Primary User Detection Research Articles

A hybrid deep learning based approach for spectrum sensing in cognitive radio

The primary user (PU) transmission is sporadic in nature, which explains why the PU is inactive during some time slots, geographic directions or frequency bands. The frequency bands where the PU is not active are called "spectrum holes". Secondary users (SUs) periodically perform sensing to detect the spectrum holes and monitor primary spectrum. For the best possible spectrum utilization, PU signal detection is very crucial. For measuring the spectrum sensing performance, two main metrics are applied, like, probability of false alarm (PFA) and probability of detection (PD). Due to PFA and PD, the conventional sensing techniques have to face issues. These two constraints used to hinder spectrum utilization. Traditional sensing strategies are mostly based on feature extraction of received signal. Advancement of artificial intelligence (AI) has reduced the inaccuracy in detection of spectrum hole. Deep learning (DL) based approaches have shown a remarkable improvement in this aspect. Hence, the present research work was undertaken to address the problem of spectrum sensing in low SNR and improves accuracy. This research penetrates into the use of deep neural network (DNN) for sensing the vacant spectrum accurately. In this article, RadioML2016.10b dataset was used for the experiments. The results are also studied. The proposed approach shows betterment in sensing than other existing spectrum detection models. DeepSenseNet model was validated through simulation results and showed that it has achieved 98.84% prediction accuracy (Pa) with 97.53% precision and 97.62% recall.

Enhancing spectrum utilization and primary user detection in cognitive radio networks through greedy cooperative spectrum sensing

The utilization of the spectrum is optimized through which primary users of modern wireless communication technologies might obtain a higher chance of detection. The research aims to study how the NI-USRP hardware platform can be used to set up greedy cooperative spectrum sensing for cognitive radio networks. Research primarily deals with energy detection and eigenvalue-based detection approaches, both of which are highly recognized for their capacity to sense the spectrum without having prior knowledge of the primary user signals. In the hardware arrangement, there is one transmitter and two cognitive radio receivers. LABVIEW makes it simple to deploy and maximizes the detection probability across a large sample. Here, it was demonstrated that cooperative spectrum sensing is superior to non-cooperative spectrum sensing, which results in a reduction in the risk of errors occurring during detection. The research discovered that the OR combination rule has a higher detection probability than the AND rule at the same time. The research emphasizes the significance of expanding cooperative spectrum sensing to improve overall detection capabilities. SNRs that are more than 10 dB allow the energy detector to operate, and the eigenvalue detector continues to work when the SNR drops to –9 dB.

A Spectrum Sensing and Allocation Model for Primary User Detection and Interference Mitigation in Television Whitespaces

A Spectrum Sensing and Allocation Model for Primary User Detection and Interference Mitigation in Television Whitespaces

One-bit spectrum sensing using Gustafson–Kessel fuzzy clustering for cognitive radio network

In this article, we address the problem of one-bit spectrum sensing using a multi-antenna array for cognitive radio networks. Spectrum sensing with one-bit analog-to-digital converter (ADC) is a far-reaching technology as it requires only a simple comparator for conversion and has the advantage of low-cost and high-speed. However, spectrum sensing with one-bit quantized data is challenging because of severe loss of information due to nonlinearities involved. Existing one-bit spectrum sensing techniques utilize arcsine law to approximate original signal covariance matrix from their quantized version and then analytically model the signal features for deriving the spectrum classification rule. However, these techniques are vulnerable to the inaccuracies of the model and it may not be at all possible to derive a classification rule for analytically intractable noise scenarios. In this article, we follow a data-driven approach for one-bit spectrum sensing and propose Gustafson–Kessel fuzzy c-means (GKFCM) clustering based algorithms for detection of primary user signals in white and correlated noise. We identify the features which are conserved in the process of quantization and form decision vectors using them. Decision vectors obtained from the received one-bit quantized data are then clustered using GKFCM to obtain the model parameters and classification is performed to identify the occupancy of the channel. Simulation results reveal that proposed algorithms provide significant improvement in detection performance compared to existing one-bit spectrum sensing techniques in literature.

