Search

The world is progressing at a rapid rate, and with continuous advancements in technology, the need for cooperative spectrum sensing has evolved. Cooperative spectrum sensing is an approach used to enhance detection performance, wherein secondary users collaborate with each other to sense the spectrum and identify spectrum holes. As technology improves over time, the spectrum becomes increasingly allocated to primary users. The role of cooperative spectrum sensing in cognitive radio fulfills the essential requirement of protecting primary users from harmful interference. The progressive evolution of technology has led to a reduction in available spectrum, prompting the emergence of the concepts of cognitive radio and cooperative spectrum sensing. To begin with, this paper introduces the fundamental application of cooperative spectrum sensing algorithms, encompassing primary and secondary users' models and an energy model. Subsequently, three machine learning algorithms, namely K-means, support vector machine, and logistic regression, are explained, and their schematics are presented in detail. The proposed model uses supervised and unsupervised learning techniques to develop a cooperative spectrum sensing framework. It compares the performance of three machine learning algorithms, K-means clustering, logistics regression and support vector machine. The comparison is based on accuracy, recall and precision metrics, and the results show that the K-means clustering algorithm has better performance than the other two algorithms. The findings highlight the superiority of the K-means clustering algorithm over logistic regression and support vector machine. [ABSTRACT FROM AUTHOR]

ABSTRACT The spectrum sensing is a major significant task in cognitive radio networks (CRNs) to avoid the unacceptable interference to primary users (PUs). Here, the threshold value determines the effectiveness of spectrum sensing and regarded as a sensing system. The fixed threshold used by the current energy detection‐based spectrum sensing (SS) techniques does not provide sufficient safety for the main users. The threshold is determined by lowering the complete probability of decision error in addition to these guidelines. Therefore, an energy detection using nonparametric amplitude quantization optimized with arithmetic optimization algorithm for enhanced spectrum sensing in CRNs (ED‐NAQ‐AOA‐SS CRN) is proposed in this paper to acquire the ideal threshold for decreasing the total error probability. The proposed method achieves greater probability of detection of 99.67%, 98.38%, 92.34%, and 97.45%, lower settling time of 98.33%, 89.34%, 83.12%, and 88.96%, and lower error rate of 93.15%, 91.25%, 79.90%, and 92.88% compared with existing techniques, like intelligent spectrum sharing and sensing in CRN with adaptive rider optimization algorithm (AROA), a novel technique for spectrum sensing in CRN utilizing fractional gray wolf optimization with the cuckoo search optimization (GWOCS), and adaptive neuro‐fuzzy inference scheme depending on cooperative spectrum sensing optimization in CRNs (ANFIS). [ABSTRACT FROM AUTHOR]

ABSTRACT A cooperative device‐to‐device (D2D) communication system with cognitive radio (CR) enabled that integrates ambient backscattering (AB) and energy harvesting is studied in this paper. Cellular network (CN) resources are exploited by D2D users. The transmission between source and destinations is hybrid (overlay or underlay) based on sensing information. In overlay, cellular users (CUs) are absent in the given channel; thus, relay transmits with high power based on available harvested energy. On the other hand, in underlay CUs are present, and they communicate with each other via base station (BS); thus, in D2D network, relay transmits with controlled power using nonorthogonal multiple‐access (NOMA) technique. In contrast to the conventional hybrid transmission, in the proposed work, relay adopts the ambient backscatter communications (ABCs) in underlay because RF signal of BS is available. The relay first backscatters the source information using available RF signal of BS; thereafter, it uses its own battery for controlled transmission and keeps the interference to the CU receiver within a tolerable limit. Spectrum sensing (SS) and a transmit power limitation are taken into consideration while developing analytical models for throughput, outage, and energy efficiency (EE). An algorithm is proposed to optimize the relay transmit power in the underlay scenario. Through meticulous simulation, the network performance in terms of throughput, outage, and EE is examined for hybrid transmission with ABCs. [ABSTRACT FROM AUTHOR]

Spectrum sensing (SS) technology is essential for cognitive radio (CR) networks to effectively identify and utilize idle spectrum resources. Due to the influence of noise characteristics in the channel, providing accurate sensing results is challenging. In order to improve the performance of SS under non-Gaussian noise and overcome the limitations of existing methods that are mostly based on a single feature, we propose a novel time-frequency cross fusion network (TFCFN). Specifically, we utilize gated recurrent units (GRU) to capture long-term dependencies in the time domain on the original signals, meanwhile, we perform a fast Fourier transform (FFT) on the original signals to obtain the frequency domain information, and subsequently use convolutional neural networks (CNN) to extract the local spatial features in the frequency domain. Ultimately, these time-domain and frequency-domain features are dynamically fused through a cross-attention mechanism to construct more comprehensive and robust features for signal classification. We use generalized Gaussian distribution (GGD) as the noise model and reconstruct the RadioML2016.10a dataset to explore the performance under various noise conditions. The experimental results show that compared with the baseline methods, TFCFN exhibits better detection ability and maintains lower complexity in both Gaussian and non-Gaussian noise environments. Notably, when the shape parameter of GGD is set to 0.5 and the signal-to-noise ratio (SNR) of the received signal is -16dB, it can maintain the probability of false alarm ( P f ) of 10% while still ensuring the probability of detection ( P d ) of over 90%. [ABSTRACT FROM AUTHOR]

A binary artificial rabbit optimization algorithm is proposed to maximize the cognitive radio (CR) network performance and fairness among users by addressing the issue of low spectrum resource utilization during the CR spectrum allocation process. Building upon the graph coloring spectrum allocation model, Sigmoid is introduced to transform the algorithm solution space. Incorporating the Lévy flight strategy for adaptive step size adjustment enhances the algorithm's flexibility and convergence accuracy. Moreover, a selective opposition strategy based on Spearman's coefficient is integrated into the algorithm to improve population diversity and convergence accuracy through reverse learning. Applying the binary artificial rabbit optimization algorithm to the CR spectrum allocation problem in the same CR network environment, we compare network efficiency and inter-user fairness through simulation experiments with the artificial rabbits optimization algorithm seagull optimization algorithm and particle swarm optimization algorithms. The experimental results show that the binary artificial rabbit optimization algorithm has higher convergence performance and global exploration ability, improves the overall network efficiency and user fairness of CR networks, and alleviates the problem of low spectrum resource utilization. [ABSTRACT FROM AUTHOR]

