Throughput Maximization Using an SVM for Multi-Class Hypothesis-Based Spectrum Sensing in Cognitive Radio

We consider a cognitive radio (CR) network that makes opportunistic access to a spectrum licensed to the primary users. During its operation, the CR network carries out spectrum sensing on a frame-by-frame basis to detect active primary users, thereby avoiding interfering with them. We formulate a collision-throughput tradeoff problem which, based on the sensing time requirement and the traffic pattern of primary users, finds optimal value for the frame duration of CR operation so that the throughput of the CR network is maximized, yet the collision probability of the primary users is not greater than a threshold. We derive the theoretical formula for achievable throughput of CR network and find the optimal solution for frame duration. Computer simulations are presented to evaluate the performance of our approach.

Energy detection based spectrum sensing has been proposed and studied widely for primary user (PU) signal detection in the literature. With the help of multiple secondary users (SU) in the cognitive radio network, various cooperative sensing schemes are investigated to enhance the energy detection performance. However, the impacts of noise power fluctuating effects on the detection performance in multipath fading and user cooperative sensing scenarios are seldom evaluated in relevant literature. In this paper, a modified version of the classic energy detection model is presented, where the noise power uncertainty (U) is introduced into threshold setting. With a reasonable approximate derivation, we provide accurate predictions of SNRwall constraints imposed by U in Rayleigh fading environments. In addition, by employing a simple hard decision fusion rule, the collaborative gain in spectrum sensing performance is also quantified with respect to U. Our analysis and numerical results confirm that collaboration can significantly improve the spectrum sensing performance in a noise power fluctuating environment.

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.

The main aim of the Spectrum Sensing (SS) in a Cognitive Radio system is to distinguish between the binary hypotheses H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> : Primary User (PU) is absent and H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> : PU is active. In this paper, Machine Learning (ML)-based hybrid Spectrum Sensing (SS) scheme is proposed. The scattering of the Test Statistics (TSs) of two detectors is used in the learning and prediction phases. As the SS decision is binary, the proposed scheme requires the learning of only the boundaries of H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> -class in order to make a decision on the PU status: active or idle. Thus, a set of data generated under H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> hypothesis is used to train the detection system. Accordingly, unlike the existing ML-based schemes of the literature, no PU statistical parameters are required. In order to discriminate between H0-class and elsewhere, we used a one-class classification approach that is inspired by the Isolation Forest algorithm. Extensive simulations are done in order to investigate the efficiency of such hybrid SS and the impact of the novelty detection model parameters on the detection performance. Indeed, these simulations corroborate the efficiency of the proposed one-class learning of the hybrid SS system.

Cooperative spectrum sensing (CSS) has been considered as a powerful approach to improve the utilization of scarce radio spectrum resources. However, spectrum sensing data falsification (SSDF) can hugely degrade the achievable detection accuracy of the primary user (PU) channel status in cognitive radio (CR) networks and thus requires learning from past CR network environments. Using the supervised learning (SL) method, a large number of labels about the real state of the channel availability need to be acquired. In a practical setting, however, obtaining off-the-shelf label data is a challenging undertaking because it necessitates cooperation among PUs and secondary users (SUs) and increases communication overhead in CR networks. Motivated by this, we solve the issue of label data scarcity in CR networks in this paper, and propose a semi-supervised learning-enabled two-stage framework for CSS against SSDF attack, in which combines the superior performance of semi-supervised support vector machine (S3VM) and the fast convergence of K-means. At last, simulation results demonstrate that S3VM is equipped with a greater detection performance than support vector data description (SVDD), and the same detection performance as support vector machine (SVM), especially it also the robustness and excellent performance using a small amount of labeled data.

Research works on cognitive radio networks (CRNs) together with energy harvesting (EH) promise to address the spectrum scarcity and limited battery power problems on the wireless communication nodes. While radio frequency (RF) signal of primary user (PU) can be used in EH, its absence (non-transmission state) offers an unused spectrum for the availability of the secondary user (SU) data transmission. Spectrum sensing (SS) process that detects the presence or the absence of PU signal, occupies a considerable slot in time-frame and energy consumption in wireless nodes, hence, both SU throughput and battery power get reduced. To address the problems, this work explores a support vector machine (SVM) based PU activity (transmit/non-transmit mode) prediction without accomplishing any SS task in the system operation. Thereafter, a Deep Q-networks (DQN) based energy and spectrum efficient routing strategy is suggested to maximize the CRNs throughput. The proposed joint EH and routing scheme improve the sum throughput, spectrum and energy efficiencies by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 33.48\%$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 34.24\%$ </tex-math></inline-formula> , and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 30.76\%$ </tex-math></inline-formula> , respectively over the conventional dedicated SS based approach.