Modified Threshold-based Spectrum Sensing Approach for VANETs

The Primary User (PU) signal detection in Cognitive Radio (CR) is crucial and is achieved through spectrum sensing techniques. The Energy Detection method is a commonly used technique, and selecting a proper threshold is essential to enhance the efficiency of the CR system. This research paper demonstrates the maximum achievable throughput and validates a Modified Threshold (MT) approach. The authors consider a scenario with multiple antennas at the receiver, where these antennas are correlated and subjected to mobility effects, and they employ the Energy Detection (ED) for spectrum sensing. The study analyzes the system's performance over a Nakagami-m fading channel, considering available correlations among the antenna elements. To compute important statistical values, the Moment Generating Function (MGF) method is employed. The research employs specialized mathematical functions, such as the Lauricella and confluent hypergeometric functions, to derive closed-form expressions for the Probability of Detection when employing the diversity technique. The results indicate a significant enhancement in the performance of the proposed algorithm when utilizing the modified threshold parameter across a wide range of Signal to Noise Ratio (SNR) values. Additionally, increasing the number of branches in the antenna system further improves detection performance. Interestingly, under high fading parameter conditions (m=4), the detection probability is found to be superior with exponential correlation among the L antenna elements compared to other available correlated branches.

A Segmentation based Classification Model for Primary User Detection Using Deep Learning Techniques

A Segmentation based Classification Model for Primary User Detection Using Deep Learning Techniques

SDR-Implementation of a Support Vector Machine-Assisted Covariance-Based Spectrum Sensing Algorithm in the Presence of Correlated Noise

Cognitive radio (CR) is an emergent technology that is considered for more efficient spectrum utilization. Spectrum Sensing (SS), i.e. detection of primary user (PU) in the frequency band, is one of the key tasks of cognitive radio systems. Recently, various works on the support vector machine (SVM)-assisted covariance-based spectrum sensing (CSS) in multi-receiver systems were presented and evaluated with computer simulations. However, the real environment is different, especially since the noise is often assumed to be spatially uncorrelated, while it is correlated in reality. Therefore, we propose a four-receivers software-defined radio (SDR)-based test platform for experimental evaluation of SVM-assisted CSS techniques. Furthermore, we implement an SVM-assisted CSS technique in the proposed platform to assess its performance when two and four receivers are activated. The measurements showed that increasing the number of receiving channels does indeed enhances the detection probability. Moreover, we evaluated the algorithm in both white and correlated noise scenarios. The evaluation demonstrates how the presence of correlated noise deteriorated the performance of the SVM-assisted CSS compared with the case where only white noise is present. More specifically, the simulations showed that to reach 99% detection probability when noise correlation is present, the implemented algorithm requires an SNR value that is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{10}~dB$</tex-math></inline-formula> greater than the one required when white noise only is present.

Collaborative and Incremental Learning for Modulation Classification With Heterogeneous Local Dataset in Cognitive IoT

Among 6G high-density connection applications, the cognitive Internet of Things (IoT) is considered to be an effective means to cope with the problem of spectrum resource shortage and reduce energy consumption. The modulation classification with the functions of primary user detection and signal recognition has become the core technology in the cognitive IoT. However, the remarkable surge in the generated data makes it intractable to store and analyze such a large amount of data in a centralized manner. Meanwhile, due to the heterogeneity of IoT devices, the collected data in distributed devices may be non-independent-identical distribution, resulting in heterogeneous local datasets problem. Therefore, we introduce the federated learning (FL) and propose federated incremental learning (FIL) method to address the problem of heterogeneous local datasets in modulation classification. Specifically, we regard the private classes of a particular local device as incremental classes and incremental learning is adopted to learn the classification knowledge of private classes. In order to avoid excessive dispersion between the global model parameters and local model parameters, we utilize knowledge distillation in the local model training stage for imposing similar constrain of parameters on different local models. Simulation results have proven the effectiveness of our proposed method.

