Browsing by Subject "Convolutional neural network (CNN)"

Now showing 1 - 2 of 2

Open Access
Deep learning in electronic warfare systems: automatic pulse detection and intra-pulse modulation recognition
(2020-12) Akyon, Fatih Cagatay
Detection and classification of radar systems based on modulation analysis on pulses they transmit is an important application in electronic warfare systems. Many of the present works focus on classifying modulations assuming signal detection is done beforehand without providing any detection method. In this work, we propose two novel deep-learning based techniques for automatic pulse detection and intra-pulse modulation recognition of radar signals. As the first nechnique, an LSTM based multi-task learning model is proposed for end-to-end pulse detection and modulation classification. As the second technique, re-assigned spectrogram of measured radar signal and detected outliers of its instantaneous phases filtered by a special function are used for training multiple convolutional neural networks. Automatically extracted features from the networks are fused to distinguish frequency and phase modulated signals. Another major issue on this area is the training and evaluation of supervised neural network based models. To overcome this issue we have developed an Intentional Modulation on Pulse (IMOP) measurement simulator which can generate over 15 main phase and frequency modulations with realistic pulses and noises. Simulation results show that the proposed FFCNN and MODNET techniques outperform the current stateof-the-art alternatives and is easily scalable among broad range of modulation types.
Open Access
Numerical weather forecasting using convolutional LSTM with attention and context matcher mechanisms
(IEEE, 2024) Tekin, Selim Furkan; Fazla, Arda; Kozat, Süleyman Serdar
Numerical weather forecasting with high-resolution physical models requires extensive computational resources on supercomputers, often making it impractical for real-life applications. Alternatively, deep learning methods can provide results within minutes of receiving data. Although baseline deep learning models can make accurate short-term predictions, their performance deteriorates rapidly as the output sequence length increases. However, many real-life scenarios require long-term prediction of certain weather features to mitigate and take advantage of the effects of high-impact weather events. In response, we introduce the Weather Model, which provides rapid and accurate long-term spatial predictions for high-resolution spatio-temporal weather data. We integrate a stacked convolutional long-short term memory (ConvLSTM) network as our building block, given its accuracy in capturing spatial data patterns through convolution operations. Furthermore, we additionally incorporate attention and context matcher mechanisms. The attention mechanism allows the effective usage of the side-information vector by selectively focusing on different parts of the input sequence at each time step. Concurrently, the context matcher mechanism enhances the network’s ability to preserve long-term dependences. Our Weather Model achieves significant performance improvements compared to baseline deep learning models, including ConvLSTM, TrajGRU, and U-Net. Our experimental evaluation involves high-scale, real-world benchmark numerical weather datasets, namely the ERA5 hourly dataset on pressure levels and WeatherBench. Our results demonstrate substantial improvements in identifying spatial and temporal correlations, with attention matrices focusing on distinct parts of the input series to model atmospheric circulations. We also compare our model with high-resolution physical models using benchmark metrics to confirm our Weather Model’s accuracy and interpretability.