The adoption of UAVs in defence and civilian sectors necessitates robust communication networks. This paper presents a routing protocol for Cognitive Radio Unmanned Aerial Vehicles (CR-UAVs) in Flying Ad-hoc Networks (FANETs). The protocol is engineered to optimize route selection by considering crucial parameters such as distance, speed, link quality, and energy consumption. A standout feature is the introduction of the Central Node Resolution Factor (CNRF), which enhances routing decisions. Leveraging the Received Signal Strength Indicator (RSSI) enables accurate distance estimation, crucial for effective routing. Moreover, predictive algorithms are integrated to tackle the challenges posed by high mobility scenarios. Security measures include the identification of malicious nodes, while the protocol ensures resilience by managing multiple routes. Furthermore, it addresses route maintenance and handles link failures efficiently, cluster formation, and re-clustering with joining and leaving new nodes along with the predictive algorithm. Simulation results showcase the protocol's self-comparison under different packet sizes, particularly in terms of end-to-end delay, throughput, packet delivery ratio, and normalized routing load. However, superior performance compared to existing methods, particularly in terms of throughput and packet transmission delay, underscoring its potential for widespread adoption in both defence and civilian UAV applications. [ABSTRACT FROM AUTHOR]

The growing demand for wireless applications has led to an increasing need for efficient spectrum utilization. Cognitive radio networks (CRNs) emerge as a promising solution to address this demand. CRNs enable unlicensed users to access the available spectrum without causing interference to primary users. However, ensuring the accurate identification of primary user emulation (PUE) attackers is critical due to the sensitivity of these networks. In this paper, an active ensemble neural network-based approach with FlightSpa optimization is employed to enhance the detection of aggressive users. Leveraging neural network (NN) classifiers, the proposed method implicitly captures complex nonlinear relationships between dependent and independent variables, allowing for a comprehensive analysis of network nodes and precise identification of malicious users. Given the network’s inherent complexity, accurately identifying malevolent users is a challenging task, which the active neural network effectively addresses. The classifier’s parameters are tuned optimally by applying a hybrid FlightSpa optimization strategy, leading to successful optimization and reliable output generation. The performance reveals that the active NN-based FlightSpa optimization achieves impressive results, with detection rates for malicious users in CRNs reaching the highest values of 84.085% for accuracy, 82.351% for sensitivity and 85.820% for specificity with 50 nodes. These outcomes surpass those obtained using earlier techniques, marking a significant advancement in the field of cognitive radio network security. [ABSTRACT FROM AUTHOR]

The basic requirement for cognitive radio networks is that secondary users sense spectrum to detect whether primary users have arrived the system. However, imperfect sensing always occurs due to the interference of unreliable channel and signal, which can negatively affect the performance of both primary users and secondary users. To better study the practical problems of cognitive radio networks, we combine queueing theory with it. In this paper, we consider cognitive radio networks with imperfect sensing based on retrial queues with feedback and priority. Using the method of probability generating function, the performance measures are solved. A bi-objective optimization model is established to minimize both system cost and users' sojourn time. From the perspective of secondary users, we analyze both individual equilibrium strategy and socially optimal strategy. Meanwhile, in order to eliminate the differences between two strategies, we impose admission fees on secondary users and obtain the optimal pricing strategy. [ABSTRACT FROM AUTHOR]

Summary: In this paper, we investigate the effectiveness of incremental relaying (IR)‐based underlay cognitive nonorthogonal multiple access (CNOMA) system with maximal ratio combining (MRC). Considering imperfect successive interference cancelation (i‐SIC), we derive analytical expressions for the outage probabilities experienced by the secondary users (SUs) and the system outage probability of the IR‐CNOMA system. We also derive the analytical expression for the outage probability of far SUs by considering IR‐CNOMA system with MRC. The maximum transmit power constraint of the secondary nodes and the interference threshold constraint of the primary receiver are considered in the analysis. We also evaluate the system throughput of IR‐CNOMA network. We also derive outage and throughput expression for the traditional cooperative relaying‐based CNOMA system (CR‐CNOMA) for the purpose of comparison. We derive analytical expression for the asymptotic outage probability and the asymptotic system outage probabilities of the SUs of IR‐CNOMA system. Further, we derive expressions for the optimal power allocation (OPA) factor at the secondary source that minimizes the asymptotic system outage probability of the IR‐CNOMA system. Through numerical and simulation investigations, we demonstrate that the system outage probability significantly lowers when the OPA factor is selected based on the criteria outlined in this work. It can be observed from the results that the IR‐CNOMA system outperforms the conventional CR‐CNOMA system in terms of outage probabilities and throughput. The outage probability is further reduced when IR‐CNOMA system with MRC is considered for the far SUs. [ABSTRACT FROM AUTHOR]

Automatic modulation recognition (AMR) stands as a pivotal operation within industrial cognitive radio systems. State-of-the-art real-valued convolutional neural networks (CNNs) have innovated modulation recognition but view complex signal components as separate real inputs, impairing signal phase integrity and model interpretability. This paper presents an innovative AMR method known as the CVCNN-LSTM. Our study aims to leverage CNNs in conjunction with a long short-term memory network (LSTM) to harness the strengths of both networks while mitigating their weaknesses. This approach enhances the learning of original I/Q signal features, thus improving recognition performance. The proposed network is composed of multiple LFLBs and an LSTM layer. Each LFLBs is primarily composed of a convolutional layer and an average pooling layer, which play a key role in enabling local correlation learning and hierarchical correlation extraction. Recognizing that I/Q data inherently possesses a complex-valued structure, we advocate using end-to-end complex-valued CNN and complex-valued LSTM instead of dual-channel real-valued networks for modulation recognition. We conduct experiments to evaluate complex-valued networks and demonstrate that our method surpasses current state-of-the-art approaches in the AMR field. [ABSTRACT FROM AUTHOR]

Copyright of Journal of National University of Defense Technology / Guofang Keji Daxue Xuebao is the property of NUDT Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Cognitive radio networks offer an alternative to low spectral availability in some frequency bands due to their high demand for frequency channels. This article proposes to improve the spectral assignment based on the combinative distance-based assessment multicriteria algorithm. The metrics obtained are compared with a simple additive weighting algorithm and a RANDOM selection. To establish the algorithm's performance, five quality-of-service metrics are used: number of handoffs, number of failed handoffs, average bandwidth, average throughput, and cumulative average delay. From the analysis of the results obtained, combinative distance-based assessment (CODAS) presented the best result for the cost metrics with the lowest levels, and for the benefit metrics, the highest levels were obtained. [ABSTRACT FROM AUTHOR]