Covariance absolute values spectrum sensing method based on two adaptive thresholds

Cognitive radio is a modern wireless communication methodology that deals with the issue of spectrum untapped. Cognitive users can continually perceive the licensed spectrum to hunt for unoccupied spots. The essential technology in cognitive radio (CR) for primary user detection is spectrum sensing. Due to signal to noise ratio (SNR), noise uncertainty in spectrum sensing would make the detection unreliable. In this paper, the two adaptive thresholds based on_covariance absolute values (TATCAV) are proposed to increase detection performance in the presence of noise uncertainty. According to the computer simulations using MATLAB 2021b, the value of the probability of detection is Pd=98.1% Compared with the results of two thresholds based on_covariance absolute values (TTCAV) Pd= 95.3% at SNR=-18, noise uncertainty Nu=1.761 dB, and using quadrature amplitude modulation (QAM). And the error rate for the proposed approach is Pe=12.1% under the same circumstances. The proposed approach results, according to the simulations, are considerably better than the results of the fixed two-threshold approach.

Performance analysis of cooperative spectrum sensing using double dynamic threshold

&lt;p&gt;Increased use of wireless technologies and in turn more utilization of available spectrum is subsequently leading to the increasing demand for wireless spectrum. This research work incorporates spectrum sensing detection consisting of a double dynamic threshold followed by cooperative type spectrum sensing. The performance has been analyzed using two modulation schemes, quadrature-amplitude-modulation (QAM) &amp;amp; binary-phase-shift-keying (BPSK). Improved probability of detection has been witnessed using the double dynamic threshold where a comparison of average values of local decision (LD) and the observed value of energy (EO) has been considered instead of using direct values of local decisions and energy. Further, the probability-of-detection ( ) is found to be better with QAM as compared to the BPSK. From the results, it has been observed that the detection of primary users is also affected by the number of samples. The simulation environment considered for this work is MATLAB and the performance of cooperative spectrum sensing for 500 and 1000 samples with -9db and -12 SNR by considering different false alarm values i. e 0.1,0.3 and 0.5 has been analyzed. The further scope shall be to enhance the primary user detection by considering different QAM schemes and different SNRs.&lt;/p&gt;

Enhanced capsule generative adversarial network for spectrum and energy efficiency of cooperative spectrum prediction framework in cognitive radio network

AbstractCognitive radio network (CRN) is an intellective technology that frequently monitors the unused licensed spectrum in a particular frequency band. When more count of secondary users enters inside the channel may results in collisions. So, the secondary user may not sense the state of inactive and busy state of the primary user resulting reduces the spectrum efficiency (SE) and energy efficiency (EE) of CRN. To overcome these issues, an enhanced capsule generative adversarial network (ECGAN) for spectrum and energy efficiency of cooperative spectrum prediction (CSP) framework in CRN is proposed to improve the channel occupancy. Initially, CRN system is designed to improve the spectrum efficiency and energy efficiency. Then, channel state of primary user is predicted by using the coupled hidden Markov model (CHMM) and ECGAN predictors. Hence, the correlation among the secondary user predict the next channel status, like AND, OR, majority rule fusion methods. The prediction error, spectrum efficiency, energy efficiency based these methods are compared with CHMM and ECGAN predictors as the process of channel occupancy based on. Also, the effect of busy and idle state prediction errors on SE is detected. The proposed method is simulated in MATLAB. Then, the performance of the proposed method shows higher energy efficiency 13.86%, 24.84%, 34.75%, compared with existing methods, such as spectrum and energy efficiency of CSP in CRN (SEE‐CSP‐CRN‐HMM), (SEE‐CSP‐CRN‐MLP), differential evolution based machine learning scheme for secure cooperative spectrum sensing system (SEE‐CSP‐CRN‐EML), cooperative spectrum sensing for primary user detection in CRN utilizing support vector machine (SEE‐CSP‐CRN‐SVM), respectively.

