Context-aware cognitive radio using deep learning

Paisana, Francisco; Selim, Ahmed; Kist, Maicon; Álvarez, Pedro; Tallon, Justin; Bluemm, Christian; Puschmann, André; DaSilva, Luiz A.

doi:10.1109/dyspan.2017.7920784

Cited by 31 publications

(15 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training set is used to train the network shown in Fig.4 for thirty times, and it will consume about 600 seconds per epoch by using the platform described in section III.A. In the training, each fully connected layer is followed by dropout [30] at ratio of 0.5 to prevent the network from overfitting. The network is trained through a forward calculation and finetuning with a stochastic gradient decent algorithm to optimize the loss until the whole network converges.…”

Section: B Training and Testing Resultsmentioning

confidence: 99%

See 1 more Smart Citation

One-Dimensional Deep Attention Convolution Network (ODACN) for Signals Classification

et al. 2020

View full text Add to dashboard Cite

Handcraft features are commonly used for signal classification, which is a time-consuming feature engineering. In order to develop a general and robust feature learning method for radio signals, a novel One-dimensional Deep Attention Convolution Network (ODACN) is proposed to automatically extract discriminative features and classify various kinds of signals. First, one-dimensional (1-D) sparse filters are designed to learn hierarchical features of raw signals. Second, an attention layer is constructed to weight and assemble feature maps, to derive more context-relevant representation. By using simple 1-D filtering, ODACN is characteristic of less parameters and lower computation complexity than traditional Convolutional Neural Networks (CNNs). Moreover, feature attention can mimic a succession of partial glimpses of humans and focus on context parts of signals, thus helps in recognizing signals even at low Signal-to-Noise Ratio (SNR). Some experiments are taken to classify 31 kinds of signals with different modulation and channel coding types, and the results show that ODACN can achieve accurate classification of very similar signals, without any prior knowledge and manual operation.

show abstract

Section: B Training and Testing Resultsmentioning

confidence: 99%

“…The traditional CNNs are good at handling 2D data and have achieved tremendous success in many engineering fields [30], [31]. Thus signals are usually converted into 2D features for the further feature learning.…”

Section: One-dimensional Deep Attention Convolution Network (Odacmentioning

confidence: 99%

One-Dimensional Deep Attention Convolution Network (ODACN) for Signals Classification

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Researchers are focus on developing systems by using SDR based special hardware like USRP and image processing methods. Among lots of platforms, SDR based platforms famous for using capturing accurate WCSI data [16,17] [24][25][26][27], which motivated us to use. However, preparing an SDR platform is always challenging because hardware error and data error can show up at any moment.…”

Section: Motivationmentioning

confidence: 99%

“…Our main concern on image classification that why we focus on building a proper CNN model to achieving better accuracy results. In a few prior research works like [16,17] [ [24][25][26][27], SDR based hardware platform used to classify wireless signals using the CNN model. However, their designed CNN model is simple and not let them achieve high accuracy.…”

Section: Motivationmentioning

confidence: 99%

“…The convolution network consists of shared weights and to achieve high translation invariance, only weighty features of the data use [23]. Prior research works like [24][25][26][27] used simple CNN model instead of complex model to classify wireless data. In this paper, we are using AlexNet and ZFNet, two renowned architecture of CNN in a parallel manner to classify data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cognitive Intelligence for Monitoring Fractured Post-Surgery Ankle Activity Using Channel Information

et al. 2020

View full text Add to dashboard Cite

In the past decades, cognitive computing and communication densely used in lots of networking areas. Current improvement in deep learning (DL) and big data analysis create great potential to analyze cognitive intelligence (CI) for many applications such as human activity monitoring and recognition through wireless communication. Cognitive intelligence and wireless communication are using to establish smart healthcare systems. Healthcare monitoring systems turn into interesting research subjects where monitoring post-operative surgical patients are the current focal point to the researcher. In this paper, we argue that deep learning along with the wireless communication technique introduces cognitive intelligence for the healthcare monitoring system. We present a deep learning based convolutional neural network (CNN) model to classify image data and a convenient and multi-functional software-defined radio (SDR) platform to detect movement of the ankle of patients who underwent ankle fracture surgery. Capturing wireless channel state information (WCSI) in the presence of the human body and classifying using CNN to observe distinct movements is the key idea of this study. A universal software radio peripheral (USRP) platform used to capture WCSI data and used for classification. AlexNet and ZFNet both are the famous architecture of CNN and used in a parallel way to classify captured WCSI-based images that converted from numeric data. The classification established on the ankles movements after surgery and classification results show that CNN provides satisfying results where test accuracy is 98.98%.

show abstract

Predicting consecutive spectrum opportunities with hidden Markov models

Bezerra

Bordim

2022

Concurrency and Computation

View full text Add to dashboard Cite

Summary The increasing demands for wireless channels to accommodate the surge of internet of things devices and the associated services exacerbated the need for flexible channel allocation strategies. Opportunistic spectrum sharing is expected to provide a more reasonable use of the limited radio frequencies by allowing the coexistence of licensed users and unlicensed users in the same frequency. This arrangement is called opportunistic channel allocation, where unlicensed users explore the channel when the licensed user is not transmitting. The challenge in opportunistic spectrum allocation is to find transmission opportunities. Statistical and machine learning techniques have been used to forecast spectrum opportunities on time slotted channels based on acquired transmission data from licensed users. However, opportunity forecast is usually limited to one slot ahead of time. This work explores the use of Hidden Markov Model (HMM) training and predicting procedures to forecast a consecutive sequence of idle slots that can be explored for opportunistic transmission. To improve prediction accuracy and speedup the training process, the proposed training procedures reduce the classification alphabet, which is achieved by balancing the number of training sequences to limit the influence of outliers and provide opportunity forecast even when the training process is executed over a limited number of observed sequences. While an aggressive predictor may explore more opportunities as compared to a conservative predictor, the former may incur in higher collision rates while the latter may explore fewer opportunities. Hence, this work proposes a metric to assist in the process of selecting the best predictor's parameters to balance collision and opportunity rates. The proposed HMM predictors are evaluated on both synthetic and real traffic traces. The results on synthetic traffic shows that, for channel load from 30%$$ 30\% $$ up to 70%$$ 70\% $$, the proposed predictors identified over 80%$$ 80\% $$ of channel opportunities with the collision rates below 10%$$ 10\% $$. On real traffic traces, where the licensed users' transmissions have deterministic inter‐packet delays, the HMM predictors identified over 95%$$ 95\% $$ of the channel opportunities with collision rates below 1%$$ 1\% $$.

show abstract

Context-aware cognitive radio using deep learning

Cited by 31 publications

References 2 publications

One-Dimensional Deep Attention Convolution Network (ODACN) for Signals Classification

One-Dimensional Deep Attention Convolution Network (ODACN) for Signals Classification

Cognitive Intelligence for Monitoring Fractured Post-Surgery Ankle Activity Using Channel Information

Predicting consecutive spectrum opportunities with hidden Markov models

Contact Info

Product

Resources

About