abdessamad ELRHARRAS | Hassan II Casablanca (original) (raw)
Uploads
Papers by abdessamad ELRHARRAS
Spectrum sensing is the most important technique used to implement cognitive radio; this approach... more Spectrum sensing is the most important technique used to implement cognitive radio; this approach allows opportunistic and dynamic allocation of spectral bands. Among the methods used for detection, there are Artificial Neural Networks (ANN) and Energy Detection (ED); those exploit the signals coming from a Fast Fourier Transformed block (FFT). In this work, we focus on improving the performance of these three blocks by performing parallel computing, and considering the fusion of the two detectors ANN and ED. In this context, we implement three algorithms on GPU, which consist on exploiting the large number of cores to perform parallel calculation. The experimental results are compared with those obtained for CPU implementations. Our study presents how calculations distribution on GPU cores influences the global performance, and how to reduce execution time by optimizing data transfer. Furthermore, by exploiting the fine-grained parallel processing, and using a suitable choice of pa...
Spectrum sensing is the critical application in cognitive radio which has been proposed in order ... more Spectrum sensing is the critical application in cognitive radio which has been proposed in order to opportunistically benefit from the unused portions of the spectrum. It has shown that the detection of energy is the most convenient method, in the case where there is no a priori information about the primary user. In this work, the implementation of the energy detection technique has been done in MatLab for an AWGN channel, the simulation show that there are a lot of problems which decrease the performance of the energy sensor; it is susceptible to uncertainty in noise power and it cannot differentiate between primary user and the others cognitive users signal. In this respect, we propose in this paper, hybrid architecture which combines the simplicity of the energy detector, and the robustness of the artificial neural networks.
We present a Hardware/software (HW/SW) implementation of an artificial neural network aims at det... more We present a Hardware/software (HW/SW) implementation of an artificial neural network aims at detecting whether or not that the targeted X ray images of breast contain cancerous cell. This system has been trained in order to approximate functions or to achieve classification from a limited number of data. We implemented our application on FPGA by using the soft processor NIOS II.
In this paper, a method of classification of handwritten signature based on neural networks, and ... more In this paper, a method of classification of handwritten signature based on neural networks, and FPGA implementation is proposed. The designed architecture is described using Very High Speed Integrated Circuits Hardware Description Language (VHDL). The proposed application consists of features extraction from handwritten digit images, and classification based on Multi Layer Perceptron (MLP). The training part of the neural network has been done by using MATLAB program; the hardware implementations have been developed and tested on an Altera DE2-70 FPGA.
In this paper, a method of classification of handwritten signature based on neural networks, and ... more In this paper, a method of classification of handwritten signature based on neural networks, and FPGA implementation is proposed. The designed architecture is described using Very High Speed Integrated Circuits Hardware Description Language (VHDL). The proposed application consists of features extraction from handwritten digit images, and classification based on Multi Layer Perceptron (MLP). The training part of the neural network has been done by using MATLAB program; the hardware implementations have been developed and tested on an Altera DE2-70 FPGA.
Comput. Inf. Sci., 2015
Spectrum sensing is the most important technique used to implement cognitive radio; this approach... more Spectrum sensing is the most important technique used to implement cognitive radio; this approach allows opportunistic and dynamic allocation of spectral bands. Among the methods used for detection, there are Artificial Neural Networks (ANN) and Energy Detection (ED); those exploit the signals coming from a Fast Fourier Transformed block (FFT). In this work, we focus on improving the performance of these three blocks by performing parallel computing, and considering the fusion of the two detectors ANN and ED. In this context, we implement three algorithms on GPU, which consist on exploiting the large number of cores to perform parallel calculation. The experimental results are compared with those obtained for CPU implementations. Our study presents how calculations distribution on GPU cores influences the global performance, and how to reduce execution time by optimizing data transfer. Furthermore, by exploiting the fine-grained parallel processing, and using a suitable choice of pa...
The concern over Smart Farming is growing, where Internet of Things (IoT) technologies are highli... more The concern over Smart Farming is growing, where Internet of Things (IoT) technologies are highlighted in the farm management cycle. Also a large amount of data is generated via different channels such as sensors, Information Systems (IS), and human experiences. A timely right decision-making by monitoring, analyzing, and creating value from these Big Data is a key element to manage and operate the farms smartly, and is also bound to technical and socio-economic constraints. Given the fact, in this research, we work on the implication of Big Data technologies, IoT, and Data Analysis in agriculture. And we propose a Smart Farming Oriented Big Data Architecture (SFOBA).
Applied Mathematical Sciences, 2014
Computer and Information Science, Jan 28, 2015
Lecture Notes in Electrical Engineering, 2016
Cognitive radio has been proposed in order to benefit opportunistically from the unused portions ... more Cognitive radio has been proposed in order to benefit opportunistically from the unused portions of the spectrum, knowing that the first and the critical phase of this approach is the spectrum sensing, wherein the cognitive user must sense his external environment, to detect and profit dynamically from the free channels. One of the most used methods to detect the holes in the frequency spectrum is the energy detection; this technique does not need any prior knowledge about the primary signal. It is simpler and it requires less sensing time. However, there are a lot of problems that decrease the performance of the energy sensor; it is susceptible to the uncertainty in noise power. In this respect, we propose in this work, hybrid architecture which combines the simplicity of the energy detector, and the robustness of artificial neural networks ANN. The Principal Component Analysis is suggested as a pre-processing module in order to extract signal features.
