Academia.eduAcademia.edu

Outline

Title
Abstract
FAQs
Figures
Introduction and Problem Statement
Neural Networks Review
Preparation Procedure
NN Pattern Selection -Results
Discussions and Conclusions
References
All Topics
Engineering
Electrical and Electronic Engineering

Neural network-based learning schemes for cognitive radio systems

Profile image of Panagiotis DemestichasPanagiotis Demestichas

2008, Computer Communications

https://doi.org/10.1016/J.COMCOM.2008.05.040
visibility

…

description

11 pages

link

1 file

Abstract

Intelligence is needed to keep up with the rapid evolution of wireless communications, especially in terms of managing and allocating the scarce, radio spectrum in the highly varying and disparate modern environments. Cognitive radio systems promise to handle this situation by utilizing intelligent software packages that enrich their transceiver with radio-awareness, adaptability and capability to learn. A cognitive radio system participates in a continuous process, the ''cognition cycle", during which it adjusts its operating parameters, observes the results and, eventually takes actions, that is to say, decides to operate in a specific radio configuration (i.e., radio access technology, carrier frequency, modulation type, etc.) expecting to move the radio toward some optimized operational state. In such a process, learning mechanisms that are capable of exploiting measurements sensed from the environment, gathered experience and stored knowledge, are judged as rather beneficial for guiding decisions and actions. Framed within this statement, this paper introduces and evaluates learning schemes that are based on artificial neural networks and can be used for predicting the capabilities (e.g. data rate) that can be achieved by a specific radio configuration. In particular, the focus in this work is placed on obtaining insight on the behavior of the presented, learning schemes, whereas useful, indicative results from the benchmarking work, conducted in order to design and use an appropriate neural network structure, are also presented and discussed. In the near future, such learning schemes are expected to assist a cognitive radio system to compare among the whole of available, candidate radio configurations and finally select the best one to operate in.

FAQs

sparkles

AI

What are the advantages of using neural networks in cognitive radio systems?add

The paper demonstrates that neural networks enhance decision-making processes in cognitive radios by learning from past experiences, potentially improving configuration selection and data rate predictions.

How does the basic neural network scheme evaluate data rates in cognitive radios?add

The basic scheme uses time-series data with an exponentially weighted moving average to derive target data rates for specific radio configurations, achieving an MSE of 0.0153 during validation.

What enhancements does the extended neural network scheme provide over the basic one?add

The extended scheme incorporates time zone parameters into its analysis, allowing for more accurate data rate predictions by considering contextual factors, with observed mean values varying by zone.

How was the neural network performance assessed in the study?add

Performance was evaluated using MSE metrics on both training and validation datasets, ensuring that the networks could generalize well to unseen data, achieving satisfactory thresholds.

Which neural network architectures were found most effective for cognitive radio applications?add

The study identified a custom feed-forward back-propagation network with two hidden layers, optimizing data rate estimates, yielding a validation MSE of 0.0637.