Higher energy consumption and poor spectrum sharing were identified as the common problems in spectrum sharing at cognitive radio networks. So to overcome these issues in the present research, a novel Horse herd-based Elman spectrum sharing model is developed. The current study found lower energy consumption and a higher spectrum sharing rate in less time. Subsequently, a channel was created for sharing the spectrum over the primary and secondary users. After that, through the proposed model, the sharing system is monitored. Moreover, the spectrum is transmitted until the false alarm is detected. The system stops the users' spectrum sharing function if a false alarm is detected. However, the spectrum is shared with every user in the designed channel. Then a case study is developed to explain the working process of the model. After that, the performance of the proposed design is detailed, and in the comparison analysis, the performance of the proposed model is compared with the existing models. Consequently, the performance of the proposed model is validated based on the throughput, outage probability, bit error rate, energy consumption, data transmission, and spectral efficiency. [ABSTRACT FROM AUTHOR]

In recent days, mobile traffic prediction has become a prominent solution for spectrum management-related operations for the next-generation cellular networks in Cognitive Radio (CR) applications. To achieve this, the binary dataset has been created from the captured data by monitoring the spectrum activities of nine different Long Term Evolution (LTE) frequency channels. We propose a Long Short Term Memory (LSTM) based Spectrum Occupancy Prediction (SOP) approach for modelling infrastructure-based cellular traffic systems. The different types of LSTM models, such as Convolutional, Convolutional Neural Network (CNN), Stacked, and Bidirectional have been generated via offline training and tested for the created binary datasets. Moreover, the prediction performance evaluation of the generated LSTM models has been calculated using Mean Absolute Error (MAE). The pro- posed LSTM-based SOP model has achieved 2.5% higher prediction accuracy than the Auto-Regressive Integrated Moving Average (ARIMA) statistical model, accurately aligning the traffic trend with the actual samples. [ABSTRACT FROM AUTHOR]

In the realm of green communications, the focus is on achieving high spectrum efficiency and low energy consumption. This paper addresses the crucial goal of reducing energy usage in green cognitive radio networks (CRNs) during communication between secondary users (SUs) and primary users (PUs). This paper proposed an energy consumption optimization model (ECOM) for green CRN utilizing collaborative spectrum sensing thereby minimizing the environmental impact and prolonging the operational lifetime of devices. The collaborative spectrum sensing proved its role in optimizing the energy consumption in the green CRN. An energy-efficient scheduling algorithm is implemented in ECOM, in which the SUs can be scheduled to perform their sensing process in a time-division manner to reduce energy consumption. Applied collaborative spectrum sensing serves as a valuable resource for researchers, network operators, and policymakers seeking to balance the increasing demand for wireless communication services with the imperative of sustainability. The simulation results and mathematical proof emphasize that ECOM demonstrates reduced energy usage and increased average effective throughput when compared to other recent models. [ABSTRACT FROM AUTHOR]

Compressive sensing (CS) has been widely used to sense the wideband spectrum with fewer measurements by taking advantage of radio spectrum underutilization. As new smart devices, such as IoT devices, smart home devices, and wearables, use batteries and have limited memory, more research is needed to reduce the overuse of cognitive radio (CR) resources through spectrum sensing. To reduce the number of compressive measurements required for spectrum recovery, researchers proposed approaches like weighted and sequential compressive sensing. In this paper, we estimate the primary user's (PU) behavior statistics and use the estimated information in a novel weighted sequential compressive spectrum sensing approach. Our proposed approach can reduce and adapt the number of measurements and the sensing time to the changing number of active channels in a dynamically changing wideband spectrum. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) has revolutionized the connectivity of physical devices, leading to an exponential increase in multimedia wireless traffic and creating substantial demand for radio spectrum. Given the inherent scarcity of available spectrum, Cognitive Radio (CR)-assisted IoT emerges as a promising solution to optimize spectrum utilization through cooperation between cognitive and IoT nodes. Unlicensed IoT nodes can opportunistically access licensed spectrum bands without causing interference to licensed users. However, energy constraints may lead to reduced cooperation from IoT nodes during the search for vacant channels, as they aim to conserve battery life. To address this issue, we propose a Punishment-reward-based Cooperative Sensing and Data Forwarding (PR-CSDF) approach for IoT data transmission. Our method involves two key steps: (1) distributing sensing tasks among IoT nodes and (2) enhancing cooperation through a reward and punishment strategy. Evaluation results demonstrate that both secondary users (SUs) and IoT nodes achieve significant utility gains with the proposed mechanism, providing strong incentives for cooperative behaviour. [ABSTRACT FROM AUTHOR]

Summary By transmitting concurrently signals of different users, non‐orthogonal multiple access (NOMA) meliorates spectral efficiency. Moreover, intelligent reflective surface (IRS) is introduced as a bridge between a NOMA transmitter and a distant NOMA receiver. This role enhances spectral efficiency and communication reliability. Furthermore, energy scavenging underlay networks (ESUNs) enable unlicensed transmitters to operate using scavenged energy on licensed spectrum allotted to licensed transmitters, enhancing both spectral and energy efficiencies. Although IRS‐assisted NOMA further meliorates spectral efficiency and communication reliability for traditional ESUNs, it faces challenges like fading severity and nonlinear energy scavenging (nlES), which directly impact scavenged energy and communication reliability. Thence, the paper assesses three pivotal performance metrics—energy efficiency, outage probability, and throughput—for IRS‐assisted NOMA in ESUNs, shortly RaNOehU, considering realistic factors such as relatively‐general Nakagami‐m fading and nlES. Demonstrative results reveal that the quantity of reflective elements, fading severity, and nlES dramatically influence the performance of RaNOehU. Notably, appropriate selection of power/time division parameters and desired data rate can hinder complete outage and attain optimum performance. In addition, the proposed IRS‐assisted NOMA is shown to outperform its reference, which is IRS‐assisted orthogonal multiple access, emphasizing the advantages of NOMA. [ABSTRACT FROM AUTHOR]

