Nowcasting plays a crucial role in responding to disasters such as flash floods by predicting rainfall in real time. However, existing nowcasting models struggle to accurately predict extreme rainfall events, which, although rare, can have devastating impacts. This challenge primarily arises because typical loss functions focus on minimizing average prediction errors rather than emphasizing the importance of extreme events, leading to their underestimation. To address this issue, this study introduces various weighted loss functions that impose greater penalties on prediction errors as rainfall intensity increases. These weighted loss functions were applied to a convolutional long short-term memory (ConvLSTM)-based nowcasting model to assess their impact on model performance. Recognizing that weighted loss functions may influence the learning of spatial patterns, and this study categorized extreme rainfall events based on their spatial characteristics and conducted a detailed performance evaluation for each type. The results demonstrated that models using weighted loss functions significantly improved the accuracy of extreme rainfall predictions compared to unweighted (UW) models. Notably, depending on the applied weighted loss functions, each model clearly exhibited its strengths and weaknesses across various extreme rainfall types. This finding suggests that selecting the best-performing weighted model based on prediction goals can lead to optimal results. Furthermore, this study revealed that the effectiveness of prediction methods varies significantly depending on the type of extreme rainfall event, indicating the need for dynamic selection of prediction methods tailored to specific condition. This article provides valuable insights into improving extreme rainfall nowcasting and is expected to contribute to enhancing disaster response systems in the future.

In this paper, we propose a simple construction for binary (n,k) linear codes using s-mark Golomb rulers. We prove that these codes are sequential-recovery locally repairable codes (LRCs) with availability 2, which can sequentially recover 5 erased symbols. We prove the necessary and sufficient condition for the proposed codes to be rate-optimal. We also prove the necessary and sufficient condition for the proposed codes to be dimension-optimal. Finally, we propose some variations of this constructions to obtain some 5-sequential recovery LRCs with availability 3. The proposed codes have higher rates and have more flexible choice for the lengths than other previously reported constructions.

In this paper, we define (m,n,r) zero-correlation-zone (ZCZ) sonar sequences and present some of their properties. We prove an upper bound on r for (m,n,r) ZCZ sonar sequences and propose a new and simple construction for (m,n,r) ZCZ sonar sequences with m = r2− 1 and any positive integer n. We also propose two constructions for (m,n,2) ZCZ-DD sonar sequences for some m and n which are some variations of well-known sonar sequence constructions. We report a lot of exhaustive search results and some interesting open problems.

In this paper, we derive a lower bound on the alphabet size of the optimal uncorrelated polyphase ZCZ sequence families, assuming that Mow’s Conjecture is true. We also propose some new optimal uncorrelated polyphase ZCZ sequence families for all the optimal ZCZ parameters. All our proposed families achieve our proposed alphabet size bound so that this bound is the minimum alphabet size (assuming Mow’s conjecture is true). We also derive some interesting facts regardless of the truth of Mow’s Conjecture: an optimal ZCZ family of normalized sequence (not necessarily uncorrelated) is always periodic complementary; therefore, a sequence generated by interleaving all the sequences in an optimal uncorrelated ZCZ family of normalized sequences is always a perfect sequence.

This paper briefly introduces the transmitter of the GNSS signal to which CSK modulation is applied, and summarizes the CSK hard/soft decision demodulation method in the corresponding receiver. Based on the summarized method, we show the hard/soft decision demodulation performance and various analyzes in an environment similar to the existing GNSS signal.

In this paper, we analyzes performance of girth-8 regular QC-LDPC codes constructed using Golomb ruler. We conducted simulations to measure FER performance of QC-LDPC codes constructed by changing the last mark of some optimal Golomb ruler and found the value of the mark that shows the best performance.