Figures (17)
Fig. 2. Cognitive radio engine. The approach that is adopted here is that a cognitive radio re- sults from the enhancement of a software radio with cognitive capabilities. Those capabilities are often provided by an intelligent software package called a cognitive engine, as proposed in [7] and depicted in Fig. 2, albeit, according to FCC [8], ‘neither having soft- ware nor being programmable are requirements of a cognitive radio’. Within the cognition cycle, the cognitive engine derives and enforces decisions to the software-based radio by continuously adjusting its parameters, observing and measuring the outcomes and taking actions to move the radio toward some desired opera- tional state [9]. Meanwhile, cognitive radios are capable of learning lessons and storing them into a knowledge base, from where they may be retrieved, when needed, to assist future decisions and ac- tions. A reasoning engine determines which actions are executable This paper contributes in this direction, by developing learning schemes that rely on artificial neural networks and aim at solving
Fig. 2. Cognitive radio engine. The approach that is adopted here is that a cognitive radio re- sults from the enhancement of a software radio with cognitive capabilities. Those capabilities are often provided by an intelligent software package called a cognitive engine, as proposed in [7] and depicted in Fig. 2, albeit, according to FCC [8], ‘neither having soft- ware nor being programmable are requirements of a cognitive radio’. Within the cognition cycle, the cognitive engine derives and enforces decisions to the software-based radio by continuously adjusting its parameters, observing and measuring the outcomes and taking actions to move the radio toward some desired opera- tional state [9]. Meanwhile, cognitive radios are capable of learning lessons and storing them into a knowledge base, from where they may be retrieved, when needed, to assist future decisions and ac- tions. A reasoning engine determines which actions are executable This paper contributes in this direction, by developing learning schemes that rely on artificial neural networks and aim at solving
Fig. 1. Simplified representation of cognitive radio cycle.
Fig. 1. Simplified representation of cognitive radio cycle.
Fig. 4. Transfer functions (a) log-sigmoid transfer function, (b) hyperbolic tangent sigmoid transfer function and (c) linear transfer function Fig. 3. Typical neural network structure.
Fig. 4. Transfer functions (a) log-sigmoid transfer function, (b) hyperbolic tangent sigmoid transfer function and (c) linear transfer function Fig. 3. Typical neural network structure.
cases, multiple types of NNs with a considerable number of adjust- able parameters have been investigated. These trial and error pro- cesses that were conducted in order to derive the best possible network in both basic and extended cases, are described and dis- cussed more analytically in the following sections. value corresponding to the input time sequence within the time window will arise from the following relation‘:
cases, multiple types of NNs with a considerable number of adjust- able parameters have been investigated. These trial and error pro- cesses that were conducted in order to derive the best possible network in both basic and extended cases, are described and dis- cussed more analytically in the following sections. value corresponding to the input time sequence within the time window will arise from the following relation‘:
Test cases examined for basic scheme Table 1
Test cases examined for basic scheme Table 1
Fig. 6. Cumulative distribution functions of input time-series.
Fig. 6. Cumulative distribution functions of input time-series.
Fig. 7. Measured performance (MSE) at both training and validation sequences for all 84 test cases —- basic scheme.
Fig. 7. Measured performance (MSE) at both training and validation sequences for all 84 test cases —- basic scheme.
Fig. 8. Neural network for the basic learning scheme.
Fig. 8. Neural network for the basic learning scheme.
Fig. 9. Measured performance (MSE) of the selected NN in training sequence - basic scheme.
Fig. 9. Measured performance (MSE) of the selected NN in training sequence - basic scheme.
Fig. 11. Weight values per time slot for the extended scheme. Step 1: Temporary target value rj?" is calculated as in Eq. (5) by applying R'”"**t as input. Step 2: The distance (absolute difference), dst,, between the tar- get value at k ie., 72"""? and mean value m, which corresponds to the considered time zone is calculated. Specifically and assuming that M is the set defined in Sub-section 4.2, if the tar- get value r2""™? equals m;¢ M, 1 <i<|M| and the mean value Mz equals m;¢ M, 1 <j <|M|, then the distance dst,, is taken equal to |i—j|, 1< ij<|M|. For example, if r2“'™? equals m, =6 Mbps and the mean value mz equals mz = 12 Mbps, the distance dst,, equals |1 —2|=1; similarly the distance from m, = 6 Mbps to m3 = 24 Mbps equals |1 — 3] = 2, and so forth. Step 3: The new weights f; (where i=1,...,n and nis the num- ber of slots in the time window) are recalculated with the use of an exponentially weighted moving average function, with a smoothing factor a, = 1 —x%k,xe¢(0---1). For example, assuming that M is the set defined in Sub-section 4.2 and that n= 10, Fig. 11 illustrates the weight values per time slot when dst,=1, ..., 5. The above rule is used so that greater distance between the calculated value and the target value should lead to a lower slope of weight decrease. A lower slope, as shown in Fig. 11, gives high importance (high weights) to past observations so as to eliminate the abovementioned distance. The slope of weight decrease is expressed by the smoothing factor a.
Fig. 11. Weight values per time slot for the extended scheme. Step 1: Temporary target value rj?" is calculated as in Eq. (5) by applying R'”"**t as input. Step 2: The distance (absolute difference), dst,, between the tar- get value at k ie., 72"""? and mean value m, which corresponds to the considered time zone is calculated. Specifically and assuming that M is the set defined in Sub-section 4.2, if the tar- get value r2""™? equals m;¢ M, 1 <i<|M| and the mean value Mz equals m;¢ M, 1 <j <|M|, then the distance dst,, is taken equal to |i—j|, 1< ij<|M|. For example, if r2“'™? equals m, =6 Mbps and the mean value mz equals mz = 12 Mbps, the distance dst,, equals |1 —2|=1; similarly the distance from m, = 6 Mbps to m3 = 24 Mbps equals |1 — 3] = 2, and so forth. Step 3: The new weights f; (where i=1,...,n and nis the num- ber of slots in the time window) are recalculated with the use of an exponentially weighted moving average function, with a smoothing factor a, = 1 —x%k,xe¢(0---1). For example, assuming that M is the set defined in Sub-section 4.2 and that n= 10, Fig. 11 illustrates the weight values per time slot when dst,=1, ..., 5. The above rule is used so that greater distance between the calculated value and the target value should lead to a lower slope of weight decrease. A lower slope, as shown in Fig. 11, gives high importance (high weights) to past observations so as to eliminate the abovementioned distance. The slope of weight decrease is expressed by the smoothing factor a.
Fig. 10. Measured performance (MSE) of the selected NN in validation sequence - basic scheme.
Fig. 10. Measured performance (MSE) of the selected NN in validation sequence - basic scheme.
Test cases examined for extended scheme Table 2
Test cases examined for extended scheme Table 2
Fig. 12. Neural network for the extended learning scheme.
Fig. 12. Neural network for the extended learning scheme.
Fig. 13. Measured performance (MSE) of the selected NN in training sequence - extended scheme.
Fig. 13. Measured performance (MSE) of the selected NN in training sequence - extended scheme.
Fig. 14. Measured performance (MSE) of the selected NN in validation sequence - extended scheme.
Fig. 14. Measured performance (MSE) of the selected NN in validation sequence - extended scheme.