With the development of wireless communication technology, the number of devices accessing the communication network is increasing. This paper addresses critical issues such as low spectrum resource utilization and the energy constraints of devices. The investigation focuses on the system performance of the shared relay Cognitive Radio Non-Orthogonal Multiple Access (CR-NOMA) network based on Simultaneous Wireless Information and Power Transfer (SWIPT) network model. Unlike the existing CR-NOMA model in which the secondary network user do not participate in the transmission of information from the primary network, we consider the secondary network near user as shared relay. The shared relay utilizes SWIPT technology to harvest energy using a nonlinear energy harvesting model. Additionally, the shared relay assists in transmitting information from the primary network base station to primary user, as well as from the secondary network base station to far user of the secondary network. Subsequently, we conduct a series of simulations to analyze the effects of Signal-to-Noise Ratio (SNR), power distribution factor, interference threshold, and time-switching (TS) factor on system performance. Furthermore, we compare and analyze the performance of our proposed network model against CR-NOMA network across three dimensions: outage probability, throughput, and energy efficiency. Our results demonstrate that the proposed network model exhibits superior outage performance and enhances user throughput compared to the CR-NOMA network. Additionally, it demonstrates improved energy efficiency compared to the shared relay CR-NOMA network, leading to an overall improvement in network performance. [ABSTRACT FROM AUTHOR]

Active switching components are employed to propose a novel compact dual-band matching network (MN) with an adapted frequency in this paper. The design method involves the integration of active switching devices, stepped impedance resonator (SIR), and microstrip transmission line into the frequency response of a dual-band MNs. The PIN diode is employed as an active switching device in the proposed structure to increase the bandwidth of the MNs and attain the desired frequency. The dual-band MNs that are being proposed are fabricated and designed on the Rogers RO4350B substrate for use in Wireless Local Area Networks (WLANs). The simulation results demonstrate a high degree of congruence with the measurements. The maximal insertion losses and return losses in the first band are -1.3 and -24 dB, respectively. In the second band, they are -2.2 and -20.69 dB. The primary benefits of the proposed dual-band MNs are their high attenuation between two passbands, frequency matching capability, suitable return loss, and low insertion loss. [ABSTRACT FROM AUTHOR]

To address the common issues in traditional convolutional neural network (CNN)-based spectrum sensing algorithms in cognitive radio networks (CRNs), including inadequate signal feature representation, inefficient utilization of feature map information, and limited feature extraction capabilities due to shallow network structures, this paper proposes a spectrum sensing algorithm based on a short-time Fourier transform (STFT) and residual attention dense network (RADN). Specifically, the RADN model improves the basic residual block and introduces the convolutional block attention module (CBAM), combining residual connections and dense connections to form a powerful deep feature extraction structure known as residual in dense (RID). This significantly enhances the network's feature extraction capabilities. By performing STFT on the received signals and normalizing them, the signals are converted into time–frequency spectrograms as network inputs, better capturing signal features. The RADN is trained to extract abstract features from the time–frequency images, and the trained RADN serves as the final classifier for spectrum sensing. Experimental results demonstrate that the STFT-RADN spectrum sensing method significantly improves performance under low signal-to-noise ratio (SNR) conditions compared to traditional deep-learning-based methods. This method not only adapts to various modulation schemes but also exhibits high detection probability and strong robustness. [ABSTRACT FROM AUTHOR]

Aiming at the problem of difficulty in allocating channels efficiently for secondary users (SU) in complex and variable cognitive radio systems, this paper proposed a high throughput channel allocation protocol, TKMA protocol. The protocol constructed an utility matrix of user and channel based on primary user (PU) activities, SU real-time service demands, and channel conditions, etc., allocated channels with the objective of maximizing the total utility value of the SU system under the premise of guaranteeing the communication quality of the PUs, and solved the problem by using the improved Kuhn-Munkras algorithm and polling scheduling. To evaluate the performance of the protocol, it established a universal multi-user multi-channel cognitive radio system model, described the packet transmission process using queuing theory, and derived the performance indicators of SU through Markov steady-state solution. The experimental results show that the use of TKMA protocol yields better results in terms of total SU system throughput, average delay, average queue length, rejection rate and other metrics compared with the previously proposed simplification algorithm and the traditional equitable random assignment protocol, which proves the effectiveness of the proposed protocol and system model. [ABSTRACT FROM AUTHOR]

The research presents innovative methods for spectrum sensing in 5G networks, using the Sqrt-Loglogish convolutional neural network (SL-CNN) and hidden orthogonal intuitionistic fuzzy Markov model (HOIFMM). The proposed methods aim to tackle issues related to detecting principal user signals accurately, mitigating interference, and efficiently utilizing the spectrum in wideband spectrum environments due to their diverse and everchanging characteristics. The Sqrt-Loglogish CNN improves spectrum sensing by addressing static threshold dependency and potential overfitting. The HOIFMM offers a complex framework for predicting sparsity levels and primary user patterns. The results highlight the effectiveness of the suggested techniques in differentiating primary user signals from noise and interference, resulting in enhanced interference management tactics and overall network performance. MATLAB simulation is performed and compared the proposed methods performance with existing state-of-the-art methods such as CNN, deep neural network (DNN), long short-term memory (LSTM) and artificial neural networks (ANN). The proposed method has outperformed existing methods in terms of sensitivity, accuracy, and precision. Future endeavors include improving these methods, investigating sophisticated machine learning algorithms, and doing realworld validations to guarantee scalability and resilience in various 5G deployment situations. This research advances the spectrum sensing capabilities in 5G networks, potentially improving efficiency, reliability, and quality of service. [ABSTRACT FROM AUTHOR]

The performance of the cognitive non-orthogonal multiple access (NOMA) systems with random locations is examined in the present paper. More precisely, the positions of the secondary receivers (SRs) are randomly distributed inside a disk with the centre at the secondary transmitter (ST). The primary networks (PNs), on the other hand, are not necessarily inside the transmission range of the ST. Two user selection schemes namely, path-loss-based and channel-gain-based approaches, are adopted to serve two SRs according to the power domain NOMA technique. Additionally, the impact of the imperfect hardware impairments (HIs) on both transmitters and receivers is also taken into consideration. Furthermore, a recently proposed transmit power allocation at the ST is employed to concurrently maximize the system performance of the secondary networks (SNs) and to strictly protect the quality-of-service (QoS) of the PNs. The correctness of the developed mathematical frameworks is corroborated by simulation results based on the Monte Carlo method. Finally, we unveil that there is no scheme that always outperforms another. Thus, an adaptive scheme can be proposed to optimize the networks' performance. Our findings also reveal that the outage probability (OP) under the harmful effect of HI can be effectively mitigated by increasing the transmit power of the ST. Additionally, increasing the transmit power of the PT is also beneficial for the SNs with the adopted transmit power allocation at the ST. [ABSTRACT FROM AUTHOR]