In recent years, channel-hopping-based medium access control protocols have been proposed to improve the capacity in a decentralized multichannel cognitive radio (CR) network without using extra control channels. Each CR user has to stochastically follow a default channel-hopping sequence in order to locate a channel and conduct its frame transmission. In this paper, theoretical analysis is conducted on the probability of channel availability and the average frame delay for primary users (PUs) by considering the impact caused by imperfect sensing of CR users and imperfect synchronization between the primary and CR networks. According to the proposed analytical model with realistic considerations, an optimal channel-hopping sequence (OCS) approach is designed for the CR users based on a dynamic programming technique. It is designed by exploiting the optimal load balance between channel availability and channel utilization within the delay constraints of PUs. By adopting the OCS approach, maximum aggregate throughput of CR users can be achieved while considering PU's quality-of-service (QoS) requirements. Moreover, in addition to the paired CR networks, the logical partition problem that occurs in generalized CR networks will also be addressed. This problem can severely degrade the aggregate throughput due to the decreased probability of connectivity between CR users, especially in a CR network with heavy traffic. Therefore, both wake-up successive contention (WSC) and wake-up counter-reset successive contention (WCSC) algorithms are proposed to increase the number of negotiations by both exploring the blind spot of imperfect sensing and amending the contention mechanisms between CR users. Compared to conventional channel-hopping sequences, numerical results illustrate that the proposed approaches can effectively maximize aggregate throughput for CR users under the QoS requirements of PUs.

In this study, a new spectrum sensing method for cognitive radio (CR) is proposed for primary user (PU) signals which are modelled as random with unknown power. It is also assumed that, being dynamic, the PU might exhibit multiple status changes in the channel within the sensing period. As a novel approach for detecting dynamic and randomly modelled PUs, we propose a cumulative sum based weighted energy detector (CuS-WED) that only utilizes the samples between the last status change point (LSCP) of the PU and the end of the sensing period contrary to using all the observed data samples in the sensing period. It is shown that the proposed sensing strategy provides noticeable performance improvement in detecting dynamic and randomly modelled PUs. Probability of detection performance of the proposed sensing scheme is investigated for strictly one, at most two, and at most three status changes of the PU. Based on simulation results, it is concluded that the proposed method performs better than the other existing techniques, especially for low probability of false alarm values.

Spectrum sensing ranks amongst cognitive radio's most critical and demanding activities. For radio resource networks to use idle licensed bandwidth, secondary users need to identify the primary user's signal precisely. Cognitive radio networks (CRNs) have been proposed to improve the efficiency of spectrum use to meet the increasing demand for wireless bandwidth; they allow channel communication between secondary (unlicensed) and primary (licensed) users without interruption. A secondary user in a CRN must at all times track the presence of a primary user's signal to prevent the primary user (PU) from attempting to intervene whether the statistical variance of the PU data traffic would boost the prediction outcome provided by multiple current teaching algorithms in estimating the PU's average OFF time and the need for the best adaptive learning to predict accurate PU activities. The objective of the work has been modeled to identify the PUs signal. First, the data has been created via quadrature phase shift keying modulation and quadrature amplitude modulation by convolutional neural network (CNN). A dataset with different orthogonal frequency‐division (OFDM) multiple access signals is generated. Then cyclo‐stationary functions are extracted using the accumulation process fast Fourier Transform. This research aims to use glow worm swarm optimization (GWSO) to automatically check for the optimal design of CNNs without any manual work being required. Also, k‐Nearest Neighbors (kNN) replaces Softmax, as it outperforms conventional CNN architectures with noisy signals. A comparison of the proposed classifier with another machine learning classifier, namely support vector machine, Decision tree (DT), clearly shows that SVM and DT are outperformed by the improved convolution neural network‐glow worm swarm optimization (ICNN‐GWSO) classification. Results of simulation show that the improved scheme outperforms conventional spectrum sensing systems, both in fading and non‐fading settings, where efficiency is measured using measures such as chance, the total likelihood of error, and the ability to optimize data transfer possibilities.

In cognitive radio networks, secondary users need to access the licensed spectrum while neither disturbing the transmissions of the primary users nor introducing much interference to other secondary users. Hence, proper power controlling is of paramount importance to ensure efficient system operation. The previous work on power control in cognitive networks mainly focuses on the interactions among a set of secondary users. The primary users' influence is rarely considered. In this paper, a new power control game model is formulated, by including the primary users into the player set of the game. The primary users are rewarded for sharing their licensed spectrum with secondary users by setting a reasonable interference cap for the secondary transmissions. However, we ensure a minimum required Quality of Service (QoS) for primary users by severely penalizing their utility if their required transmission quality is not met. Simultaneously, the secondary users achieve energy efficient transmissions while not introducing much interference to both the primary and other secondary users. We prove that the proposed cognitive power control game has a unique Nash equilibrium (NE). The numerical examples show that the proposed game theoretical power control algorithm can provide the secondary users with energy efficient transmissions and the primary users with reasonable monetary rewards, while not compromising their required QoS.

In this paper, one of the most important aspect of the Cognitive Radio i.e. Spectrum Sensing is considered. A new technique based on Multiple Input Multiple Output (MIMO) antennas is proposed for Spectrum Sensing in Cognitive Radio. This paper shows the results which prove that the use of MIMO antennas increases the Probability of Detection and decreases the Probability of Missed Detection of the target. Thus, this technique increases the chance to exploit spectrum resources more efficiently, so that spectrum scarcity problem is alleviated.

