Focused Time Delay Neural Network Research Articles

Neuro-computational frameworks for non-linear stochastic wireless channels

Inter symbol interference and related distortions observed in high data rate wireless systems are mitigated using equalizers and channel estimation methods which are traditionally based on data driven tapped delay line (TDL) models. With increased fading due to mobility of transmit receive terminals, data driven schemes prove to be inefficient. Further, the TDL model assumes the dispersive channel to be linear in nature unlike the practical case which is non-linear, resulting in degraded performance. In such a situation, extrapolation of the channel state information is required which has been done by using artificial neural networks (ANN) optimized for non-linear channels. We explore a few neuro-computational approaches namely focused time delay neural network, feed forward ANN (FANN)–decision feedback equalizer (DFE) (FANN–DFE) and a functional link neural network while modeling non-linear wireless channels. Further, implementation of FANN–DFE based algorithm has been carried out using a specialized hardware framework to show the reduction of computational latency during training to establish its suitability for real time applications.

Neuro identification for flexible cantilever beam structure

In this study, system identification of a flexible beam which undergoes a small deflection due to the disturbance force was conducted. Recursive Least Square (RLS) and several intelligent Neuro identifications were established to acquire the dynamic model of the beam based on One Step-Ahead (OSA) prediction. Comparative assessment between conventional and intelligent methods was investigated in this study with the aim to find the best model that can represents the flexible beam. Neural Network (NN) can be classified according to their structures and learning algorithms. For the purpose of this study, Focused Time-Delay Neural Network (FTDNN), Distributed Time-Delay Neural Network (DTDNN), Nonlinear Autoregressive with Exogeneous Variables (NARX), Elman NN and Radial Basis Function (RBF) were compared their effectiveness in emulating the dynamic behaviour of the system. The performance of the network was observed based on its capability to represents the system with the lowest Mean- Squared-Error (MSE) between the actual and predicted output responses of the models.

Estimation and Forecast of Wind Power Generation by FTDNN and NARX-net based models for Energy Management Purpose in Smart Grids

This paper is focused on the prediction and forecast of climate time series, particularly useful for planning and management of the power grid, by artificial neural networks. An appropriate prediction and forecast of climate variables, indeed, improves the overall efficiency and performance of renewable power plants connected to the power grid. On such a basis, the application of suitable Artificial Neural Networks (ANNs) to the field of wind power generation is proposed. In particular, two dynamic recurrent ANNs, i.e., the Focused Time-Delay Neural Network (FTDNN) and the Nonlinear autoregressive network with exogenous inputs (NARX), are used to develop a model for the estimate and forecast of daily wind speed. Results, applied to a turbine model, allow the produced power to be calculated for energy management and planning purpose in smart grids.

Neural network approaches for noisy language modeling.

Text entry from people is not only grammatical and distinct, but also noisy. For example, a user's typing stream contains all the information about the user's interaction with computer using a QWERTY keyboard, which may include the user's typing mistakes as well as specific vocabulary, typing habit, and typing performance. In particular, these features are obvious in disabled users' typing streams. This paper proposes a new concept called noisy language modeling by further developing information theory and applies neural networks to one of its specific application-typing stream. This paper experimentally uses a neural network approach to analyze the disabled users' typing streams both in general and specific ways to identify their typing behaviors and subsequently, to make typing predictions and typing corrections. In this paper, a focused time-delay neural network (FTDNN) language model, a time gap model, a prediction model based on time gap, and a probabilistic neural network model (PNN) are developed. A 38% first hitting rate (HR) and a 53% first three HR in symbol prediction are obtained based on the analysis of a user's typing history through the FTDNN language modeling, while the modeling results using the time gap prediction model and the PNN model demonstrate that the correction rates lie predominantly in between 65% and 90% with the current testing samples, and 70% of all test scores above basic correction rates, respectively. The modeling process demonstrates that a neural network is a suitable and robust language modeling tool to analyze the noisy language stream. The research also paves the way for practical application development in areas such as informational analysis, text prediction, and error correction by providing a theoretical basis of neural network approaches for noisy language modeling.