Although the design of spectrum switching has been studied, little is known about how random user movement affects the handoff. This issue can occur when a user moves to a new location. In this paper, the authors present a framework that verifies the necessity of spectrum handoff to improve the performance of the system by employing machine learning (ML) techniques. Some of these include the Logistic Regression, KNN Algorithm, SVM Algorithm, Naïve Bayes Classifier, Decision Tree Classification and Random Forest Algorithm. The system is implemented on a real-time dataset where all the users are separated in power domain using the concept of non-orthogonal multiple access (NOMA) technique. The dataset values are prepared using a software-defined radio experimental setup, which is used to analyse the performance of various ML techniques in terms of confusion matrix, specificity, precision, F1_score, sensitivity and accuracy. The performance of proposed system is compared with the literature and shown a significant improvement that proves the evidence of our findings. [ABSTRACT FROM AUTHOR]

This letter introduces the design of an estimator for parameters of Middleton Class A noise using its canonical formula and classical numerical methods. The main focus is to acquire parameters to characterize communication channels in intelligent systems or those based on cognitive paradigms. A comprehensive analysis of the first-order characteristics of the Middleton Class A noise model is conducted to establish the foundational understanding necessary for developing the presented estimator model, named impulsive noise estimator with minimization methods (INEMM). Subsequently, the method is introduced, substantiated, and compared to various established estimators concerning precision and complexity. Results show a distinct advantage in terms of overall performance. [ABSTRACT FROM AUTHOR]

This letter focuses on the field-programmable gate array (FPGA) implementation of a Class A Middleton noise estimator, aiming to enhance its efficiency and performance. The inherent algorithm of the estimator undergoes strategic enhancements, leveraging median approximations. This endeavor leads to the development of a more streamlined and expeditious architecture. The research not only introduces the refined architecture but also conducts a comparative analysis of its attributes. The outcomes of this investigation show the benefits of algorithmic optimization, as the execution times achieved in hardware significantly surpass those attainable through software-based implementation. This underscores the practicality of the algorithmic refinement and also the notable advantages of the FPGA-based execution in terms of computational speed. [ABSTRACT FROM AUTHOR]

Spectrum sensing is pivotal in cognitive radio (CR), a burgeoning technology for optimizing radio spectrum utilization. Traditional spectrum sensing techniques like energy detection, matching filter, and cyclic stationary detection have been proposed, which rely on prior knowledge and models. These techniques suffer from challenging issues such as missed detection and false alarms, which impede the effective utilization of the spectrum. Inaccurate assumptions or limited knowledge can hinder detection. To tackle these challenging issues, we propose a novel deep learning-oriented spectrum sensing (DLoSS) technique and highlight the use of deep neural networks (DNNs) for cooperative spectrum sensing (CSS) model. Specifically, we propose a “DLSpectSenNet,” a DLoSS-based model, utilizes structural information from incoming modulated signals for spectrum sensing. Particularly, we combine convolutional neural network (CNN) and long-short-term memory (LSTM) network in series, extracting hidden spatial information and temporal data, respectively. The simulation results using the RadioML2016.10b dataset, show the proposed DLSpectSenNet’s improved detection performance, especially under low SNR conditions, surpassing traditional cooperative algorithms. It outperforms previous models, enabling improved spectrum detection. [ABSTRACT FROM AUTHOR]

Automatic modulation classification models based on deep learning models are at risk of being interfered by adversarial attacks. In an adversarial attack, the attacker causes the classification model to misclassify the received signal by adding carefully crafted adversarial interference to the transmitted signal. Based on the requirements of efficient computing and edge deployment, a lightweight automatic modulation classification model is proposed. Considering that the lightweight automatic modulation classification model is more susceptible to interference from adversarial attacks and that adversarial training of the lightweight auto‐modulation classification model fails to achieve the desired results, an adversarial attack defense system for the lightweight automatic modulation classification model is further proposed, which can enhance the robustness when subjected to adversarial attacks. The defense method aims to transfer the adversarial robustness from a trained large automatic modulation classification model to a lightweight model through the technique of adversarial robust distillation. The proposed method exhibits better adversarial robustness than current defense techniques in feature fusion based automatic modulation classification models in white box attack scenarios. [ABSTRACT FROM AUTHOR]

Summary Perfect detection of unused/underutilized frequency band of the primary user (PU) in the cognitive radio communication systems is an essential component for its efficient utilization by the secondary/cognitive user (CU) without any interference. In the cooperative spectrum sensing scenario, the presence or absence of PU is determined by analyzing the decisions of each CU engaged in the collaboration. In this paper, we have emphasized on the significance of spectrum sensing duration in the optimal threshold selection approach in the cooperative spectrum sensing scenarios for a range of the received PU signal signal‐to‐noise ratio (SNR) from low to high values. The sensing duration parameter is taken as ts = 2.5 ms in the higher SNR environment, whereas in the lower SNR environment when the optimal threshold condition fails, ts is kept dynamic to satisfy minimum target values of sensing performance parameters. We have computed the spectrum sensing performance parameters and throughput using the same approach and compared the results with those obtained from the constant false‐alarm rate (CFAR), constant detection rate (CDR), and minimized error probability (MEP) threshold selection approaches. The spectrum sensing performance computed by the proposed approach is significantly improved as compared with the other reported approaches. At SNR = −24 dB, global probability of false alarm Qf = 0.002 and global probability of detection Qd = 0.999 are achieved. However, the throughput is relatively less for low SNR values. [ABSTRACT FROM AUTHOR]

A minimal interrupted communication link set-up is the primary objective of the medium access control (MAC) layer. The MAC layer is responsible for accessing the communication channel. At MAC, the control information is shared through the control channel to select a collision-free data channel. Therefore, the sharing of control information prior to data channel selection plays a pivotal role in achieving desired QoS in cognitive radio (CR) technology. Random data channel selection is used in conventional CR scenarios. The random selection lacks in performance due to the unpredictability of primary user (PU) appearances in the data channels. This unpredictability in PU appearances will increase the number of handoffs in the scenario and sometimes may cause communication link failure. These adverse effects can be avoided if a common data channel selection scheme is implemented. The common data channel selection scheme selects a channel that has fewer chances of PU appearance. In the reported research paper, the work has been carried out in two parts. In the first part, a primary user free channel list (PCL) is formed at each node. Each node shares its PCL with all neighbour nodes. The sharing of PCL helps in implementing a common data channel selection scheme among nodes. The selected common channel is used for data transmission. The conventional OSI layered structure is used to observe the performances. In the second part, a cross-layer framework is proposed to improve the performance. The cross-layering has been done between the network and MAC layer. The proposed cross-layer framework helps in sharing PCL in an efficient manner. The need for an extra beacon to transfer PCL is avoided in the proposed cross-layer design. The performance of control overhead and average delay becomes better. The simulation results show that the proposed cross-layer design approach performs satisfactorily in terms of packet delivery ratio, average throughput, average delay, average energy consumption and control overhead. The observed results are also tested in the worst scenarios. The proposed scheme has applications in small throughput scenarios like cognitive radio aided sensor networks. [ABSTRACT FROM AUTHOR]

