Ensemble convolutional neural networks for automatic fusion recognition of multi‐platform radar emitters

Abstract

Presently, the extraction of hand-crafted features is still the dominant method in radar emitter recognition. To solve the complicated problems of selection and updation of empirical features, we present a novel automatic feature extraction structure based on deep learning. In particular, a convolutional neural network (CNN) is adopted to extract high-level abstract representations from the time-frequency images of emitter signals. Thus, the redundant process of designing discriminative features can be avoided. Furthermore, to address the performance degradation of a single platform, we propose the construction of an ensemble learning-based architecture for multi-platform fusion recognition. Experimental results indicate that the proposed algorithms are feasible and effective, and they outperform other typical feature extraction and fusion recognition methods in terms of accuracy. Moreover, the proposed structure could be extended to other prevalent ensemble learning alternatives.

Automatic Classification of Basic Nursing Teaching Resources Based on the Fusion of Multiple Neural Networks

“…e mutual information method only considers the influence of positive correlation attributes on the importance of feature items. e square root fitting test is used to measure the lack of lexical and category independence [17,18]. e larger x 2 is, the smaller the independence is and the greater the correlation is, which is defined as follows:…”

Global ECG Classification by Self-Operational Neural Networks With Feature Injection

“…Time-frequency analysis is a crucial technique for extracting intra-pulse features of radar emitters, transforming signals into time-frequency images (TFIs). Notable methods in time-frequency analysis encompass the short-time Fourier transform (STFT) [3], Wigner-Ville distribution (WVD) [4], and Choi-Williams distribution (CWD) [5], among others. These methods have gained popularity for their ability to provide a detailed representation of signals across both time and frequency domains.…”

Ensemble the recent architectures of deep convolutional networks for skin diseases diagnosis

“…Handcrafted feature extraction in recognition algorithms running on radar systems is another complex application example that ensembling CNNs have overcome. In the 2017 study by Zhou et al the difficulty of extracting features from the radar signal and the observed changes performed in real time were automatically applied with multiple CNNs 6 . In their application, images generated by stacking multiple oscilloscope screens, similar to the channels of an RGB image, served as input for CNN models.…”

“…In the 2017 study by Zhou et al the difficulty of extracting features from the radar signal and the observed changes performed in real time were automatically applied with multiple CNNs. 6 In their application, images generated by stacking multiple oscilloscope screens, similar to the channels of an RGB image, served as input for CNN models. Multiple CNN models were trained using different platforms' time-frequency images of emitter signals.…”