Spatial-Temporal Hybrid Feature Extraction Network for Few-Shot Automatic Modulation Classification

Label noise is omnipresent in the annotations process and has an impact on supervised learning algorithms. This work focuses on the impact of label noise on the performance of learning models by examining the effect of random and class-dependent label noise on a binary classification task: quality assessment for photoplethysmography (PPG). PPG signal is used to detect physiological changes and its quality can have a significant impact on the subsequent tasks, which makes PPG quality assessment a particularly good target for examining the impact of label noise in the field of biomedicine. Random and class-dependent label noise was introduced separately into the training set to emulate the errors associated with fatigue and bias in labeling data samples. We also tested different representations of the PPG, including features defined by domain experts, 1D raw signal and 2D image. Three different classifiers are tested on the noisy training data, including support vector machine (SVM), XGBoost, 1D Resnet and 2D Resnet, which handle three representations, respectively. The results showed that the two deep learning models were more robust than the two traditional machine learning models for both the random and class-dependent label noise. From the representation perspective, the 2D image shows better robustness compared to the 1D raw signal. The logits from three classifiers are also analyzed, the predicted probabilities intend to be more dispersed when more label noise is introduced. From this work, we investigated various factors related to label noise, including representations, label noise type, and data imbalance, which can be a good guidebook for designing more robust methods for label noise in future work.

Section: Introductionmentioning

confidence: 99%

Impact of Label Noise on the Learning Based Models for a Binary Classification of Physiological Signal

Ding

Pereira

Xiao

et al. 2022

“…In the cognitive radio (CR) communication system, the signal blind detection technology can achieve the accurate discovery, information restoration, and user identity adjudication of illegal authorized users, and further conduct location monitoring, spectrum suppression, and investigation and forensics to protect the safe use of radio spectrum resources by legitimate users, which is of great practical importance. In this paper, the main goal of blind detection of communication signals is the detection and preliminary automatic modulation identification (AMC) of communication signals under blind information conditions (i.e., reception of non-cooperating party signals) [ 3 , 4 , 5 , 6 ].…”

Section: Introductionmentioning

confidence: 99%

“…Some existing studies have obtained good results on AMC tasks by transforming signals into time-frequency images and then inputting them into classifiers built by neural networks [ 12 , 13 ]. Che et al proposed a spatial-temporal hybrid feature extraction network for AMC, which maps wireless communication signals into spatial feature space and temporal feature space, respectively, to improve the effectiveness in the few-sample AMC task [ 5 ]. After the development in recent years, AMC techniques based on deep learning have become mature [ 12 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Blind Detection of Broadband Signal Based on Weighted Bi-Directional Feature Pyramid Network

Guo

Yao

et al. 2023

With the development of wireless technology, signals propagating in space are easy to mix, so blind detection of communication signals has become a very practical and challenging problem. In this paper, we propose a blind detection method for broadband signals based on a weighted bi-directional feature pyramid network (BiFPN). The method can quickly perform detection and automatic modulation identification (AMC) on time-domain aliased signals in broadband data. Firstly, the method performs a time-frequency analysis on the received signals and extracts the normalized time-frequency images and the corresponding labels by short-time Fourier transform (STFT). Secondly, we build a target detection model based on YOLOv5 for time-domain mixed signals in broadband data and learn the features of the time-frequency distribution image dataset of broadband signals, which achieves the purpose of training the model. The main improvements of the algorithm are as follows: (1) a weighted bi-directional feature pyramid network is used to achieve a simple and fast multi-scale feature fusion approach to improve the detection probability; (2) the Efficient-Intersection over Union (EIOU) loss function is introduced to achieve high accuracy signal detection in a low Signal-Noise Ratio (SNR) environment. Finally, the time-frequency images are detected by an improved deep network model to complete the blind detection of time-domain mixed signals. The simulation results show that the method can effectively detect the continuous and burst signals in the broadband communication signal data and identify their modulation types.

“…To work well under the conditions of limited training samples, the idea of task-oriented training in few shot learning (FSL) [ 4 , 5 , 6 , 7 , 8 , 9 , 10 ] can be combined. However, not all FSL methods are applicable.…”

Section: Introductionmentioning

confidence: 99%

SAR Target Recognition with Limited Training Samples in Open Set Conditions

Zhou

Zhang

Liu

et al. 2023

It is difficult to collect training samples for all types of synthetic aperture radar (SAR) targets. A realistic problem comes when unseen categories exist that are not included in training and benchmark data at the time of recognition, which is defined as open set recognition (OSR). Without the aid of side-information, generalized OSR methods used on ordinary optical images are usually not suitable for SAR images. In addition, OSR methods that require a large number of samples to participate in training are also not suitable for SAR images with the realistic situation of collection difficulty. In this regard, a task-oriented OSR method for SAR is proposed by distribution construction and relation measures to recognize targets of seen and unseen categories with limited training samples, and without any other simulation information. The method can judge category similarity to explain the unseen category. Distribution construction is realized by the graph convolutional network. The experimental results on the MSTAR dataset show that this method has a good recognition effect for the targets of both seen and unseen categories and excellent interpretation ability for unseen targets. Specifically, while recognition accuracy for seen targets remains above 95%, the recognition accuracy for unseen targets reaches 67% for the three-type classification problem, and 53% for the five-type classification problem.