Cognitive radio networks (CRN) enable wireless devices to sense the radio spectrum, determine the frequency state channels, and reconfigure the communication variables to satisfy Quality of Service (QoS) needs by reducing energy utilization. In CRN, spectrum sensing is an essential process that is highly challenging and can be addressed by several traditional techniques, such as energy detection, match filtering, etc. For now, the current models' performance is impacted by the comparatively low Signal to Noise Ratio (SNR) of recognized signals and the insignificant quantity of traditional signal samples. This research proposals a new spectral sensing technique for cognitive radio networks (SST-CRN) that addresses the drawbacks of predictable energy detection models. With the use of a deep belief network (DBN), the suggested model contributes to accomplish a nonlinear threshold based on the chicken swarm algorithm (CSA). The proposed DBN enabled SST-CRN technique goes through two phases in a organized process: offline and online. Throughout the offline phase, the DBN model is methodically trained on pre-gathered data, developing the aptitude to identify problematic patterns and examples from the spectral features of the radio environment. This stage involves extensive feature extraction, validation, and model development to ensure that the DBN can professionally represent complicated spectral dynamics. Additionally, online spectrum sensing is conducted during the real communication phase to enable real-time adaptation to dynamic changes in the spectrum environment. Offline spectrum sensing is typically performed during a devoted sensing period before actual communication begins. When combined with DBN's deep learning capabilities and CSO's innate nature-inspired algorithms, a synergistic framework is created that enables CRNs to explore and allocate incidences on their own with astonishing accuracy. The proposed solution considerably improves the spectrum efficiency and resilience of CRNs by harnessing the power of DBN, which leads to more effective resource utilization and less interference. The Simulation results show that our proposed strategy produces more accurate spectrum occupancy assessments. The result parameters such as probability of detection, SNR of -24dB, the SST-CRN perfect has increased a developed Pd of 0.810, whereas the existing methods RMLSSCRN-100 and RMLSSCRN-300 have accomplished a lower Pd of 0.577 and 0.736, respectively. Our deep learning methodology uses convolutional neural networks to automatically learn and adapt to dynamic and complicated radio environments, improving accuracy and flexibility over classic spectrum sensing approaches. Future research might focus on improving CSO algorithms to better optimize the spectrum sensing process, enhancing the reliability of DBN-enabled sensing techniques. [ABSTRACT FROM AUTHOR]

Traditional heterogeneous networks (HetNets) are constrained by their hardware design and configuration. These HetNets have a limited ability to adapt to variations in network dynamics. Software-defined radio technology has the potential to address this adaptability issue. In this paper, we introduce a software-defined radio (SDR)-based long-term evolution licensed assisted access (LTE-LAA) architecture for next-generation communication networks. We show that with proper design and tuning of the proposed architecture, high-level adaptability in HetNets becomes feasible with a higher throughput and lower power consumption. Firstly, maximizing the throughput and minimizing power consumption are formulated as a constrained optimization problem. Then, the obtained solution, alongside a heuristic solution, is compared against the solutions to existing approaches, showing our proposed strategy is drastically superior in terms of both power efficiency and system throughput. This study is then concluded by employing artificial intelligence techniques in multi-objective optimization, namely random forest regression, particle swarm, and genetic algorithms, to balance out the trade-offs between maximizing the throughput and power efficiency and minimizing energy consumption. These investigations demonstrate the potential of employing the proposed LTE-LAA architecture in addressing the requirements of next-generation HetNets in terms of power, throughput, and green scalability. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel system integrating reconfigurable intelligent surfaces (RISs) with cognitive radio (CR) technology, presenting a forward-looking solution aligned with the evolving standards of 6G and beyond networks. The proposed RIS-assisted CR networks operate with a base station (BS) transmitting signals to two users, the primary user (PU) and secondary user (SU), through direct and reflected signal paths, respectively. Our mathematical analysis focuses on deriving expressions for SU in the RIS-assisted CR system, validated through Monte Carlo simulations. The investigation covers diverse aspects, including the impact of the signal-to-noise ratio (SNR), power allocations, the number of reflected surfaces, and blocklength variations. The results provide nuanced insights into RIS-assisted CR system performance, highlighting its sensitivity to factors like the number of reflectors, fading severity, and correlation coefficient. Careful parameter selection, such as optimizing the configuration of reflectors, is shown to prevent a complete outage, showcasing the system's robustness. Additionally, the work suggests that the optimization of reflectors configuration can significantly enhance overall system performance, and RIS-assisted CR systems outperform reference schemes. This work contributes a thorough analysis of the proposed system, offering valuable insights for efficient performance evaluation and parameter optimization in RIS-assisted CR networks. [ABSTRACT FROM AUTHOR]

Summary: Cognitive Radio Ad Hoc Networks (CRAHNs) are an essential method for resolving conflicts between extreme spectrum scarcity and rapid traffic increase while maintaining high‐quality service for consumers. However, the coexistence of primary and secondary users represents a critical challenge for reasonable resource allocation in order to provide a sustaining system performance. Many approaches have been developed to efficiently allocate resources; however, these methods are currently limited by things like user collision, strange traffic networks, and high data transmission error rates. To address these constraints, this paper proposes a policy‐configured reinforcement learning‐based ad hoc network (AHN) model. To obtain the ideal policy configuration for the network, the system first models the cognitive radio (CR) network, in which nodes are initialised and grouped employing the Link Reliability K‐Means clustering Algorithm (LR‐KMA). The available spectrum was then detected and separated into multiple bands utilizing coherent‐based detection (CBD) and signal source identification employing the Parzen–Rosenblatt Window‐based Restricted Boltzmann Machine (PRW‐RBM). Next, the learning model for the resource allocation process employs the Weibull Distribution‐based Blue Monkey Optimization (WD‐BMO) approach to pick the relevant bands. Finally, the experimental results were analyzed in order to evaluate the proposed resource allocation model's performance in CRAHNs. When compared with previous findings, the proposed method improves resource utilization by 5%, the proposed model achieves a 7% higher throughput, and the PRW‐RBM's accuracy improves classification accuracy by 1.07%. [ABSTRACT FROM AUTHOR]

