The propagation analysis of an automotive radar using the finite-difference time-domain (FDTD) method uses tens of billions of analysis cells, and the simulation time is several days to several weeks. This study developed a method of accelerating the FDTD simulation for automotive radar analysis using a cluster-type supercomputer with multiple GPUs. An analysis region was divided into blocks, which were allocated to GPUs. The start time of the operation of each GPU was controlled according to radio wave propagation from a radiation antenna. The proposed method reduced the total simulation time by approximately 34% compared to a conventional method.

In this paper, we propose a deep-learning-based sequential phishing detection to improve the security and speed of the phishing detection. In our proposed method, phishing websites are detected in three phases: the URL, domain, and HTML analysis phases. In these phases, URLs, DNS records, and HTML contents are input to CNN-BiLSTMs (Convolutional Neural Network-Bidirectional Long Short Term Memory), respectively. Through experiments, we show that our proposed method is faster than the existing detection method, in which URLs and HTML contents are input to a CNN-BiLSTM simultaneously.

This study proposes a faster-than-Nyquist (FTN) single-carrier multiple-input multiple-output (MIMO) signaling scheme considering the effects of colored noise. In this approach, frequency-domain equalization (FDE) is considered as a practical receiver. Moreover, since FTN signaling generally induces colored noise, the proposed scheme considers its impact on FDE weight generation for improved detection accuracy of FTN-MIMO signaling. The effectiveness of the proposed scheme is demonstrated in terms of both bit error rate (BER) and throughput via the compression factor parameters and MIMO antenna configuration through computer simulations.