نشریه مهندسی مخابرات جنوب
پیاپی 43 (بهار 1401)

Wideband linear frequency modulation (LFM) signals are widely used in systems such as radar, sonar and mobile. DOA estimation algorithms LFM signals rely on a large number of time frames (Snapshot). Therefore, they are not suitable for low power and real-time applications. In this paper, we present a low computation based estimation method based on the SPICE algorithm for linear arrays using the Fourier transform FFT. The method is to first develop the dechirping algorithm for LFM signals using the FFT converter, then modify the SPICE algorithm for uniform linear arrays (ULA), then obtain the DOA calculations for a few number of snapshots with low computational volumes. The simulation results of the proposed MSPICE algorithm (i.e. modified SPICE algorithm) show that the proposed algorithm is better than other algorithms such as MUSIC and SPICE. We have shown that the proposed method has acceptable accuracy in both low SNRs and low error in high SNRs. In addition, in terms of computational volume, it is much less than the MUSIC and SPICE methods and will have a good robustness compared to other methods.

In this paper, a four-band microstrip antenna using metamaterial structure is presented. The antenna proposed in this article, has an asymmetric structure and it is made up of two identical radiation elements and ground plane at the back. In this design, each radiation element consists of a reciprocating capacitor, which is attached to a slotted triangular acetate at the end. The proposed antenna has a miniature dimension of 10×15 mm2. Also the proposed antenna has four resonances in frequencies of 3.19 GHz (1.25%), 6.22 GHz (10.74%), 7.18 GHz (5.19%) and 9.12 GHz (4.52%). Pattern and gain of the antenna are acceptable at resonant frequencies and the good consistency between the test result and the simulation confirm the accuracy of the antenna design. The proposed antenna is made and tested in the antenna laboratory. The results of the measurements are presented in the paper. Good consistency between test and simulation results confirms the accuracy and efficiency of the proposed multirole microstrip antenna design.

Classification of hyperspectral images is one of the most important processes on these images. Hyperspectral images are high dimensional, so classification of these images is difficult. Therefore, methods that extract low-dimensional subspace structures from the hyperspectral image are considered. The low-rank representation method can extract the low-dimensional subspace structure in the data. This method considers the global structure of the data. In this paper, to preserve the global and local structure in the data, spares and low-rank representation feature extraction method based on spectral and spatial information is presented. The data structure is better revealed using this model, and the discrimination of the features is increased. In this model, each pixel is expressed by a linear combination of dictionary atoms. In addition, to solve the optimization problem, the alternating direction method of multipliers has been used. The simulation results show that the proposed model has better results than other methods.

Social networks have found many applications in people's daily lives today so that identifying the behavior of members of these types of networks and associations within them is of particular importance. Due to the structure and the way of communication between the members of social networks, some members within this type of network have more important roles than other members. In this study, a method for identifying more important communities was discussed. For this purpose, new features were introduced using network centralization features and then the importance of this type of feature was investigated by Rough set theory. The experimental results showed that by increasing the number of popular nodes among a community where introduced in this study and at the same time decreasing the value of density, betweenness, and closeness centrality features of that community, it causes the effect of the number of popular nodes on the community will remain more evident.

In this paper, a novel plasmonic optical switch based on graphene is designed and simulated with better features and four channels. Each of these channels has an equal width of 20nm which constructed from graphene ribbons spaced at a distance of 30nm from each other. The operation of device is based on the modulation of graphene chemical potential in each channel by using applied voltage. This means that by application of a proper voltage the transmission of input in every channel can be controlled and hence the switching can be occurred. Performance characteristics of this optical switch achieves a relatively high transmission with very low loss at the frequency of 30THz. Also in this structure the amount of crosstalk is remarkably low. This structure has been investigated in different chemical potentials in which one of it’s best status occurs in chemical potential of 0.4eV. All analyses and simulations have been performed in Lumerical and MATLAB.

Photon crystals are on of the most common structures for designing and implementing electronic optical circuits. The photon gap band is of great importance in these structures. The most common mathematical methods for analyzing and calculating photon band diagrams and extracting the forbidden photon band of photon crystals is the flat wave expansion method. This method has many complexities that require special commercial software to perform its calculations. In two-dimensional photonic crystals, two parameters in the photon band are affected, which are the refractive index of the dielectric material and the ratio of the radius of the holes to the constant of the crystal lattice (r / a). In this paper, using mathematics and linear regression, simpler relations for extracting the photon band range of photon crystals are presented. Comparison of the results obtained from the relations with the results obtained with BandSOLVE software for photonic crystals shows that the maximum difference between the two calculation methods for fl and fu is 0.007 and 0.028, respectively, and the average difference is , respectively 0.004 and is 0.01.