a comparative study for the proposed method and some existing techniques are given. They clearly show that the RCGA identification method performs better in terms of precision, time of convergence and simplicity of programming.

Quantity and Quality of Caregivers' Linguistic Input to 18-Month and 3-Year-Old Children Who Are Hard of Hearing.

The primary objective of this study was to examine the quantity and quality of caregiver talk directed to children who are hard of hearing (CHH) compared with children with normal hearing (CNH). For the CHH only, the study explored how caregiver input changed as a function of child age (18 months versus 3 years), which child and family factors contributed to variance in caregiver linguistic input at 18 months and 3 years, and how caregiver talk at 18 months related to child language outcomes at 3 years. Participants were 59 CNH and 156 children with bilateral, mild-to-severe hearing loss. When children were approximately 18 months and/or 3 years of age, caregivers and children participated in a 5-min semistructured, conversational interaction. Interactions were transcribed and coded for two features of caregiver input representing quantity (number of total utterances and number of total words) and four features representing quality (number of different words, mean length of utterance in morphemes, proportion of utterances that were high level, and proportion of utterances that were directing). In addition, at the 18-month visit, parents completed a standardized questionnaire regarding their child's communication development. At the 3-year visit, a clinician administered a standardized language measure. At the 18-month visit, the CHH were exposed to a greater proportion of directing utterances than the CNH. At the 3-year visit, there were significant differences between the CNH and CHH for number of total words and all four of the quality variables, with the CHH being exposed to fewer words and lower quality input. Caregivers generally provided higher quality input to CHH at the 3-year visit compared with the 18-month visit. At the 18-month visit, quantity variables, but not quality variables, were related to several child and family factors. At the 3-year visit, the variable most strongly related to caregiver input was child language. Longitudinal analyses indicated that quality, but not quantity, of caregiver linguistic input at 18 months was related to child language abilities at 3 years, with directing utterances accounting for significant unique variance in child language outcomes. Although caregivers of CHH increased their use of quality features of linguistic input over time, the differences when compared with CNH suggest that some caregivers may need additional support to provide their children with optimal language learning environments. This is particularly important given the relationships that were identified between quality features of caregivers' linguistic input and children's language abilities. Family supports should include a focus on developing a style that is conversational eliciting as opposed to directive.

Scalability of Asynchronous Networks Is Limited by One-to-One Mapping between Effective Connectivity and Correlations.

Network models are routinely downscaled compared to nature in terms of numbers of nodes or edges because of a lack of computational resources, often without explicit mention of the limitations this entails. While reliable methods have long existed to adjust parameters such that the first-order statistics of network dynamics are conserved, here we show that limitations already arise if also second-order statistics are to be maintained. The temporal structure of pairwise averaged correlations in the activity of recurrent networks is determined by the effective population-level connectivity. We first show that in general the converse is also true and explicitly mention degenerate cases when this one-to-one relationship does not hold. The one-to-one correspondence between effective connectivity and the temporal structure of pairwise averaged correlations implies that network scalings should preserve the effective connectivity if pairwise averaged correlations are to be held constant. Changes in effective connectivity can even push a network from a linearly stable to an unstable, oscillatory regime and vice versa. On this basis, we derive conditions for the preservation of both mean population-averaged activities and pairwise averaged correlations under a change in numbers of neurons or synapses in the asynchronous regime typical of cortical networks. We find that mean activities and correlation structure can be maintained by an appropriate scaling of the synaptic weights, but only over a range of numbers of synapses that is limited by the variance of external inputs to the network. Our results therefore show that the reducibility of asynchronous networks is fundamentally limited.

