C24. Automatic modulation recognition in wireless systems using cepstral analysis and neural networks

Automatic modulation recognition is a rapidly evolving area of signal analysis. In recent years, interest from the academic and military research institutes has focused around the research and development of modulation recognition algorithms. Any communication intelligence (COMINT) system comprises three main blocks: receiver front-end, modulation recogniser and output stage. Considerable work has been done in the area of receiver front-ends. The work at the output stage is concerned with information extraction, recording and exploitation and begins with signal demodulation, that requires accurate knowledge about the signal modulation type. There are, however, two main reasons for knowing the current modulation type of a signal; to preserve the signal information content and to decide upon the suitable counter action, such as jamming. Automatic Modulation Recognition of Communications Signals describes in depth this modulation recognition process. Drawing on several years of research, the authors provide a critical review of automatic modulation recognition. This includes techniques for recognising digitally modulated signals. The book also gives comprehensive treatment of using artificial neural networks for recognising modulation types. Automatic Modulation Recognition of Communications Signals is the first comprehensive book on automatic modulation recognition. It is essential reading for researchers and practising engineers in the field. It is also a valuable text for an advanced course on the subject.

Automatic modulation recognition is very important for communication intelligence applications such as signal interception for satellite communications, civil authority, surveillance and threat analysis. Automatic digital modulation recognition is a system that automatically classifies the modulation type of the intercepted signal. This paper presents a method for the automatic recognition of digital modulations without a priori knowledge of the signal parameters. This algorithm works at low signal to noise ratio and get nearly to 100 % of correct classification percentage at 0 dB which is effective in Electronic warfare and communication surveillance applications. This method can recognize common single-carrier modulated satellite communication signals such as M-ary phase-shift keying, M-ary quadrature amplitude modulation and M-ary amplitude phase shift keying which used in DVB-S2. This algorithm is performed by a sequential steps to blindly classify the signal using PSD of instantaneous amplitude, kurtosis and HOS (cumulants).

Automatic Digital Modulation Recognition (ADMR) is becoming an interesting problem with various civil and military applications. In this paper, an ADMR algorithm in Multi-Carrier Code Division Multiple Access (MC-CDMA) systems using Discrete Transforms (DTs) and Mel-Frequency Cepstral Coefficients (MFCCs) is proposed. This algorithm uses various DT techniques such as the Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT) and Discrete Sine Transform (DST) with MFCCs to extract features from the modulated signal and a Support Vector Machine (SVM) to classify the modulation orders. The proposed algorithm avoids over fitting and local optimal problems that appear in Artificial Neural Networks (ANNs). Simulation results shows the classifier is capable of recognizing the modulation scheme with high accuracy up to 90% - 100% using DWT, DCT and DST for some modulation schemes over a wide Signal-to-Noise Ratio (SNR) range in the presence of Additive White Gaussian Noise (AWGN) and Rayleigh fading channel, particularly at a low Signal-to-Noise ratios (SNRs).

Wavelet transform-based methodologies for both Automatic Modulation Recognition (AMR) and Demodulation of digitally modulated communications signals can be utilized in an enabling platform for the implementation of a new class of communications systems. In particular, such techniques could enable the development of agile radio transceivers for use in both commercial and military applications. Such radio transceivers would have the ability to transmit and receive signals using many different modulation schemes while employing a common receiver architecture based on a single demodulator. In this dissertation, the development of AMR and Demodulation techniques are based on the relatively new mathematical theory of Wavelet Transforms (WTs). Information-bearing signals acquired by communications receivers are transformed into the wavelet-domain using the Continuous Wavelet Transform (CWT) and then applied to signal processing algorithms that also use the CWT in conjunction with pattern recognition techniques. In particular, the method of template-matching is used for both the AMR and Demodulation processes. Signal templates characterizing various modulated signals are used for both processes. The signal templates are determined based on the signal features present in the fractal patterns of their corresponding scalograms for specific modulation schemes as they appear in the wavelet-domain. The algorithms developed in this work are capable of both classifying the method of modulation used in the acquired signal, as well as subsequently automatically demodulating the signal to recover the message. The classes of digitally modulated signals considered in this work include variants of the Amplitude-, Frequency-, Phase-Shift Keying modulation families, i.e., ASK, FSK, and PSK, respectively, and multiple-level Quadrature Amplitude Modulation (M-ary QAM) families. The AMR and Demodulation performances are evaluated in the presence of Additive White Gaussian Noise (AWGN) over a wide range of Signal-to-Noise Ratio (SNR) values. Through extensive Monte Carlo computer simulations it is determined that the average correct classification rates using wavelet-based AMR for PSK, ASK, and QAM are over 98%, and over 90% for FSK signals, all at an SNR of 0 dB. The Bit Error Rate (BER) performance obtained using wavelet-based Demodulation is at least one order of magnitude better than the matched filter-based BER performance realized for the modulation schemes considered.

