Academic literature on the topic 'Mean square Canny error'

In order to improve the performance of LMS (least mean square) algorithm by decreasing the amount of calculations this research proposes to make an update on each step only for those elements from the input data set, that fall within a small window W near the separating hyperplane surface. This work aims to describe in detail the results that can be achieved by using the proposed LMS with window learning algorithm in information systems that employ the methodology of neural network for the purposes of classification.

In this dissertation, new theoretical results are obtained for bounding convergence and mean-square error in conditional coding. Further new statistical methods for the practical application of conditional coding are developed. Criteria for the uniform convergence are first examined. Conditional coding Markov chains are aperiodic, π-irreducible, and Harris recurrent. By applying the general theories of uniform ergodicity of Markov chains on genera l state space, one can conclude that conditional coding Markov cha ins are uniformly ergodic and further, theoretical convergence rates based on Doeblin's condition can be found. Conditional coding Markov chains can be also viewed as having finite state space. This allows use of techniques to get bounds on the second largest eigenvalue which lead to bounds on convergence rate and the mean-square error of sample averages. The results are applied in two examples showing that these bounds are useful in practice. Next some algorithms for perfect sampling in conditional coding are studied. An application of exact sampling to the independence sampler is shown to be equivalent to standard rejection sampling. In case of single-site updating, traditional perfect sampling is not directly applicable when the state space has large cardinality and is not stochastically ordered, so a new procedure is developed that gives perfect samples at a predetermined confidence interval. In last chapter procedures and possibilities of applying conditional coding to mixture models are explored. Conditional coding can be used for analysis of a finite mixture model. This methodology is general and easy to use.

Speech corrupted by background noise (or noisy speech) can cause misinterpretation and fatigue during phone and conference calls, and for hearing aid users. Noisy speech can also severely impact the performance of speech processing systems such as automatic speech recognition (ASR), automatic speaker verification (ASV), and automatic speaker identification (ASI) systems. Currently, deep learning approaches are employed in an end-to-end fashion to improve robustness. The target speech (or clean speech) is used as the training target or large noisy speech datasets are used to facilitate multi-condition training. In this dissertation, we propose competitive alternatives to the preceding approaches by updating two classic robust speech processing techniques using deep learning. The two techniques include minimum mean-square error (MMSE) and missing data approaches. An MMSE estimator aims to improve the perceived quality and intelligibility of noisy speech. This is accomplished by suppressing any background noise without distorting the speech. Prior to the introduction of deep learning, MMSE estimators were the standard speech enhancement approach. MMSE estimators require the accurate estimation of the a priori signal-to-noise ratio (SNR) to attain a high level of speech enhancement performance. However, current methods produce a priori SNR estimates with a large tracking delay and a considerable amount of bias. Hence, we propose a deep learning approach to a priori SNR estimation that is significantly more accurate than previous estimators, called Deep Xi. Through objective and subjective testing across multiple conditions, such as real-world non-stationary and coloured noise sources at multiple SNR levels, we show that Deep Xi allows MMSE estimators to produce the highest quality enhanced speech amongst all clean speech magnitude spectrum estimators. Missing data approaches improve robustness by performing inference only on noisy speech features that reliably represent clean speech. In particular, the marginalisation method was able to significantly increase the robustness of Gaussian mixture model (GMM)-based speech classification systems (e.g. GMM-based ASR, ASV, or ASI systems) in the early 2000s. However, deep neural networks (DNNs) used in current speech classification systems are non-probabilistic, a requirement for marginalisation. Hence, multi-condition training or noisy speech pre-processing is used to increase the robustness of DNN-based speech classification systems. Recently, sum-product networks (SPNs) were proposed, which are deep probabilistic graphical models that can perform the probabilistic queries required for missing data approaches. While available toolkits for SPNs are in their infancy, we show through an ASI task that SPNs using missing data approaches could be a strong alternative for robust speech processing in the future. This dissertation demonstrates that MMSE estimators and missing data approaches are still relevant approaches to robust speech processing when assisted by deep learning.<br>Thesis (PhD Doctorate)<br>Doctor of Philosophy (PhD)<br>School of Eng & Built Env<br>Science, Environment, Engineering and Technology<br>Full Text

Microscopic traffic simulation is useful tool in analysing traffic and estimating the capacity and level of service of road networks. In this thesis, the four legged Idrottsparken roundabout in the city of Norrkoping in Sweden is analysed by using the microscopic traffic simulation package AIMSUN. For this purpose, data regarding traffic flow counts, travel times and queue lengths were collected for three consecutive weekdays during both the morning and afternoon peak periods. The data were then used in model building for simulation of traffic of the roundabout. The Root Mean Square Error (RMSE) method is used to get the optimal parameter value between queue length and travel time data and validation of travel time data are carried out to obtain the basic model which represents the existing condition of the system. Afterward, the results of the new models were evaluated and compared to the results of a SUMO model for the same scenario model. Based on calibrated and validated model, three alternative scenarios were simulated and analysed to improve efficiency of traffic network in the roundabout. The three scenarios includes: (1) add one free right turn in the north and east sections; (2) add one free right turn in the east and south sections; and (3) addition of one lane in roundabout. The analysis of these scenarios shows that the first and second scenario are only able to reduce the queue length and travel time in two or three legs, while the third scenario is not able to improve the performance of the roundabout. In this research, it can be concluded that the first scenario is considered as the best scenario compared to the second scenario and the third scenario. The comparison between AIMSUN and SUMO for the same scenario shows that the results have no significance differences. In calibration process, to get the optimal parameter values between the model measurements and the field measurements, both of AIMSUN and SUMO uses two significantly influencing parametersfor queue and travel time. AIMSUN package uses parameter of driver reaction time and the maximum acceleration, while SUMO package uses parameter of driver imperfection and also the driver rection time.

This paper investigates the effect of parameter selection for the decision feedback equalization (DFE) on communication performance through a dispersive underwater acoustic wireless channel (UAWC). A DFE based on minimum mean-square error (MMSE-DFE) criterion has been employed in the implementation for evaluation purposes. The output from the MMSE-DFE is input to the decoder to estimate the transmitted bit sequence. The main goal of this experimental simulation is to determine the best selection, such that the reduction in the computational overload is achieved without altering the performance of the system, where the computational complexity can be reduced by selecting an equalizer with a proper length. The system performance is tested for BPSK, QPSK, 8PSK and 16QAM modulation and a simulation for the system is carried out for Proakis channel A and real underwater wireless acoustic channel estimated during SPACE08 measurements to verify the selection.

ITC/USA 2014 Conference Proceedings / The Fiftieth Annual International Telemetering Conference and Technical Exhibition / October 20-23, 2014 / Town and Country Resort & Convention Center, San Diego, CA<br>Channel equalizers based on minimum mean square error (MMSE) and zero forcing (ZF) criteria have been formulated for a general scalable multiple input multiple output (MIMO) system and implemented for a 2x2 MIMO system with spatial multiplexing (SM) for Rayleigh channel associated with additive white Gaussian noise. A model to emulate transmitters and receivers on a spinning vehicle has been developed. A transceiver based on the BLAST architecture is developed in this work. A mathematical framework to explain the behavior of the ZF and MMSE equalizers is formulated. The performance of the equalizers has been validated for a case with one of the communication entities being a spinning aero vehicle. Performance analysis with respect to variation of angular separation between the antennas and relative antenna gain for each case is presented. Based on the simulation results a setup with optimal design parameters for placement of antennas, choice of the equalizers and transmit power is proposed.