Energy Saving in Five-Axis Machine Tools Using Synchronous and Contouring Control and Verification by Machining Experiment

This paper deals with a controller design for five-axis machine tools, which typically consist of three translational and two rotary axes. The flexibility to change the relative orientation between a tool and the workpiece in a five-axis machine tool offers many advantages over conventional 3-D machining. Because machine tools operate around the clock in industrial applications all over the world, there is a requirement not only for high-speed and high-precision operation but also for reduced energy consumption. This paper presents a novel controller design that aims to reduce energy consumption in five-axis machine tools. The proposed design is based on a contouring controller that has been proposed for use in high-precision machining in machine tool systems. The design also takes into consideration the relative orientation error between a tool and the workpiece in five-axis machining. The synchronous controller is used to reduce synchronous error between two parallel actuators, which is a cause of significant mechanical damage. The contouring control system has a degree of freedom (DOF) in the direction tangential to the desired contour curve because this directional error does not directly affect the contouring performance, and hence, this DOF is employed for consumed energy reduction in this study. The experimental results of feed drive motion (without machining) demonstrate the effectiveness of the proposed controller, which reduces energy consumption by 13.2%. In addition, after applying to the machining experiment, the effectiveness was increased to 27.9% because of largely reducing the control input variance caused by the machining task.

Assessment of wind energy potential for selecting wind turbines: An application to Thailand

Wind energy assessment plays a critical role on determining installation of wind turbines in many countries worldwide. Wind analysis and economic analysis are proposed to systematically assess the potential of electrical energy production for commercial wind turbines. Statistical information of wind data and power curves of wind turbines are applied for determination of factors in decision making such as annual energy production and capacity factor. Accordingly, financial indices and reductions of both fossil fuel consumption and carbon dioxide generation are examined for installing potential wind turbines. The uncertainties of financial indices are determined under variances of parametric inputs within confidence intervals. Case studies are investigated in the central region of Thailand, which has a high demand for electricity. Annual mean wind speed is between 3m/s and 5m/s along directions of tropical monsoons. With the proposed methodology, a wind turbine with low cut-in wind speed, such as the Vestas™ V60 850kW, can be recommended as one of the most feasible wind turbines for electricity generation in green economy under weak and moderate wind conditions in the considered sites. The annual energy production has the most impact on the uncertainty in the net present value, among other financial parameters.

Energy Saving in Feed Drive Systems Using Sliding-Mode-Based Contouring Control With a Nonlinear Sliding Surface

Reduction of contour error, which is defined as the shortest distance between the actual position of the cutting tool and the reference trajectory, is an essential requirement in machining by multiaxis feed drive systems. In addition, because these machines operate all the time in industrial applications all over the world, reduction of the consumed energy in these machines contributes to environmental, natural resources, and energy problems. This paper presents a novel sliding mode contouring controller with a nonlinear sliding surface to improve the machining accuracy of the biaxial feed drive systems. Because the contour error is more important than the tracking error with respect to each feed drive axis, the contour error component is included in the proposed sliding surface. The advantage of including the contour error in the proposed sliding surface is that the damping ratio of the system changes from its initial low value to final high value as the contour error changes from high value to small value and vice versa. Hence, the proposed algorithm allows a closed-loop system to simultaneously achieve low energy consumption and small settling time, resulting in a smaller error. By using the proposed method, experimental results for a biaxial feed drive system show a significant performance improvement in terms of the contour error. In addition, the proposed approach reduces the control input variance by about 41.2 % and 14.9 % for x - and y-axis, respectively, and consumed energy by about 23.7 % and 5.5 % for the respective axis from the conventional method with a linear sliding surface.

Simulation-based baggage handling system merge analysis

This paper details the investigation into the design and control of merging bottlenecks of conveyor-based baggage handling systems, encompassing the merging control algorithm and the impact of the merge’s physical layout. A methodology for the analysis of simulation model results is presented. Results show that the layout of the merge influences bag throughput and when the physical configuration is in a preferred position, input variance has no effect on bag throughput performance. These results have potential application to other material handling systems, such as those used in manufacturing and warehousing.

Combined outputs variance constrained and input variance constrained design for flight control

In this article, an algorithm combining two different variance constrained control strategies (i.e. output variance constrained: OVC controller and input variance constrained: IVC controller) is developed. Performance of the resulting novel control technique called as combined output and input variance constrained controller is evaluated on helicopters. Their linearized state-space models are used to see feasibility and efficiency of this new strategy (i.e., output and input variance constrained). Closed loop responses of some outputs and inputs of interest and also some outputs out of interest are investigated while using output and input variance constrained for air vehicle flight control system. Then, closed loop performance of output and input variance constrained is compared with respect to the two other classical control techniques (i.e. outputs variance constrained and input variance constrained). Finally, robustness of output and input variance constrained with respect to the modeling uncertainities is examined.