Summary: Effective, reliable, and rapid spectrum sensing in concordance with maximum throughput efficiency is necessary at the cognitive radio network to maximize the network's utilization. An adaptive coalition using Kullback–Leibler divergence (KLD)‐based cooperative spectrum sensing scheme (KLDCSS) with a nominal sensing time to optimize the spectral efficiency is presented in this paper. In a coalition‐based spectrum sensing scheme, cooperating secondary users (SUs) are allocated to different coalitions to sense different Primary User (PU) channels. The same cognitive users allocated to sense the primary user channels are allotted to detect in different sensing phases. In the proposed sensing scheme, the cooperating SUs are adaptively allocated to different coalitions to maximize the throughput efficiency. Collaborating cognitive users with a standard signal‐to‐noise ratio and distinct energy levels that are characterized by the localized probabilities of detection and false alarm is considered. An adaptive coalition and user allocation algorithm are proposed to optimize the average opportunistic throughput and diminish the sensing overhead. The expression for throughput efficiency in terms of average throughput and sensing accuracy is derived. In this scheme, the SU computes the log‐likelihood ratio (LLR) from the acquired sample and forwards the same to the fusion center. The fusion center then accumulates the LLR values computed by the SU to estimate the likelihood of spectrum availability. Simulation results highlighting the competitive edge of the proposed scheme, with higher throughput efficiency over various numbers of SUs, coalitions, and signal‐to‐noise ratios, are presented. [ABSTRACT FROM AUTHOR]

Spectrum sensing technology can effectively improves the utilization of spectrum resources by empowering unauthorized users to access the idle spectrum opportunely. This paper proposes a new spectrum sensing algorithm to determine whether a signal exists under impulsive noise by only using the sampled data. First, this method uses the Similarity Measure of samples to construct the test statistic. Theoretical analysis shows that it can suppress the influence of impulsive noise, realize accurate signal detection and improve the reliability of the system with fewer samples. Then, the probability distribution of test statistic is investigated through statistical theory analysis, under the both case of the absence and the presence of primary user signals, respectively, and the probabilities of detection and the false alarm are derived. After that, we give the detection threshold for a given false-alarm probability. Finally, experimental simulation confirms the effectiveness of the proposed algorithm and the correctness of theoretical analysis. [ABSTRACT FROM AUTHOR]

Current cutting-edge solutions to the spectrum shortage problem are unable to meet the growing demand for a limited spectrum. A key dimension beyond state-of-the-art solutions is to exploit the free spectrum more effectively. Although various schemes have been proposed for trading spectrum, few studies have focused on optimal admission of spectrum requests for maximizing service providers' (SP's) profit. Thus, this timely study presents a novel intelligent admission scheme for spectrum requests from the perspective of a non-cooperative game, in which the information of all participants (customers and providers) is incomplete to others, and each player wishes to maximize its benefit. The proposed control admission policy may evict clients in-service to release spectrum for serving certain, e.g., wealthy clients. Evicted clients are compensated using a dynamic strategy that adopts greedy game theory to capture the conflict of interest between SP and evicted users. Simulation experiment results validate and demonstrate the feasibility and efficiency of the proposed scheme, compared to a benchmark reinforcement learning approach and another widely used scheme for admission and eviction control of cognitive radio users. [ABSTRACT FROM AUTHOR]

In wireless sensor networks, the expensive and scarce resources are energy and frequency spectrum. To overcome this scarcity of spectrum, cognitive radio has been introduced in WSNs. The licensed band is utilized by the primary users in cognitive radio whereas the secondary users can use the licensed channels. For improving the network lifetime, overall network scalability, and energy consumption, the clustering technique is used to group the sensor nodes into clusters. Clustering algorithms must be energy efficient because of the difficulties in replacing or recharging the batteries of nodes. While designing the algorithms, more constraints result in the clustering for CRSN as the dynamic spectrum access has been addressed. We propose an energy-efficient fuzzy clustering and congestion control algorithm (EFCCA) in this paper to improve energy efficiency. With the consideration of spectrum availability, queue length, and residual energy, the election of cluster head contributes to the energy efficiency of the algorithm. The active Queue Management algorithm is used to monitor and control the congestion rate. The proposed EFCCA's performance evaluates with the help of comparison with other clustering methods and it shows enhanced performance in energy efficiency and lifetime based on the outcomes of experimental investigation. [ABSTRACT FROM AUTHOR]

In cognitive radio network (CRN), effective spectrum management provides better quality of service. The spectrum is limited but the significance of the spectrum is increasing at each network generation. Due to the ineffective spectrum allocation policies, several researches have indicated that a vast segment of the licensed radio is not viably used. A CRN is an intelligent spectrum utilization innovation that provides better spectrum interface. Spectrum sensing detects unused spectrum in the manner that protects interferences to the authorized users. In principle, the secondary user (SU) receives the primary user (PU) signal and reports it to the fusion center for decision or spectrum allocation. The SU cooperates in the detection of the presence or absence of the PU. This type of spectrum sensing is called cooperative spectrum sensing. However, the significance of this type of spectrum sensing is blurred by the security problems. Malicious users can deliberately report misleading information regarding the presence of the PU. Hence, in this paper, a support vector machine learning algorithm is proposed to statistically learn the behavior of the malicious users and it classifies the legitimate SU and malicious users. A particle swarm optimization algorithm is also integrated to learn the smallest possible distinguishable malicious users' energy report deviation from the legitimate SUs. The probability of detection and energy of detection have been applied to evaluate the contribution of the proposed method. Finally, the simulation results have confirmed that better spectrum management can be derived from the proposed statistical approach. [ABSTRACT FROM AUTHOR]

In the conventional LTE homogeneous network, sufficient transmit power of user equipment (UE) is determined by open-loop power control (OL-PC) and closed-loop power control (CL-PC) schemes. However, in a Het-Net environment, setting the UE’s transmit power requires delicate responsiveness to handle the severe and complicated uplink interference. In this paper, an interference-aware uplink power control mechanism based on Heuristic game theory approach is proposed for devices coexisting in a heterogeneous wireless network. Various wireless constraints like channel response, path loss, fading, shadowing, interference and metrics like SNR, SINR, throughput and bit rates are considered. Uplink power is controlled by suitably selecting the penalization factor (β) based on a simple Heuristic game theory approach considering the possible wireless constraints of each user depending on its location in the cell under consideration. The algorithm is framed in such a way to reduce inter-cell interference, limit transmit power, enhance bit rates and throughput of users. A significant improvement of 5.2% in the user coverage/ distribution is achieved as a result of interference management compared to conventional power control scheme and power control with convex pricing. [ABSTRACT FROM AUTHOR]

