TY - JOUR
T1 - Deep Neural Networks for Spectrum Sensing
T2 - A Review
AU - Syed, Sadaf Nazneen
AU - Lazaridis, Pavlos I.
AU - Khan, Faheem A.
AU - Ahmed, Qasim Zeeshan
AU - Hafeez, Maryam
AU - Ivanov, Antoni
AU - Poulkov, Vladimir
AU - Zaharis, Zaharias D.
N1 - Publisher Copyright:
© 2013 IEEE.
PY - 2023/8/25
Y1 - 2023/8/25
N2 - As we advance towards 6G communication systems, the number of network devices continues to increase resulting in spectrum scarcity. With the help of Spectrum Sensing (SS), Cognitive Radio (CR) exploits the frequency spectrum dynamically by detecting and transmitting in underutilized bands. The performance of 6G networks can be enhanced by utilizing Deep Neural Networks (DNNs) to perform SS. This paper provides a detailed survey of several Deep Learning (DL) algorithms used for SS by classifying them as Multilayer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, combined CNN-LSTM architectures and Autoencoders (AEs). The works are discussed in terms of the input provided to the DL algorithm, data acquisition technique used, data pre-processing technique used, architecture of each algorithm, evaluation metrics used, results obtained, and comparison with standard SS detectors. This survey further provides an overview of traditional Machine Learning (ML) algorithms and simple Artificial Neural Networks (ANNs) while highlighting the drawbacks of conventional SS approaches for completeness. A description of some publicly available Radio Frequency (RF) datasets is included and the need for comprehensive RF datasets and Transfer Learning (TL) is discussed. Furthermore, the research challenges related to the use of DL for SS are highlighted along with potential solutions.
AB - As we advance towards 6G communication systems, the number of network devices continues to increase resulting in spectrum scarcity. With the help of Spectrum Sensing (SS), Cognitive Radio (CR) exploits the frequency spectrum dynamically by detecting and transmitting in underutilized bands. The performance of 6G networks can be enhanced by utilizing Deep Neural Networks (DNNs) to perform SS. This paper provides a detailed survey of several Deep Learning (DL) algorithms used for SS by classifying them as Multilayer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, combined CNN-LSTM architectures and Autoencoders (AEs). The works are discussed in terms of the input provided to the DL algorithm, data acquisition technique used, data pre-processing technique used, architecture of each algorithm, evaluation metrics used, results obtained, and comparison with standard SS detectors. This survey further provides an overview of traditional Machine Learning (ML) algorithms and simple Artificial Neural Networks (ANNs) while highlighting the drawbacks of conventional SS approaches for completeness. A description of some publicly available Radio Frequency (RF) datasets is included and the need for comprehensive RF datasets and Transfer Learning (TL) is discussed. Furthermore, the research challenges related to the use of DL for SS are highlighted along with potential solutions.
KW - 6G
KW - Cognitive radio
KW - Autoencoders
KW - Convolutional Neural Networks (CNNs)
KW - Deep Learning (DL)
KW - Deep Neural Networks (DNNs)
KW - Machine Learning (ML)
KW - Multilayer Perceptrons (MLPs)
KW - Long Short-Term Memory (LSTM) Networks
KW - Spectrum Sensing
KW - machine learning (ML)
KW - multilayer perceptrons (MLPs)
KW - deep learning (DL)
KW - deep neural networks (DNNs)
KW - convolutional neural networks (CNNs)
KW - cognitive radio
KW - long short-term memory (LSTM) networks
KW - spectrum sensing
UR - http://www.scopus.com/inward/record.url?scp=85168289988&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2023.3305388
DO - 10.1109/ACCESS.2023.3305388
M3 - Review article
AN - SCOPUS:85168289988
VL - 11
SP - 89591
EP - 89615
JO - IEEE Access
JF - IEEE Access
SN - 2169-3536
M1 - 10217791
ER -