Multimodal Learning for Classification of Solar Radio Spectrum

Abstract

Abstract-This paper proposes the first attempt to utilize multimodal learning method for the representation learning of the solar radio spectrums. The solar radio signals sensed from different frequency channels, which present different characteristics, are regarded as different modalities. We employ a multimodal neural network to learn the representations of the solar radio spectrum, which can distinguish the differences and learn the interactions between different modalities. The original solar radio spectrums are firstly pre-processed, including normalization, denoising, channel competition and etc., before being fed into the multimodal learning network. Experimental results have demonstrated that the proposed multimodal learning network can learn the representation of the solar radio spectrum more effectively, and improve the classification accuracy.

A Deep Learning Framework Using Passive WiFi Sensing for Respiration Monitoring

“…O'shea et.al proposed CNNs for signal modulation recongition [22]. Other researchers have targeted applications like solar radio burst classification [9], and CSIbased fingerprinting for indoor localization [11].…”

“…A deep learning network uses a cascade of multiple layers of non-linear processing units, where each successive layer takes the output from the previous layer as an input [9]. At each abstraction level, higher order features are constructed from the lower level features from the previous layers [10].…”

Auto Recognition of Solar Radio Bursts Using the C-DCGAN Method

“…For example, Gu et al [13] used a combination of principal component analysis (PCA) and support vector machine (SVM) for the mentioned purpose, yet the obtained accuracy of recognition needs to be improved. To do this, Chen et al [14] applied the multimodal network to auto-classify types of radio bursts, and later, they also tried the method of the deep belief network (DBN) [15] and the convolutional neural network (CNN) [16]. In addition, Yu et al [17] classified the solar radio data by using the long short-term memory network (LSTM) and obtained some improvement of the classification accuracy.…”

LSTM neural network for solar radio spectrum classification

“…It has been shown that the proposed model can learn better representation of a solar radio spectrum, and thus achieve higher accuracy of classification beyond the traditional support vector machine (SVM) (Suykens & Vandewalle 1999) coupled with principal component analysis (PCA) (Jolliffe 2011;Wold et al 1987). In Chen et al (2015); Ma et al (2017), an AE (Vincent et al 2008(Vincent et al , 2010 was explored for spectrum classification. In addition, the multiple modality concept (Guillaumin et al 2010;Ngiam et al 2011) was introduced to exploit the correlations between adjacent frequency channels, where each channel was regarded as one modality.…”

“…In this paper, all the models we have developed for spectrum classification in Chen et al (2016Chen et al ( , 2015; Ma et al (2017); Chen et al (2017b); Yu et al (2017) are summarized and compared. In particular, the LSTM model is extended and further enriched.…”