In cognitive radio networks, channel assembling allows secondary users (SUs) to expand network capacity and improve spectrum utilization. Scheduling strategies only based on heterogeneous service classification cannot guarantee the delivery priority of vital elastic services in special scenarios such as emergency rescue. Therefore, a priority scheduling strategy with reserved queue (Ps-rq) is proposed in this work. A static factor is defined to classify SUs into elastic services and real-time services based on message type, while a dynamic factor is defined to differentiate high-priority elastic services based on information validity, message correlation and message size. The high-priority users in the interrupted elastic services are placed in the reserved queue to ensure its services. Accordingly, the scheduling algorithm and the dynamic channel access process is presented. A continuous-time Markov chain analysis is conducted and all possible transition states, trigger events, transition rates and transition conditions of the system starting from a general state are derived. Furthermore, evaluation indexes of system performance are obtained. Study cases and simulation results prove that the proposed strategy can enhance network capacity, reduce blocking probability and forced termination probability for secondary users, and notably enhance the performance of high-priority elastic services. In addition, we analyze the characteristics of Ps-rq through a comprehensive comparison with four other schemes. The experiment proves that the Ps-rq strategy can effectively improve the service quality of the vital elastic services on the basis of providing fair scheduling. [ABSTRACT FROM AUTHOR]

The detection and characterisation of electromagnetic signals within a specific frequency range, known as spectrum sensing, plays a crucial role in Cognitive Radio Networks (CRNs). The CRNs aim to adapt their communication parameters to the surrounding radio environment, thereby improving the efficiency and utilisation of the available radio spectrum. Spectrum sensing is particularly important in device‐to‐device (D2D) communication when operating independently of the cellular network infrastructure. The Medium Access Control (MAC) protocol coordinates device communication and ensures interference‐free operation of the CRN coexisting with the primary cellular network. A spectrum sensing strategy at the MAC layer for cognitive D2D communication. The strategy focuses on reducing the overall sensing period allocated at the MAC layer by having each Cognitive D2D User (cD2DU) sense a smaller subset of available channels while maintaining the same sensing time for cellular user detection at the physical layer. To achieve this, the concept of concurrent groups of D2D devices is introduced in proximity, which are formed by using unique IDs of cD2DUs during the device discovery stage. Each concurrent group senses a specific portion of the cellular user band in a shorter time, resulting in a reduced overall sensing period. In addition to mitigating traffic congestion through data diversion from the cellular network, the proposed strategy facilitates the concurrent sensing of multiple channels by cD2DUs within the underutilised cellular user band. This leads to extended data transmission periods, increased network throughput, and effective offloading of the cellular network. The effectiveness of the proposed work is evaluated by considering factors, such as network throughput and transmission time. Simulation results confirm the effectiveness of the approach in improving spectrum utilisation and communication efficiency in multi‐channel Cognitive D2D Networks (cD2DNs). [ABSTRACT FROM AUTHOR]

Cooperative communication through energy harvested relays in Cognitive Internet of Things (CIoT) has been envisioned as a promising solution to support massive connectivity of Cognitive Radio (CR) based IoT devices and to achieve maximal energy and spectral efficiency in upcoming wireless systems. In this work, a cooperative CIoT system is contemplated, in which a source acts as a satellite, communicating with multiple CIoT devices over numerous relays. Unmanned Aerial Vehicles (UAVs) are used as relays, which are equipped with onboard Energy Harvesting (EH) facility. We adopted a Power Splitting (PS) method for EH at relays, which are harvested from the Radio frequency (RF) signals. In conjunction with this, the Decode and Forward (DF) relaying strategy is used at UAV relays to transmit the messages from the satellite source to the CIoT devices. We developed a Multi-Objective Optimization (MOO) framework for joint optimization of source power allocation, CIoT device selection, UAV relay assignment, and PS ratio determination. We formulated three objectives: maximizing the sum rate and the number of admitted CIoT in the network and minimizing the carbon dioxide emission. The MOO formulation is a Mixed-Integer Non-Linear Programming (MINLP) problem, which is challenging to solve. To address the joint optimization problem for an epsilon optimal solution, an Outer Approximation Algorithm (OAA) is proposed with reduced complexity. The simulation results show that the proposed OAA is superior in terms of CIoT device selection and network utility maximization when compared to those obtained using the Nonlinear Optimization with Mesh Adaptive Direct-search (NOMAD) algorithm.

Smart transportation systems implemented through vehicular ad hoc networks (VANET) offer significant potential to improve safety. However, the network faces critical challenges related to security, as well as inadequate spectrum sensing and management. To address these issues, researchers have utilized cognitive radio and machine learning technologies. Although, previous survey studies have provided a valuable foundation for understanding the use of cognitive radio in VANET, not all have systematically investigated its impact on mitigating spectrum sensing and management issues or the role of machine learning in supporting cognitive radio functionality. Furthermore, the effects of security issues on both VANET and cognitive radio enhanced VANET have not been consistently examined. This survey aims to systematically review the application of cognitive radio and machine learning approaches to address the identified challenges in smart transportation networks, offering valuable research opportunities for future investigations. The paper extensively explores state-of-the-art approaches and focuses on: (1) Assessing the impact of cognitive radio and machine learning on spectrum sensing and management in smart transportation networks and (2) Evaluating the impact of security issues on both VANET and cognitive radio enhanced VANET.

Abstract Automatic modulation classification models based on deep learning models are at risk of being interfered by adversarial attacks. In an adversarial attack, the attacker causes the classification model to misclassify the received signal by adding carefully crafted adversarial interference to the transmitted signal. Based on the requirements of efficient computing and edge deployment, a lightweight automatic modulation classification model is proposed. Considering that the lightweight automatic modulation classification model is more susceptible to interference from adversarial attacks and that adversarial training of the lightweight auto‐modulation classification model fails to achieve the desired results, an adversarial attack defense system for the lightweight automatic modulation classification model is further proposed, which can enhance the robustness when subjected to adversarial attacks. The defense method aims to transfer the adversarial robustness from a trained large automatic modulation classification model to a lightweight model through the technique of adversarial robust distillation. The proposed method exhibits better adversarial robustness than current defense techniques in feature fusion based automatic modulation classification models in white box attack scenarios.