CNN-SVM Spectrum Sensing in Cognitive Radio Based on Signal Covariance Matrix

Among the basic components of cognitive radio technology is spectrum sensing technology. Cooperative spectrum sensing is also a crucial factor research directions of spectrum sensing. Due to the problems of channel fading and hidden nodes, the spectrum sensing based on a single node cannot meet the requirements of the cognitive radio system for the primary user (PU) detection performance. A spectrum sensing algorithm is based on CNN-SVM and covariance matrix. In this algorithm, the covariance matrix of the signal is input into the convolutional neural network for feature extraction. Secondly, the result of the fully connected layer is taken as the input of the support vector machine for training to obtain a classifier, and finally, spectrum sensing is performed. The simulation outcomes demonstrate that the algorithm suggested in this research works outperform other algorithms in this paper. When the signal-to-noise ratio is -14 dB, the false alarm probability is 0.1, the number of secondary users is 32, and the detection probability reaches 0.81.

Analysis of the Impact of Detection Threshold Adjustments and Noise Uncertainty on Energy Detection Performance in MIMO-OFDM Cognitive Radio Systems.

Due to the capability of the effective usage of the radio frequency spectrum, a concept known as cognitive radio has undergone a broad exploitation in real implementations. Spectrum sensing as a core function of the cognitive radio enables secondary users to monitor the frequency band of primary users and its exploitation in periods of availability. In this work, the efficiency of spectrum sensing performed with the energy detection method realized through the square-law combining of the received signals at secondary users has been analyzed. Performance evaluation of the energy detection method was done for the wireless system in which signal transmission is based on Multiple-Input Multiple-Output—Orthogonal Frequency Division Multiplexing. Although such transmission brings different advantages to wireless communication systems, the impact of noise variations known as noise uncertainty and the inability of selecting an optimal signal level threshold for deciding upon the presence of the primary user signal can compromise the sensing precision of the energy detection method. Since the energy detection may be enhanced by dynamic detection threshold adjustments, this manuscript analyses the influence of detection threshold adjustments and noise uncertainty on the performance of the energy detection spectrum sensing method in single-cell cognitive radio systems. For the evaluation of an energy detection method based on the square-law combining technique, the mathematical expressions of the main performance parameters used for the assessment of spectrum sensing efficiency have been derived. The developed expressions were further assessed by executing the algorithm that enabled the simulation of the energy detection method based on the square-law combining technique in Multiple-Input Multiple-Output—Orthogonal Frequency Division Multiplexing cognitive radio systems. The obtained simulation results provide insights into how different levels of detection threshold adjustments and noise uncertainty affect the probability of detection of primary user signals. It is shown that higher signal-to-noise-ratios, the transmitting powers of primary user, the number of primary user transmitting and the secondary user receiving antennas, the number of sampling points and the false alarm probabilities improve detection probability. The presented analyses establish the basis for understanding the energy detection operation through the possibility of exploiting the different combinations of operating parameters which can contribute to the improvement of spectrum sensing efficiency of the energy detection method.

A novel primary user detection using OFDM internal structures on Raspberry Pi

A novel primary user detection using OFDM internal structures on Raspberry Pi

A novel primary user detection using OFDM internal structures on Raspberry Pi

A novel primary user detection using OFDM internal structures on Raspberry Pi

Joint optimisation techniques for trade-off aware spectrum sensing in cognitive radio network

Cognitive radio (CR) network is considered a promising domain to enhance spectrum efficiency to access underutilised frequency bands. However, due to the influence of channel fading and shadowing, accuracy in primary user (PU) detection by CR gets hampered. This paper designs a joint optimisation technique for spectrum sensing in CR network to optimise energy, delay, and throughput with increased sensing accuracy. Initially, simple energy detection is exhibited for sensing the presence of PU in band. Further, the algorithm is developed to achieve an energy-throughput trade-off, and delay-throughput trade-off. Hence, the optimisation algorithm for detecting energy, reducing delay, and enhancing throughput are developed to optimise complete sensing performance. Furthermore, the joint optimisation model assists in acquiring trade-offs amongst energy, delay, and throughput. The assessment of the technique is performed using delay, energy, and throughput. Moreover, the software-defined radio (SDR) configuration is performed for validating the result.

