Malware detection in current times is increasingly important due to the presence of dangerous malicious software (malware) as well as ransomware in digital cyberspace. Conventional approaches such as using malware features (either static or dynamic or hybrid) have been applied for detection. Advances in Deep Learning (DL) has attracted a lot of interests in applications of malware detection. In particular, the file binaries are fed in to the DL neural network for training and testing. Despite the theoretical basis, we find that overfitting has occurred despite applying precaution, such as applying sufficient dropout layers. Limitations can also be attributed to the final classification layers (fully connected shallow network and softmax classification). In this paper, we apply transfer learning using ShuffleNet and DenseNet-201, two models trained on large dataset to recognize daily objects. This is done with the entire layers frozen to prevent overfitting and an optimal Error correction output code (ECOC) ensemble configuration of Support Vector Machines (SVM). Several ECOC coding matrices were applied, e.g., One vs. All (OVA), One vs. One (OVO), Dense Random (DR), and Sparse Random (SR). Each of these configurations represents varying complexity and ensemble size and, hence, a tradeoff between computation reduction and complex non-linear separation appears. Given that the continuous values of SVM parameters may take up high computation for acquiring the optimal parameter configuration, we apply discrete values combination using a grid search approach for parameter optimization. We test the proposed model on Malimg, MaleVis, virus-MNIST, and Dumpware10 datasets. The results show better/comparable accuracy compared with the existing work. The best/average accuracy values for each dataset over 10 trials are: Malimg (99.14%/98.87%), MaleVis (95.01%/93.91%), Virus-MNIST (86.36%/85.79%), Dumpware10 (96.62%/95.79%).

THz imaging, which is a THz wave spectroscopic diagnosis, was developed to solve the weakness of existing imaging techniques. One of the THz imaging techniques is THz near-field imaging. However, this technique has an issue that is only able to capture breast tissue image with the thickness of 20 μm. Optimum use of a proper antenna is expected to solve this problem. In this paper, we propose a rectangular patch microstrip array antenna design modified with inset line and rectangular slot to image thicker breast tissue. The antenna has a center frequency, bandwidth, gain, horizontal beamwidth and vertical beamwidth of 0.312 THz, 22.68 GHz, 5.6 dB, 86.5 degrees and 47.1 degrees, respectively. This antenna is expected to be able to image the breast tissue with the thickness of 2 mm and the distance between the transmitter and the receiver antenna of 4.5 mm.

The evolution of communication network comes from the need for users to stay connected anytime and anywhere. Modernization of the network continues to be done to increase bandwidth capacity in order to obtain an increase in multimedia services, especially in urban areas. Of various transmission media, optical fiber is the best choice to support reliable communication networks. One of the optical fiber communication technologies is Fiber to the Home (FTTH). This study proposes FTTH network design with Gigabit Capable Passive Optical Network (GPON) technology aimed at housing in urban areas. The design uses aerial and duct-aerial cabling systems. The design analysis is carried out based on the parameters of the link power budget, rise time, Bit Error Rate (BER), components and costs required. The value of the power budget link for the uplink and downlink are -22.792 dBm and -23.120 dBm. The rise time value for uplink and downlink are 0.256 ns and 0.258 ns, while the BER value is 14.628 x 10-12. The calculation results show that the design meets the required parameters, hence, the proposed network can be implemented. For the cost components required by the aerial system, it is 11.56% less than of the ductaerial sistem.

For application in a room-temperature microbolometer to detect terahertz waves, “narrow-width effects” on the temperature coefficient of resistance (TCR) and resistivity (ρ) for platinum (Pt) and titanium (Ti) nanowires are investigated. The TCR and ρ of a thermistor could be described by empirical formulas with average measured widths for higher precision. The effect of width on TCR in Pt is lower than that in Ti. The noise equivalent power of the microbolometer could be improved by a factor of 2.3 by replacing the Ti thermistor with a Pt thermistor of the same width (100 nm). The results provide a detailed understanding of the narrow-width effect on TCR for Pt and Ti, which, to our knowledge, is lacking at present.

The need for antennas with small dimensions with high performance has prompted various research to conduct experiments in terms of the structure and shape of the antenna and terms of material. This study proposes utilizing both artificially and engineered magneto-dielectric (MD) materials’ characteristics to miniaturize antenna dimension and improve antenna performance. The MD characteristics are realized through the SRR array structure as an antenna performance enhancer; and material synthesis as an antenna dimension reducer. MD material has been successfully synthesized, extracted, and characterized from a mixture of magnetite powder (Fe ₃ O ₄ ), PDMS elastic polymer material (polydimethylsiloxane), and BaFe ₁₂ O ₁₉ . The parametric study results show that the antenna designed on MD material Sample C (PDMS: Fe ₃ O ₄ =1: 0.6) with a 3 $\times $ 3 SRR configuration gives the most optimal response. Simulations, optimizations, and measurements have been carried out to obtain an MD antenna that works at a frequency of 3.5 GHz, with a fractional bandwidth of 10% (360 MHz), radiation efficiency of 58.54%, and a maximum gain of up to 4.33 dBi. This antenna measures 30 $\times30$ mm or 64% smaller than conventional antennas. The proposed stacked MD antenna is compatible with 5G technology in the mid-band frequency range.

Phased array antennas play a critical role in satellite communication systems, where minimizing peak sidelobe levels (PSLL) is essential to prevent interference and enhance signal quality. This paper presents a novel hybrid optimization approach, grey wolf optimizer with Nelder-Mead local search (GWO-NM), to suppress PSLL effectively in both linear antenna arrays (LAA) and planar antenna arrays (PAA). The proposed method combines the global search capability of greywolf optimizer (GWO) with the Nelder-Mead local refinement to address the PSLL suppression challenge. The optimization process was applied to both 32-element and 100-element LAA, achieving PSLL reductions to -52.63 dB and -56.39 dB, respectively, and these results are also extended to the corresponding PAA configuration. The results were further validated through full-wave electromagnetic simulations in CST Microwave Studio 2024, considering practical effects such as mutual coupling and material losses. Also, this suppression comes with trade-offs, including a widened first null beamwidth from 18.34° to 19.78° for the 32-element arrays and from 6.47° to 9.71° for the 100-element arrays in both LAA and their corresponding PAA compared the conventional algorithm, along with a marginal decrease of 0.5-1 dB in directivity observed in electromagnetic simulations. The GWO-NM algorithm outperformed other metaheuristic optimization techniques, including PSO, ACO, CSO, ABCO, and conventional GWO, demonstrating superior sidelobe suppression across different array configurations while maintaining an acceptable trade-off with other key radiation parameters.