Design of a Smart Adaptive Control System

In industrial processes, it is necessary to maintain the user-specified control performance in order to achieve desired productivity. This paper describes a design scheme of smart adaptive controller based on the above strategy. In the proposed method, variance of control error and input are evaluated on-line. Moreover, control parameters are adjusted only when the user-specified control performance can not be obtained. Control parameters are calculated directly from closed-loop data and they are adjusted by 1-parameter tuning. The effectiveness of the proposed method is verified by using a simulation example and experiment of temperature control system.

Gaussian Process Model Based Performance Assessment of Control Loop

Practical constraints on control systems such as valve constraints imply that performance of minimum variance control is most likely not achievable. Instead, the objective may be expressed as minimizing a linear quadratic function of input and output variances instead. The evaluated performance index is dependent on the model and in terms of accuracy of the model it is only as accurate as the quality of the data available. This work proposed a Gaussian process model indicator to provide a trust region on the performance index and to provide a framework of judging what improvement is feasible. The model can be updated adaptively and in turn the performance. Based on the indication of the model quality the reliability of the evaluated controller performance index can be obtained.

Mean-Square Stability Analysis of a Normalized Least Mean Fourth Algorithm for a Markov Plant

Recently, it has been shown that the stability of the least mean fourth (LMF) algorithm depends on the nonstationarity of the plant. The present paper investigates the possibility of overcoming this problem by normalization of the weight vector update term. A rigorous mean-square stability analysis is provided for a recent normalized LMF algorithm, which is normalized by a term that is second order in the estimation error and fourth order in the regressor. The analysis is done for a Markov plant with a stationary white input with even probability density function and a stationary zero-mean white noise. It is proved that the mean-square deviation (MSD) of the algorithm is bounded for all finite values of the input variance, noise variance, initial MSD, and mean-square plant parameter increment. Analytical results are supported by simulations.

Breeding objectives for pigs in Kenya. II: economic values incorporating risks in different smallholder production systems.

This study estimated economic values for production traits (dressing percentage (DP), %; live weight for growers (LWg), kg; live weight for sows (LWs), kg) and functional traits (feed intake for growers (FEEDg), feed intake for sow (FEEDs), preweaning survival rate (PrSR), %; postweaning survival (PoSR), %; sow survival rate (SoSR), %, total number of piglets born (TNB) and farrowing interval (FI), days) under different smallholder pig production systems in Kenya. Economic values were estimated considering two production circumstances: fixed-herd and fixed-feed. Under the fixed-herd scenario, economic values were estimated assuming a situation where the herd cannot be increased due to other constraints apart from feed resources. The fixed-feed input scenario assumed that the herd size is restricted by limitation of feed resources available. In addition to the tradition profit model, a risk-rated bio-economic model was used to derive risk-rated economic values. This model accounted for imperfect knowledge concerning risk attitude of farmers and variance of input and output prices. Positive economic values obtained for traits DP, LWg, LWs, PoSR, PrSR, SoSR and TNB indicate that targeting them in improvement would positively impact profitability in pig breeding programmes. Under the fixed-feed basis, the risk-rated economic values for DP, LWg, LWs and SoSR were similar to those obtained under the fixed-herd situation. Accounting for risks in the EVs did not yield errors greater than ±50% in all the production systems and basis of evaluation meaning there would be relatively little effect on the real genetic gain of a selection index. Therefore, both traditional and risk-rated models can be satisfactorily used to predict profitability in pig breeding programmes.

Dynamics of self-sustained asynchronous-irregular activity in random networks of spiking neurons with strong synapses.

