Kais Hassan

Automatic modulation recognition using wavelet transform and neural networks in wireless systems

Modulation type is one of the most important characteristics used in signal waveform identification. In this paper, an algorithm for automatic digital modulation recognition is proposed. The proposed algorithm is verified using higher-order statistical moments (HOM) of continuous wavelet transform (CWT) as a features set. A multilayer feed-forward neural network trained with resilient backpropagation learning algorithm is proposed as a classifier. The purpose is to discriminate among different M-ary shift keying modulation schemes and the modulation order without any priori signal information. Pre-processing and features subset selection using principal component analysis is used to reduce the network complexity and to improve the classifiers performance. The proposed algorithm is evaluated through confusion matrix and false recognition probability. The proposed classifier is shown to be capable of recognizing the modulation scheme with high accuracy over wide signal-to-noise ratio (SNR) range over both additive white Gaussian noise (AWGN) and different fading channels.

Blind Digital Modulation Identification for Spatially-Correlated MIMO Systems

Modulation type is one of the most important characteristics used in signal waveform identification and classification. Spatial correlation is a crucial factor for practical multiple-input multiple-output (MIMO) systems. This paper addresses the problem of blind digital modulation identification in spatially-correlated MIMO systems. The proposed algorithm is verified using higher order statistical moments and cumulants of the received signal. The purpose is to discriminate among different M-ary shift keying linear modulation schemes without any priori signal information. This study employs several MIMO techniques to identify the modulation with and without channel state information (CSI). The proposed classifier shows a high identification performance in acceptable signal-to-noise ratio (SNR) range.

Blind Modulation Identification for MIMO Systems

Modulation type is one of the most important characteristics used in signal waveform identification and classification. In this paper, an algorithm for blind digital modulation identification for multiple-input multiple-output (MIMO) systems is proposed. The suggested algorithm is verified using higher order statistical moments and cumulants of the received signal. A multi-layer neural network trained with resilient backpropagation learning algorithm is proposed as a classifier. The purpose is to discriminate among different M-ary shift keying linear modulation types and the modulation order without any priori signal information. This study covers different MIMO systems with and without channel state information (CSI). The proposed classifier is evaluated through the probability of identification where we show that our proposed algorithm is capable of identifying the modulation scheme with high accuracy in excellent signal-to-noise ratio (SNR) range.

Multiple-Antenna-Based Blind Spectrum Sensing in the Presence of Impulsive Noise

Cognitive radio (CR) was proposed as a solution to the spectrum scarcity problem. One of the basic functions of any CR is spectrum sensing. Most existing works on spectrum sensing consider the Gaussian noise assumption. In practice, this assumption is not always valid since several existing noise types exhibit non-Gaussian and impulsive behavior. Hence, it is very beneficial to study spectrum sensing in the presence of impulsive noise. In this paper, we propose two new multiple-antenna-based spectrum sensing methods, assuming that the underlying noise follows a symmetric α-stable distribution. This assumption is justified by a distribution fitting of some measurements of the noise acting on the GSM-R antennas onboard trains. The first proposed sensing method is based on the covariation properties of α-stable processes, whereas the second proposed method has the strategy of filtering the corrupted signals before applying a traditional spectrum sensing method. These two methods do not require a priori knowledge about the primary-user signal. Simulation results show that the proposed algorithms provide good spectrum sensing performance in the presence of α-stable distributed impulsive noise.

Automatic modulation recognition using wavelet transform and neural network

Modulation type is one of the most important characteristics used in signal waveform identification. An algorithm for automatic modulation recognition has been developed and presented in this study. The suggested algorithm is verified using higher order statistical moments of wavelet transform as a features set. A multi-layer neural network with resilient backpropagation learning algorithm is proposed as a classifier. The purpose is to discriminate different M-ary shift keying modulation types and modulation order without any priori signal information. Pre-processing and features subset selection using principal component analysis will reduce the network complexity and increase the recognizer performance.

Predicted Eigenvalue Threshold Based Spectrum Sensing with Correlated Multiple-Antennas

In this paper, we consider the problem of sensing a primary user in a cognitive radio network by employing multiple-antennas at the secondary user. Among the many spectrum-sensing methods, the predicted eigenvalue threshold (PET) based method is a promising non-parametric blind method that can reliably detect the primary users without any prior information. Also, a simplified PET sensing method, which needs to compare only one eigenvalue to its threshold, is introduced. A performance comparison between the proposed method and other existing methods is provided. Spatial antenna correlation at the secondary user is a crucial factor for practical systems. The effect of the spatial correlation presence on the different sensing methods is investigated.

Non-parametric multiple-antenna blind spectrum sensing by predicted eigenvalue threshold

In this paper, we consider the problem of sensing a primary user in a cognitive radio network by employing multiple antennas at the secondary user. Among the many spectrum-sensing methods, the predicted eigenvalue threshold (PET) based method is a promising non-parametric blind method that can reliably detect the primary users without any prior information. Then, a simplified PET sensing method, which needs to compare only one eigenvalue to its threshold, is introduced. Compared with the original PET sensing algorithm, the simplified algorithm significantly reduces the computational complexity without any loss in performance. A performance comparison between the proposed method and other existing methods is provided.

Blind detection of the number of transmitting antennas for spatially-correlated MIMO systems

The blind interception process of multiple-input multiple-output (MIMO) signals have recently gained more attention. The number of transmitting antennas is an important information for many blind MIMO algorithms. In this paper, several algorithms for blind detection of the number of transmitting antennas are presented. These different algorithms are evaluated and a performance comparison is presented. Spatial correlation is a crucial factor for practical MIMO systems. This paper addresses for the first time the problem of blind detection of the number of transmitting antennas in spatially correlated MIMO systems.

A cost-effective approach for spectrum sensing using beamforming

Spectrum sensing (SS) is one of the principal challenges on which the mobile communication is based on. Identifying the available frequency bands, also called white spaces, is the main issue. A novel blind approach for SS in the narrowband context is proposed in order to improve the signal detection. Considering a channel with its angle of arrival (AoA), we use beamforming technique to exploit the maximum and minimum angular energy. Both theoretical developments of the threshold and performance analysis are developed. To validate our contribution, the analytical results of the performance developed in this paper are compared with those from simulation. A comparison of state-of-the-art SS method using the eigenvalue decomposition is provided which brings an interesting trade-off between complexity and performance. Finally, simulation results considering the probability of misdetection under very low signal-to-noise ratio (SNR) are presented.

Blind Spectrum Sensing Using Extreme Eigenvalues for Cognitive Radio Networks

Here, a new spectrum sensing method, called mean-to-square extreme eigenvalue (MSEE), is proposed. Considering a multiple antenna communication system, the proposal is drawn from the arithmetic-to-geometric mean (AGM) algorithm using only the smallest and the largest eigenvalues of the covariance matrix of the received signal. The aim of MSEE is to avoid the heavy computational costs of AGM method. First, based on the random matrix theory, a theoretical development to set the threshold of the proposal is provided. Then, the validity of the expression is verified by simulations. Finally, simulation results show an interesting performance of MSEE compared with several spectrum sensing methods in the literature.