Related papers

Technique for Predicting Data Rate for Cognitive Radio Using Neural Network
IJERA Journal

Significant problems confronting wireless communications, are scarcity and deployment difficulty. The deployment problem is a process problem i.e., frequency allocation is fixed and is done so by complex collaboration and coordination between countries and systems, respectively. The limited available spectrum and the inefficiency in the spectrum usage necessitate a new communication paradigm to exploit the existing wireless spectrum opportunistically called cognitive radio technology. In this paper we have proposed a learning technique using neural network to be used in cognitive radio so that it can predict the data rate. Performance of learning scheme has been observed using Matlab simulations and found that Focused Time Delay Neural Networks are good candidate for cognitive radios.

View PDFchevron_right
Learning-Based Approaches for Intelligent Cognitive Radio
Nada Abdel khalek

2020

Learning-Based Approaches for Intelligent Cognitive Radio

View PDFchevron_right
Cognitive radio engine learning adaptation
Martins Olaleye

2016 10th International Conference on Software, Knowledge, Information Management & Applications (SKIMA)

Cognitive radio (CR) system being an intelligencebased communication device has been considered as the next generation emerging technologies to Wireless Communication Systems (WCS). This CR's embedded-intelligent agent is called Cognitive Engine (CE), and is responsible for the dynamic adaptation between the WCS's environment and the radio operational parameters. As a result of CR's intelligence capability, the WCS's quality of service (QoS) and its connectivity operations get enhanced. In order to evaluate the CR engine performance in respect to its learning, timing, and its computational performances. This paper proposes an alternative state-of-the-art enhanced CR learning engine based on Random Neural Network (RNN). Unlike Artificial Neural Network (ANN) systems, RNN establishes strong data generalization, converges faster and produces relatively smaller levels of prediction errors. Subjected to similar environmental conditions, the simulation cumulative results show that the performance of the proposed RNN system is satisfactory and produces 36.895% performance improvement above the ANN learning engine.

View PDFchevron_right
Performance evaluation of artificial neural network-based learning schemes for cognitive radio systems
Kostas Tsagkaris

Computers & Electrical Engineering, 2010

This work investigated the potential of shea butter oil (SBO) as feedstock for synthesis of biodiesel. Due to high free fatty acid (FFA) of SBO used, response surface methodology (RSM) was employed to model and optimize the pretreatment step while its conversion to biodiesel was modeled and optimized using RSM and artificial neural network (ANN). The acid value of the SBO was reduced to 1.19 mg KOH/g with oil/ methanol molar ratio of 3.3, H 2 SO 4 of 0.15 v/v, time of 60 min and temperature of 45 C. Optimum values predicted for the transesterification reaction by RSM were temperature of 90 C, KOH of 0.6 w/v, oil/ methanol molar ratio of 3.5, and time of 30 min with actual shea butter oil biodiesel (SBOB) yield of 99.65% (w/w). ANN combined with generic algorithm gave the optimal condition as temperature of 82 C, KOH of 0.40 w/v, oil/methanol molar ratio of 2.62 and time of 30 min with actual SBOB yield of 99.94% (w/w). Coefficient of determination (R 2 ) and absolute average deviation (AAD) of the models were 0.9923, 0.83% (RSM) and 0.9991, 0.15% (ANN), which demonstrated that ANN model was more efficient than RSM model. Properties of SBOB produced were within biodiesel standard specifications.