Random networks of integrate-and-fire neurons with strong current-based synapses can, unlike previously believed, assume stable states of sustained asynchronous and irregular firing, even without external random background or pacemaker neurons. We analyze the mechanisms underlying the emergence, lifetime and irregularity of such self-sustained activity states. We first demonstrate how the competition between the mean and the variance of the synaptic input leads to a non-monotonic firing-rate transfer in the network. Thus, by increasing the synaptic coupling strength, the system can become bistable: In addition to the quiescent state, a second stable fixed-point at moderate firing rates can emerge by a saddle-node bifurcation. Inherently generated fluctuations of the population firing rate around this non-trivial fixed-point can trigger transitions into the quiescent state. Hence, the trade-off between the magnitude of the population-rate fluctuations and the size of the basin of attraction of the non-trivial rate fixed-point determines the onset and the lifetime of self-sustained activity states. During self-sustained activity, individual neuronal activity is moreover highly irregular, switching between long periods of low firing rate to short burst-like states. We show that this is an effect of the strong synaptic weights and the finite time constant of synaptic and neuronal integration, and can actually serve to stabilize the self-sustained state.

Improving the approximation ability of Volterra series identified with a cross-correlation method

This paper proposes an improvement in cross-correlation methods derived from the Lee–Schetzen method, in order to obtain a lower mean square error in the output for a wider range of the input variances. In particular, each Wiener kernel is identified with a different input variance and new formulas for conversion from Wiener to Volterra representation are presented.

Context-dependent coding in single neurons.

The linear-nonlinear cascade model (LN model) has proven very useful in representing a neural system's encoding properties, but has proven less successful in reproducing the firing patterns of individual neurons whose behavior is strongly dependent on prior firing history. While the cell's behavior can still usefully be considered as feature detection acting on a fluctuating input, some of the coding capacity of the cell is taken up by the increased firing rate due to a constant "driving" direct current (DC) stimulus. Furthermore, both the DC input and the post-spike refractory period generate regular firing, reducing the spike-timing entropy available for encoding time-varying fluctuations. In this paper, we address these issues, focusing on the example of motoneurons in which an afterhyperpolarization (AHP) current plays a dominant role regularizing firing behavior. We explore the accuracy and generalizability of several alternative models for single neurons under changes in DC and variance of the stimulus input. We use a motoneuron simulation to compare coding models in neurons with and without the AHP current. Finally, we quantify the tradeoff between instantaneously encoding information about fluctuations and about the DC.

Minimum Input Variances for Modelling Rainfall-runoff Using ANN

This paper presents the study of possible input variances for modeling the long-term runoff series using artificial neural network (ANN). ANN has the ability to derive the relationship between the inputs and outputs of a process without the physics being provided to it, and it is believed to be more flexible to be used compared to the conceptual models [1]. Data series from the Kurau River sub-catchment was applied to build the ANN networks and the model was calibrated using the input of rainfall, antecedent rainfall, temperature, antecedent temperature and antecedent runoff. In addition, the results were compared with the conceptual model, named IHACRES. The study reveal that ANN and IHACRES can simulate well for mean runoff but ANN gives a remarkable performance compared to IHACRES, if the model customizes with a good configuration.

Propagation-of-uncertainty from contact angle and streaming potential measurements to XDLVO model assessments of membrane–colloid interactions

Membrane fouling assessed from a fundamental standpoint within the context of the Derjaguin–Landau–Verwey–Overbeek (DLVO) model. The DLVO model requires that the properties of the membrane and foulant(s) be quantified. Membrane surface charge (zeta potential) and free energy values are characterized using streaming potential and contact angle measurements, respectively. Comparing theoretical assessments for membrane–colloid interactions between research groups requires that the variability of the measured inputs be established. The impact that such variability in input values on the outcome from interfacial models must be quantified to determine an acceptable variance in inputs. An interlaboratory study was conducted to quantify the variability in streaming potential and contact angle measurements when using standard protocols. The propagation of uncertainty from these errors was evaluated in terms of their impact on the quantitative and qualitative conclusions on extended DLVO (XDLVO) calculated interaction terms. The error introduced into XDLVO calculated values was of the same magnitude as the calculated free energy values at contact and at any given separation distance. For two independent laboratories to draw similar quantitative conclusions regarding membrane–foulant interfacial interactions the standard error in contact angle values must be⩽2.5°, while that for the zeta potential values must be⩽7mV.