This paper aims to perform modeling of Taiwanese farm and ecotourism data using compositional models as a probabilistic approach and to compare its results with the performance of an artificial neural network approach. Authors use probabilistic compositional models together with the artificial neural network as a classifier and compare the accuracy of both approaches. The probabilistic model structure is learned using hill climbing algorithm, and the weights of multilayer feedforward artificial neural network are learned using an R implementation of H2O library for deep learning. In case of both approaches, we employ a non-exhaustive cross-validation method and compare the models. The comparison is augmented by the structure of the compositional model and basic characterization of artificial neural network. As expected, the compositional models show significant advantages in interpretability of results and (probabilistic) relations between variables, whereas the artificial neural network provides more accurate yet “black-box” model.

Automatic modulation recognition (AMR) stands as a crucial core technology within the realm of signal processing and perception, playing a significant part in harsh electromagnetic environments. The time–frequency image (TFI) of communication signals can manifest modulation characteristics and serve as a foundation for signal modulation recognition and classification. However, under the influence of the electromagnetic environment, communication signals are exposed to varying degrees of interference, which poses a challenge to the recognition of modulation types. Taking into account the effects of interference and channel fading, this paper introduces a communication signal modulation recognition algorithm based on deep learning (DL) and time–frequency analysis. This approach employs short-time Fourier transform (STFT) to generate time–frequency diagrams from time-domain signals. Subsequently, it binarizes the image and feeds it as input data to the neural network. Our research presents a composite deep convolutional neural network (CNN) architecture known as the composite dense-residual neural network (CDRNN). This architecture focuses on enhancing the feature extraction and identification, aiming to achieve accurate recognition of modulation types in harsh electromagnetic environments. Finally, simulation results validate that the proposed deep learning algorithm holds remarkable advantages in boosting the accuracy of modulation type recognition with better adaptability. The algorithm shows better performance even in harsh electromagnetic environments. When the signal-to-noise ratio (SNR) is 18 dB, the recognition accuracy can reach 92.1%.

An approach based on artificial neural networks is proposed for the analysis of electromechanical modes in Taiwan power system. To evaluate the dynamic performance of a power system in system operation and planning, the dominant eigenvalues for the worst-damped electromechanical mode must be computed. A multilayer feedforward artificial neural network is developed. It is well known that eigenvalues are complicated functions of many system variables such as bus loads, line flows, generation schedule, bus voltages etc. An important procedure in neural network design is to select those features which most affect system eigenvalues. A clustering artificial neural network is thus designed for feature selection. To demonstrate the effectiveness of the approach, results from eigenvalue analyses of Taiwan power system are reported.

This paper presents a method for incorporating a priori information about an uncertain nonlinear system into the structure of a multilayer feedforward artificial neural network. Known information is incorporated into the activation function of the network output layer. An algorithm is derived for backpropagating the error and updating adjustable parameters within this layer that is consistent with existing supervised learning techniques. The developed technique is applied to the identification of a dynamic system and compared with conventional feedforward artificial neural network identifier. Results exhibit an improvement in the quality of the identification model and an increase in the rate of convergence. As a practical application, a prior information is utilized for identification of switched reluctance motor characteristics on the basis of experimental measurements. The results further demonstrate that artificial neural networks employing a priori information converge faster, require fewer adjustable weights, and more accurately predict the system of interest.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;Heat treated cast aluminum components like engine blocks and cylinder heads can develop significant amount of residual stress and distortion particularly with water quench. To incorporate the influence of residual stress and distortion in cast aluminum product design, a rapid simulation approach has been developed based on artificial neural network and component geometry characteristics. Multilayer feed-forward artificial neural network (ANN) models were trained and verified using FEA residual stress and distortion predictions together with part geometry information such as curvature, maximum dihedral angle, topologic features including node's neighbors, as well as quench parameters like quench temperature and quench media.&lt;/div&gt;&lt;/div&gt;

An inverse analysis using artificial intelligence based on the guided ultrasonic waves is proposed for effective identification of damage in thick steel beams for the purpose of structural health monitoring applications. Parameterized modeling for finite element analysis is applied to constitute the damage parameter database cost-effectively. For signal processing and feature extraction, wavelet transform is employed. A novel feature extraction technique, damage characteristic points, is applied to constitute the database for pattern recognition procedures. Using the extracted metrics, a multilayer feedforward artificial neural network under supervision of an error-backpropagation algorithm is developed and trained. The generalization performance of the artificial neural network has been examined experimentally. Results illustrate that the proposed metrics together with artificial neural network technique are powerful tools for effective identification of damage in the case of thick structures.