Comparison of Four Neural Network Learning Methods Based on Genetic Algorithm for Non-linear Dynamic Systems Identification

Non-linear dynamical systems are difficult to control due to the model uncertainties and external disturbances that may occur in these systems. This paper addresses the problem of identification using dynamic neural networks (DNNs) based on genetic algorithm (GA) for nonlinear dynamic systems. Four different dynamic neural networks are used for identification of the same nonlinear dynamic system, using the genetic algorithm (GA) to train the LayerRecurrent Network (LRN), Focused Time-Delay Neural Network (FTDNN), the Elman Network, and Nonlinear Autoregressive Network with Exogenous inputs (NARX). The simulation results show the generalization ability of the four dynamic neural networks which provide the high precision of model of the nonlinear dynamic system. Also this paper illustrates the advantages and disadvantages of the different dynamic neural networks trained by GA.

Learning Location From Sequential Signal Strength Based on GSM Experimental Data

This study proposes two positioning algorithms, called the focused time delay neural network (FTDNN) and the distributed time delay neural network (DTDNN), to efficiently learn the mobile location from sequential received signal strength (RSS). By embedding the temporal structures of RSS into the spatial structures of networks, the proposed algorithms can extract location information from temporal variation of RSSs rather than removing them. This study applies the proposed algorithms to actual Global System for Mobile communications (GSM) networks, collecting realistic RSS measurements across a campus using commercially available mobile phones. On-site experiments clearly demonstrated that both algorithms provide better positioning accuracy than the traditional Cell-ID method, the Bayesian approach, and the multilayer perceptron with variations, where DTDNN achieves the best performance.

An Artificial Intelligence-based language modeling framework

The text entry interaction between human and computer could be noisy. The typing stream is a reflection of users’ typing behavior that includes ones’ fluency of vocabulary, typing habits and typing performance. This research develops an original intermediate layer language modeling framework called Adaptive Language Modeling Intermediate Layer (ALMIL), which can be seen as a communication language layer between human and computer to analyze noisy language stream data and provide users with Text Prediction and Text Correction functions. Then, the ALMIL framework is applied to QWERTY keyboard producing an Intelligent Keyboard hybrid framework, which can be used to analyze users’ typing stream, and accordingly correct typing mistakes and predict users typing intention. An extendable Focused Time-Delay Neural Network (FTDNN) n-gram prediction algorithm is also developed to learn from the users’ typing history and produce text entry prediction and correction based on historical typing stream. Overall, this research pioneers a comprehensive neural network language modeling process based on cross-experiments between extendible inputs, hidden neurons and noise rates, while statistics based traditional methods have failed to consider possible noises of a language stream. The Computer Simulation Results demonstrate that the proposed Intelligent Keyboard performs better than the conventional keyboard in Text Prediction and Text Correction. In practice, a higher prediction rate could be achieved by combining the First Three Hitting Rate with an English word dictionary.

Creep Predicting Model in Masonry Structure Utilizing Dynamic Neural Network

Problem statement: When the loads are applied to a brickwork structure, visco-elastic behavior upon their stress-strain relationships is exhibited, where the response can be classified into two separate parts: an instantaneous elastic strains and time-dependent creep strains. The creep strain represents the non- instantaneous strain that happens with time when the stress is sustained. Through the previous century, along with the alter in brickwork construction, A chain of creep tests on brickwork has shown that creep in brickwork be able to result in deformation that rise gradually with the way of time. Brickwork has considerable creep strain that is complicated to predict because of its reliance on several unrestrained parameters (e.g., relative humidity, time of load application, stress level). Dependable and precise prediction models for the long term, time-dependent creep deformation of brickwork structures are required. Artificial Neural Network (ANN) models have been determined useful and efficient especially in such problems for which the characteristics of the processes are difficult to describe using numerical models. Approach: This study introduces a creep prediction model based Focused Time-Delay Neural Network (FTDNN) which could detect and consider within its architecture the time dependency which is major factor in creep deformation in brickwork structure. Results: Performance of the proposed FTDNN model was examined with experimental creep data from brickwork assemblages collected over the last 15 years. Results showed that the FTDNN model has a relatively small prediction error compared to the other models with the error less than 15%. Conclusion: The results showed that the FTDNN model outperformed the existing ANN models and significantly enhance the accuracy of creep prediction.