View PDFchevron_right
Performance of Cognitive Radio Using Adaptive Neural Fuzzy Inference System
mithra venkatesan

2015

Over the last several years the world of wireless communications has undergone some critical changes, which have brought it at the front position of international research and development interest, finally resulting in the introduction of a huge number of innovative technologies and associated products such as WiFi, WiMax, 802.20, 802.22, wireless mesh networks and software defined radio. Cognitive radio system uses a base of software defined radio technology and makes use of intelligent software packages that enhance their transceivers with the extremely smart properties of selfresponsiveness, flexibility and ability to learn. A cognitive radio system has the capability to change its operating parameters, monitor the results and, finally take actions, that is to say, decide to operate in a specific radio configuration (i.e. radio access technology, carrier frequency, modulation type, etc.), expecting to move the radio toward some optimized operational state. For such a process, lea...

View PDFchevron_right
Comparative Analysis of ANN Techniques for Predicting Channel Frequencies in Cognitive Radio
Adnan Waqar

International Journal of Advanced Computer Science and Applications, 2017

Demand of larger bandwidth increases the spectrum scarcity problem. By using the concepts of Cognitive radio we can achieve an efficient spectrum utilization. The cognitive radio allows the unlicensed user to share the licensed user band. To sense the accessibility of vacant channel and allocation of licensed user band is provided by Machine learning techniques because this decision need to be very fast and accurate. It is based on certain factors (such as Power, Bandwidth, antenna parameters, etc.). In this paper, we used neural network to propose this decision of resource allocation more accurately by providing bandwidth, power, antenna gain, azimuth, angle of elevation and location as a supplements factors to increase the predicting accuracy of Available channel frequencies for secondary user in particular bands. The comparative analysis is done between artificial neural network techniques to determine the maximum decision accuracy in order to design a suitable neural network str...

View PDFchevron_right
Survey of Artificial Intelligence Approaches in Cognitive Radio Networks
yasmina el morabit

Journal of information and communication convergence engineering, 2019

This paper presents a comprehensive survey of various artificial intelligence (AI) techniques implemented in cognitive radio engine to improve cognition capability in cognitive radio networks (CRNs). AI enables systems to solve problems by emulating human biological processes such as learning, reasoning, decision making, self-adaptation, self-organization, and self-stability. The use of AI techniques is studied in applications related to the major tasks of cognitive radio including spectrum sensing, spectrum sharing, spectrum mobility, and decision making regarding dynamic spectrum access, resource allocation, parameter adaptation, and optimization problem. The aim is to provide a single source as a survey paper to help researchers better understand the various implementations of AI approaches to different cognitive radio designs, as well as to refer interested readers to the recent AI research works done in CRNs.

View PDFchevron_right
A Relationship Between Cognitive Information Processing in Learning Theory and Machine Learning Techniques in Cognitive Radios
Zhaneta Tasheva

SOCIETY. INTEGRATION. EDUCATION. Proceedings of the International Scientific Conference

The relationship between cognitivism as learning theory in education and machine learning is characterized in this survey paper. The cognitivism describes how learning occurs through internal processing of information and thus leads to understanding and retention. Cognitive information processing plays an active role to understand and process information that learner receives and relates it to already known and stored within learner’s memory. Thus, the cognitive approach defines learning as a change in knowledge which is stored in learner’s memory, and not a change in learner’s behaviour. In regard with importance of various learning problems to designing cognitive communications systems the two main classification categories of learning techniques are explained. Furthermore, the cognitive radio learning algorithms that have been proposed are described. Finally, the similarities and differences among the principles of learning theories and machine learning are discussed.

View PDFchevron_right
Spectrum Sensing using Probabilistic Neural Networks in Cognitive Radio Network
Pavithra Roy

2015

Prediction of potential idle times of different channels based on the past information allows a cognitive radio to select the best channels for control and data transmission. For a successful cognitive radio, learning is one of the most important factors. Cognitive radio with enhanced Artificial Intelligence (AI) techniques leads to design and development of algorithm that enable computer to learn. The learning mechanisms are capable of utilizing measurements sensed from the environment, gathered experience and stored knowledge to improve the radio operation and to guide decisions and actions. This paper presents the learning schemes that are based on artificial neural networks; in particular, Neural Network (NN) and probabilistic neural network (PNN) . Based on the intelligent radio learning schemes, an effective training approach is proposed. Simulations demonstrate that the novel technique beats existing hierarchical approaches, and present better performance.