Reinforcement learning-based cooperative sensing in cognitive radio networks for primary user detection

Reinforcement learning-based cooperative sensing in cognitive radio networks for primary user detection

Reinforcement learning-based cooperative sensing in cognitive radio networks for primary user detection

Reinforcement learning-based cooperative sensing in cognitive radio networks for primary user detection

PU-DetNet: Deep Unfolding Aided Smart Sensing Framework for Cognitive Radio

Spectrum sensing in cognitive radio (CR) paradigm can be broadly categorized as analytical-based and data-driven approaches. The former is sensitive to model inaccuracies in evolving network environment, while the latter (machine learning (ML)/deep learning (DL) based approach) suffers from high computational cost. For devices with low computational abilities, such approaches could be rendered less useful. In this context, we propose a deep unfolding architecture namely the Primary User-Detection Network (PU-DetNet) that harvests the strength of both: analytical and data-driven approaches. In particular, a technique is described that reduces computation in terms of inference time and the number of floating point operations (FLOPs). It involves binding the loss function such that each layer of the proposed architecture possesses its own loss function whose aggregate is optimized during training. Compared to the state-of-the-art, experimental results demonstrate that at SNR= -10 dB, the probability of detection is significantly improved as compared to the long short term memory (LSTM) scheme (between 39% and 56%), convolutional neural network (CNN) scheme (between 45% and 84%), and artificial neural network (ANN) scheme (between 53% and 128%) over empirical, 5G new radio, DeepSig, satellite communications, and radar datasets. The accuracy of proposed scheme also outperforms other existing schemes in terms of the F1-score. Additionally, inference time reduces by 91.69%, 90.90%, and 93.15%, while FLOPs reduces by 62.50%, 56.25%, 64.70% w.r.t. LSTM, CNN and ANN schemes, respectively. Moreover, the proposed scheme also shows improvement in throughput by 56.39%, 51.23%, and 69.52% as compared to LSTM, CNN and ANN schemes respectively, at SNR = -6 dB.

Location-aware spectrum sensing for cognitive visible light communications over multipath channels

Accurate spectrum sensing is crucial for cognitive visible light communication (CVLC). However, due to multiple light emitting diodes (LEDs) and indoor reflections, a CVLC channel often shows a multipath characteristic which is difficult for a secondary user (SU) to estimate before spectrum sensing, thereby leading to a degraded sensing accuracy. To tackle this issue, we propose a location-aware spectrum sensing scheme for the CVLC system, where its multipath channel can be effectively estimated based on the location information of an SU. By considering different a priori information, we propose three primary user (PU) detection algorithms. Specifically, we first propose a location-aware likelihood-ratio-test detection (LLD) algorithm to achieve an optimal sensing performance by using a priori knowledge of signal and noise variances. To reduce the computational complexity of LLD, we propose a more efficient location-aware semi-blind detection (LSD) algorithm, which requires no a priori knowledge of signal variance. Further, to tackle noise uncertainty, we propose a location-aware near-blind detection (LND) algorithm, which does not require either signal or noise variance. To evaluate the performance of the proposed algorithms, we also develop their respective analytical models based on the multipath channel. It is found that the analytical models can accurately match the results obtained by simulations, and LLD and LSD can improve detection probability by 10% compared with the conventional energy detection (ED) scheme. Moreover, by incorporating LSD with cyclic-prefix-based detection (CD), the detection probability can be further improved by ∼20% and 8% compared with ED and CD, respectively. Finally, LND outperforms all the other algorithms when the noise uncertainty is over 0.5 dB, and it is found that both LSD and LND are robust to receiver tilt and LSD is robust to location error.