نشریه صنایع الکترونیک
سال نهم شماره 3 (پیاپی 35، پاییز 1397)

Evaluation plays an important role in the interactive question answering (IQA) systems. In the context of evaluating IQA systems, there is partially no specific methodology for evaluating these systems in general. The main problem with designing an assessment method for IQA systems lies, in fact, that is rarely possible to predict interaction part. To this end, human needs to be involved in the evaluation process. In this paper, an appropriate model is presented by introducing a set of built in features for evaluating IQA systems. To conduct the evaluation process, four IQA systems were considered and based on the conversation was exchanged between users and systems, the number 540 samples were considered as suitable data to create a test and training set. After performing the preprocessing on the conversation, the statistical characteristics of the conversation extracted and base on that characteristics matrix was formed. Finally, using linear and nonlinear regression, the human thinking was predicted that the nonlinear power regression with 0.15 MSE was the best model.

Today's optimization is one of the most used areas in various sciences. Different algorithms are proposed to solve these problems. The SSPCO algorithm is one of the newest optimization algorithms. Simulation algorithms also have many challenges that are important in providing algorithms. Simulation is a tool to provide the real-world for applying the optimization algorithm. Meta-models are approaches to deal with simulation problems including heavy load of calculations and scalable runtime. Meta-model encounters some problems itself such as making the simulation model far away from its real-world. SSPCO Algorithm is one of the newest meta-heuristic optimization algorithms which model the behavior of partridge-bird chickens. In this paper, we tried to remove the meta-model as well as reducing the time to reach the optimal solution using SSPCO Algorithm. Simulation results which are presented in four ways of running the algorithm show that the simulation new methods could reduce the runtime by removing meta-model.

The HF surface wave radar, due to its transcendental horizon, makes it possible to detect the targets that the horizon prevents them from detecting in a vast area of the ocean and sea. However, in the event of high level of interference with colored clutter, the efficiency of detecting conventional processes (which is dependent on data) is reduced. Also, the Bragg component of the sea is another factor that prevents the detection of targets. In this paper, the general - purpose M-AR-GLR detector is used in a special case for HF surface wave radar and on the practical samples of sea clutter. It has been shown that with increasing the number of received pulses, the efficiency of detector becomes better. The efficiency of detector is improved by reducing the spatial dependency of the received samples. The detector maintains its performance with little secondary data. In the little signal to interference ratio, the proposed detector maintains its efficiency and is better than the GLR's Kelly. Hence, this detector is effective for sea clutter and HF surface wave radar.

Nowadays pre- and post-event disaster images are available in very high resolution by aerial imagery. Detecting buildings from aerial image in urban areas is a challenging task. Besides, finding equivalent buildings in pre- and post-disaster images is more challenging and hard to acquire. Accurate building detection in pre- and post-disaster images play significant role in damage assessment algorithms. This issue brought us to develop a common space for describing image pairs. To bate the problem, a method using PPI (Pixel Purity Index) and ISRA (Image Space Reconstruction Algorithm) is proposed. Some endmembers are extracted from pre-disaster image using PPI algorithm and both images are described according to these endmembers by ISRA. Our method could be a part of many conventional building detection algorithms and enable them being used in damage assessment algorithms. We embedded the proposed method into recently building detection algorithms to demonstrate the performance. It shows significant improvement in pre- and post-disaster images. It also shows that the purpose of finding equivalent buildings in pre- and post-disaster images is fully satisfied.

A method is proposed in this paper to enhance the security of communication systems. This method utilizes four cryptographic algorithms instead of one cryptographic algorithm. This will force the system to use one quadrature of each algorithm’s output stream. The Zynq system-on-chip from Xilinx Co. is used here and its fpga part is used for hardware implementation of these four cryptographic algorithms. Utilizing the four cryptographic algorithms leads to more hardware resource usage and is a challenging case. Partial reconfiguration is used here for solving this problem. Because of hardware implementation of these cryptographic algorithms, timing parameters of this design are all acceptable. Results show that not only the timing parameters are improved but also an enhanced security is achieved by using the four cryptographic algorithms instead of one cryptographic algorithm. A hardware resource utilization improvement of about 63% is achieved in comparison with traditional static implementation of the hardware.

In many applications such as IOT and different parts of the command and controll networks it is essential that information is transferred via network. In applications where the FPGAs are used as processors, due to cost and size limitiation, network layers need to be implemented on FPGA. Also with implementing all layers on FPGA it is possible to encrypt each layer individually. The purpose of this paper presents designing a node in the network based on IEEE802.3 standards so that it can be implemented on FPGA. In this design architecture combination of the data link, network and transport layer is considered. In order to increase speed and reducehe number of used LUTs, the design is based on pipleline. The chosen protocol of the design are IEEE802.3 for the data link layer, IPv4 for network layer and UDP for transport layer. The architecture mentioned above was implemented on xc6slx9-2 chip. Accourding to the results obtained in practice and software synthesise, about 25 percent of the FPGA's logic cells are consumped and 140 Mhz clock speed for the program is obtained.

This study has designed, analyzed a three-axis capacitor accelerator in which micro-electromechanical systems technology is employed. The accelerometers are generally divided into three categories of single axis, two axes, and three axes in terms of their ability to measure acceleration. In the suggested structure in this paper, acceleration measurements are carried out on all the three axes simultaneously with the use of a mass and spring system, which makes it possible to achieve a high sensitivity at a low occupancy level without losing other accelerator factors. Due to taking difference in this structure, the effect of any axis acceleration on the measured acceleration of the other two axes has been very low. If any external factor (e.g., electromagnetic waves) changes the value of a single capacitor, the original output of the capacitor measured for detecting acceleration does not change. That is, designing is done in such a way that the acceleration of each axis does not affect the other two axes and the external factors cannot also disrupt the system performance. Among the other important characteristics of the accelerometers are dynamic range, operating frequency, and sensitivity. As an advantage, this study covers a dynamic range up to 1000g and an operating frequency up to 20 kHz. The accelerometer sensitivity has been obtained to be 4fF/g in the z axis direction and 9fF/g in the x and y axes directions.

In this paper, the formalism of intrinsic losses, i.e. the below-band gap, thermalisation, Boltzmann, Carnot, and emission losses, is presented for the quantum wire intermediate band solar cells (QWr-IBSCs). For this purpose, a theoretical framework based on the principle of detailed balance approach with the help of physical origins of losses discussed in terms of photon absorption and emission in the presence of IB, is employed. Then, the intrinsic losses for a reported structure of QWr-IBSC where the intermediate band (IB) is introduced by embedding a stack of InxGa1-xAs QWr in the intrinsic layer of p-i-n GaAs solar cell, is calculated by presented formalism in this paper. Regarding that the calculation of position of IB, which is equivalent to the first eigen-energy of QWr, and the width of IB needed to obtain the intrinsic loss components, they are calculated by a finite element method in the context of Schrödinger equation and the tight binding method, respectively. Furthermore, the effect of indium molar fraction on the intrinsic loss components of GaAs/InxGa1-xAs QWr-IBSC is investigated.