This paper is concerned with optical character recognition (OCR) system for Bangla conjunctive characters. A method is proposed giving emphasis on the identification of the characters using the proposed methodology. Here generalization is achieved by preprocessing the characters before presenting them to the system for classification. The pre-processing systematically functions isolating the characters from BMP images, as well as noise removal, scaling and binary image conversion. The method uses a flexible matching between sample data and training data components applying Multi-layered Feed-forward Artificial Neural Network. A number of parameters of Neural Network are estimated so that the system performance is improved in comparison with normal training. Classification of character is defined as error minimization among the possible training set. Also, a measure of the amount of distortion for this training is given. Application of Artificial Neural Network in character recognition has made the system faster with optimum performance.

This paper presents the development of several multilayer feed-forward artificial neural network (ANN) models aimed at the simulation and parametrization of semiconductor optical amplifiers (SOAs). In the direct approach, it is assumed that the SOA physical parameters are known, and the ANNs can reproduce the results of detailed SOA models up to 20 times faster. In the inverse approach, it is assumed that the SOA is a grey-box with a known model but unknown parameters, and the ANNs infer the SOA model parameters from the input-output data. These results provide efficient surrogate models to represent SOAs in future digital twins for on-line optimization of ultra-wide band transmission systems that employ the SOA as the repeater element.

Gray layer technology represents a novel method for incorporating prior information about an uncertain nonlinear system into the structure of a multilayer feedforward artificial neural network (ANN). A prescriptive technique is developed that sets or constrains the weights of a particular layer, designated the gray layer, according to a known or partially known function that is related in some manner to the system of interest. An algorithm is derived for backpropagating the error through the gray layer and adjusting the parameters of the gray layer that is consistent with other ANN learning techniques. Gray layer technology is applied to the identification of two different partially known dynamic systems. Results demonstrate a significant improvement in the quality of the identification model and an increase in the rate of convergence.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Hypergame is a typical framework to deal with decision situations with multiple conflicting players who perceive the conflict differently and subjectively. In the face of opponent players' different outcome preferences information over the outcome space perceived by different decision makers using fuzzy linguistic values, a concept of fuzzy relational degree between different fuzzy numbers sequences is first presented in this paper, and based on fuzzy relational degree analysis, the weight of each decision maker can be evaluated and the integrated fuzzy outcome preferences can be obtained. Then, a multi-layer feedforward artificial neural network (ANN) is designed and trained to learn the criterions of comparing trapezoidal fuzzy numbers. Using the trained ANN, the integrated fuzzy outcome preferences can be ranked automatically. At last, a numerical example is provided to illustrate the proposed method.

Evaluation of neural network based simulation models for the prediction of site specific soil properties of the large agricultural areas will provide important technological benefits for the efficient use of land resources and Agricultural Decision Making System. In this study, an artificial intelligence model was investigated for simulation of site specific iron (Fe) and zinc (Zn) levels in the soils of apple orchard by using a feedforward multilayered Artificial Neural Network (ANN). The measurements of Fe and Zn were made for the topsoil (0-25 cm) and subsoil samples (25-50 cm) collected from forty five different coordinates within intervals of 20 x 10 m based on a grid sampling system in the east and north directions. The measured coordinate values were reflected to the input layer of the developed two input one output feedforward multilayer of the ANN model. Respectively, the measured Fe and Zn values were reflected to its output layer, and then the training stage was started. In the training, back propagation algorithms were used to get the most suitable NN structure for the prediction. As a result of varied trainings, the ANN having two hidden layers and ten neurons produced the best estimation values for site specific Fe and Zn estimations. The most suitable prediction values (R2 = 0.98 and R2 = 0.97, P<0.01) were obtained from the ANN with 10.10.1 structure for site specific Fe and Zn levels in the topsoil. When the simulated values obtained from the ANN model were compared with the measured values, it was observed that successful results were achieved based on varied training values. Hence, the ANN based model should be calibrated for varied training conditions depending on different soil conditions to get more reliable results. Key words: Articifial neural network, neural structures, iron, zinc, apple orchard.