Spectrum sensing is the most important technique used to implement cognitive radio; this approach... more Spectrum sensing is the most important technique used to implement cognitive radio; this approach allows opportunistic and dynamic allocation of spectral bands. Among the methods used for detection, there are Artificial Neural Networks (ANN) and Energy Detection (ED); those exploit the signals coming from a Fast Fourier Transformed block (FFT). In this work, we focus on improving the performance of these three blocks by performing parallel computing, and considering the fusion of the two detectors ANN and ED. In this context, we implement three algorithms on GPU, which consist on exploiting the large number of cores to perform parallel calculation. The experimental results are compared with those obtained for CPU implementations. Our study presents how calculations distribution on GPU cores influences the global performance, and how to reduce execution time by optimizing data transfer. Furthermore, by exploiting the fine-grained parallel processing, and using a suitable choice of pa...
Spectrum sensing is the critical application in cognitive radio which has been proposed in order ... more Spectrum sensing is the critical application in cognitive radio which has been proposed in order to opportunistically benefit from the unused portions of the spectrum. It has shown that the detection of energy is the most convenient method, in the case where there is no a priori information about the primary user. In this work, the implementation of the energy detection technique has been done in MatLab for an AWGN channel, the simulation show that there are a lot of problems which decrease the performance of the energy sensor; it is susceptible to uncertainty in noise power and it cannot differentiate between primary user and the others cognitive users signal. In this respect, we propose in this paper, hybrid architecture which combines the simplicity of the energy detector, and the robustness of the artificial neural networks.
We present a Hardware/software (HW/SW) implementation of an artificial neural network aims at det... more We present a Hardware/software (HW/SW) implementation of an artificial neural network aims at detecting whether or not that the targeted X ray images of breast contain cancerous cell. This system has been trained in order to approximate functions or to achieve classification from a limited number of data. We implemented our application on FPGA by using the soft processor NIOS II.
In this paper, a method of classification of handwritten signature based on neural networks, and ... more In this paper, a method of classification of handwritten signature based on neural networks, and FPGA implementation is proposed. The designed architecture is described using Very High Speed Integrated Circuits Hardware Description Language (VHDL). The proposed application consists of features extraction from handwritten digit images, and classification based on Multi Layer Perceptron (MLP). The training part of the neural network has been done by using MATLAB program; the hardware implementations have been developed and tested on an Altera DE2-70 FPGA.
In this paper, a method of classification of handwritten signature based on neural networks, and ... more In this paper, a method of classification of handwritten signature based on neural networks, and FPGA implementation is proposed. The designed architecture is described using Very High Speed Integrated Circuits Hardware Description Language (VHDL). The proposed application consists of features extraction from handwritten digit images, and classification based on Multi Layer Perceptron (MLP). The training part of the neural network has been done by using MATLAB program; the hardware implementations have been developed and tested on an Altera DE2-70 FPGA.
Comput. Inf. Sci., 2015
Spectrum sensing is the most important technique used to implement cognitive radio; this approach... more Spectrum sensing is the most important technique used to implement cognitive radio; this approach allows opportunistic and dynamic allocation of spectral bands. Among the methods used for detection, there are Artificial Neural Networks (ANN) and Energy Detection (ED); those exploit the signals coming from a Fast Fourier Transformed block (FFT). In this work, we focus on improving the performance of these three blocks by performing parallel computing, and considering the fusion of the two detectors ANN and ED. In this context, we implement three algorithms on GPU, which consist on exploiting the large number of cores to perform parallel calculation. The experimental results are compared with those obtained for CPU implementations. Our study presents how calculations distribution on GPU cores influences the global performance, and how to reduce execution time by optimizing data transfer. Furthermore, by exploiting the fine-grained parallel processing, and using a suitable choice of pa...
The concern over Smart Farming is growing, where Internet of Things (IoT) technologies are highli... more The concern over Smart Farming is growing, where Internet of Things (IoT) technologies are highlighted in the farm management cycle. Also a large amount of data is generated via different channels such as sensors, Information Systems (IS), and human experiences. A timely right decision-making by monitoring, analyzing, and creating value from these Big Data is a key element to manage and operate the farms smartly, and is also bound to technical and socio-economic constraints. Given the fact, in this research, we work on the implication of Big Data technologies, IoT, and Data Analysis in agriculture. And we propose a Smart Farming Oriented Big Data Architecture (SFOBA).
Applied Mathematical Sciences, 2014
Computer and Information Science, Jan 28, 2015
Lecture Notes in Electrical Engineering, 2016
Cognitive radio has been proposed in order to benefit opportunistically from the unused portions ... more Cognitive radio has been proposed in order to benefit opportunistically from the unused portions of the spectrum, knowing that the first and the critical phase of this approach is the spectrum sensing, wherein the cognitive user must sense his external environment, to detect and profit dynamically from the free channels. One of the most used methods to detect the holes in the frequency spectrum is the energy detection; this technique does not need any prior knowledge about the primary signal. It is simpler and it requires less sensing time. However, there are a lot of problems that decrease the performance of the energy sensor; it is susceptible to the uncertainty in noise power. In this respect, we propose in this work, hybrid architecture which combines the simplicity of the energy detector, and the robustness of artificial neural networks ANN. The Principal Component Analysis is suggested as a pre-processing module in order to extract signal features.