In cognitive radios, the first and crucial task is to detect the spectrum holes or the presence of primary user by employing spectrum sensing. However, there exist two kinds of detection errors (i.e., miss-detection error and false-alarm error), which degrade the sensing performance severely. Aiming at minimizing the probability of total detection errors, based on the law of total probability we propose a novel target function (which also denotes the so-called comprehensive sensing performance metric) for optimization of sensing parameters. According to such established target function, we then apply the method of extremum seeking to develop the optimization of time-bandwidth product and sensing threshold jointly for local spectrum sensing with energy detection. Moreover, we further derive the closedform expression of the optimal fusion rule for cooperative spectrum sensing by performing the discrete difference operation on the target function. Finally, the optimal area of joint parameters selection for local spectrum sensing and the effect of different detecting channel conditions on the optimization for cooperative spectrum sensing are evaluated numerically.

Cooperative spectrum sensing is envisaged to increase reliability in cognitive radio networks. Several users cooperate to detect the availability of a wireless channel. In this framework, some malicious users may affect the decisions taken by the cognitive users by providing false information, introducing spectrum sensing data falsification (SSDF). In this paper, the effect of multiple malicious users on the energy efficiency of a cognitive radio network is given. A low-overhead security protocol is proposed to address SSDF attacks under a trade-off between energy efficiency and security. An analytical study is provided to set the optimal number of security bits required to maximize energy efficiency. Simulation results show a considerable improvement on the achievable energy efficiency, and the optimal number of bits explicitly depends on the selected fusion rule, the number of malicious users and the number of legitimate users.

With tremendous growth in wireless services, the demand for radio spectrum has significantly increased. However, spectrum resources are scarce and most of them have been already licensed to existing operators. Recent studies have shown that despite claims of spectral scarcity, the actual licensed spectrum remains unoccupied for long periods of time. Thus, cognitive radio (CR) systems have been proposed in order to efficiently exploit these spectral holes, when licensed primary users (PUs) are not present. CRs or secondary users (SUs) are wireless devices that can intelligently monitor and adapt to their environment, hence, they are able to share the spectrum with the licensed PUs, operating when the PUs are idle. In this chapter, we discuss spectrum sensing and access in cognitive radio networks. Depending on the nature of the activities in the network, different categories of MAB problems have been formulated. Solutions and regret bounds are provided.

Motivated by the low utilization of the licensed spectrum across many frequency bands, sensing-based oppor- tunistic spectrum access has recently emerged as an alternative to the outdated exclusive spectrum access policy. Under this new paradigm, a secondary (unlicensed) user monitors a primary (licensed) frequency band for a given and opportunistically transmits if it does not detect any ongoing licensed operations. Evidently, selection of the parameters involves balanc- ing a tradeoff between the speed and the quality with which the secondary user senses the licensed band. With the average throughput as the performance criterion, we obtain the parameters so as to optimize the performance of the secondary user while providing the primary user with its desired level of interference protection. I. INTRODUCTION As evidenced by recent measurements, many frequency bands across the licensed spectrum are significantly under- utilized (1), (2). This finding suggests that the spectrum scarcity, as perceived today, is largely due to the inefficient fixed frequency allocations rather than the physical shortage of the spectrum and has led the regulatory bodies to consider the opportunistic access to the temporally/spatially unused licensed bands (a.k.a. the white spaces) as a means to improve the efficiency of spectrum usage. In the absence of cooperation or signalling between the primary licensee and the secondary users, spectrum availability for the opportunistic access may be determined by direct spectrum where the secondary user monitors a licensed band for a given sensing time and opportunistically transmits if it does not detect any ongoing licensed operations. This approach is particularly appealing due to its low deployment cost and its compatibility with legacy primary users and is being considered for inclusion in the upcoming IEEE 802.22 standard for opportunistic access to the TV spectrum (3). Due to their ability to autonomously detect and to react to the changes in the spectrum usage, secondary users equipped with the spectrum capability may be considered as a primitive form of the cognitive radio (4). Design of any scheme involves balancing a tradeoff between the quality and the speed of through an appropriate selection of the time. As we shall illustrate, in the context of spectrum sensing, may be fine- tuned to enhance the secondary users' perceived quality-of- service (QoS) as long as the regulatory constraint for the protection of the primary users against harmful interference is satisfied. In particular, we will obtain the optimum times at different stages of the spectrum to maximize the average throughput of the secondary user. In this paper, simple energy detection (a.k.a. radiometry) (5) is chosen as the underlying detection scheme. In general, when some information about the structure of the primary signal is available, ad hoc feature-detectors offer a better performance (6). We note, however, that the methodology employed in this paper may be applied to optimize different spectrum sensors once the quality is characterized in terms of the time. The remainder of this paper is organized as follows. The regulatory constraints on spectrum are described in the following section. Section 3 provides an overview of the energy-based spectrum sensing. The optimum times for different stages of the spectrum are derived in Section 4. Finally, this paper is concluded in Section 5.