View PDFchevron_right
STUDY OF ROLE OF MACHINE LEARNING TECHNIQUES IN RADIO NETWORKS
Journal ijmr.net.in(UGC Approved)

Cognitive radio offers the promise of intelligent radios that can gain from and adapt to their environment. To date, most cognitive radio research has concentrated on policy-based radios that are hard-coded with a rundown of principles on how the radio ought to act in specific situations. Some work has been done on radios with learning engines custom fitted for unmistakable applications. This article depicts a solid model for generic cognitive radio to use a learning engine. The objective is to fuse the consequences of the learning engine into a predicate math-based reasoning engine with the goal that radios can recollect exercises learned previously and acted rapidly. 1. INTRODUCTION

View PDFchevron_right

References (22)

  1. European Radiocommunications Committee (ERC), European table of frequency allocations and utilizations frequency range 9 kHz to 275 GHz, ERC Report 25, January 2002.
  2. Susan Moore, Managing the radio spectrum: hands-on or back-off?, IT Prof 6 (2) (2004) 49-55.
  3. P. Leaves, K. Moessner, R. Tafazolli, D. Grandblaise, D. Bourse, R. Tonjes, M. Breveglieri, Dynamic spectrum allocation in composite reconfigurable wireless networks, Commun. Mag. IEEE 42 (5) (2004) 72-81.
  4. S. Haykin, Cognitive radio: brain-empowered wireless communications, IEEE J. Select. Areas Commun. 23 (2) (2005) 201-220.
  5. J. Mitola et al., Cognitive radio: making software radios more personal, IEEE Pers. Commun. 6 (4) (1999) 13-18.
  6. J. Mitola, Cognitive radio: an integrated agent architecture for software defined radio, Doctor of Technology, Royal Institute of Technology (KTH), Stockholm, Sweden, 2000.
  7. C. Clancy, J. Hecker, E. Stuntebeck, T. O'Shea, Applications of Machine Learning to Cognitive Radio Networks, Wireless Communications, IEEE [see also IEEE Personal Communications], vol. 14, no. 4, August 2007, pp. 47-52.
  8. Federal Communication Commission, Facilitating opportunities for flexible, efficient and reliable spectrum use employing cognitive radio technologies', ET Docket 03-108, Notice of Inquiry and Proposed Rulemaking and Order, May 2003.
  9. B. Le, T.W. Rondeau, C.W. Bostian, Cognitive radio realities, Wirel. Commun. Mob. Comput. 7 (9) (2007) 1037-1048.
  10. A.P. Rewagad, V.L. Soanawane, Artificial neural network based short term load forecasting, TENCON '98, in: Proceedings of the IEEE Region 10 International Conference on Global Connectivity in Energy, Computer, Communication and Control, vol. 2, 1998, pp. 588-595.
  11. H. Demuth, M. Beale, M. Hagan, Matlab Neural Network Toolbox User's Guide, Version 5.1, The MathWorks Inc., 2007.
  12. Rosenblatt, Frank The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain, Cornell Aeronautical Laboratory, Psychological Review, vol. 65, no. 6, 1958, pp. 386-408.
  13. B. Widrow, S.D. Sterns, Adaptive Signal Processing, Prentice-Hall, New York, 1985.
  14. J.L. Elman, Finding structure in time, Cogn. Sci. 14 (1990) 179-211.
  15. J.J. Hopfield, Neural networks and physical systems with emergent collective computational abilities, Proc. Natl. Acad. Sci. USA 79 (1982) 2554-2558.
  16. B. Le, T.W. Rondeau, D. Maldonado, C.W. Bostian, Modulation identification using neural network for cognitive radios, SDR Forum Technical Conference, Anaheim, CA, 2005.
  17. M. Gandetto, M. Guainazzo C.S. Regazzoni, Use of time-frequency analysis and neural networks for mode identification in a wireless software-defined radio approach, Eurasip JASP, Special Issue on Non Linear Signal Processing and Image Processing, vol. 13, 2004, pp. 1778-1790.
  18. NIST/SEMATECH e-Handbook of Statistical Methods, available at http:// www.itl.nist.gov/div898/handbook/, September 2002.
  19. T. Masters, Practical Neural Network Recipes in C++, Academic Press, 1993.
  20. S. Haykin, Neural Networks, A Comprehensive Foundation, second ed., Prentice Hall, Upper Saddle River, NJ, 1999.
  21. B. Hassibi, D. Stork, G. Wolff, Optimal Brain Surgeon and general network pruning, in: Proceedings of the IEEE International Conference on Neural Networks, 1993, pp. 293-299.
  22. E. Coersmeier, M. Hoffmann, H. Bothe, Combining cognitive radio and software radio approach for low complexity receiver architecture, ISART 2007, Boulder, Colorado, US, February 27, 2007.

Related papers

Improved Learning Scheme for Cognitive Radio using Artificial Neural Networks
rita mahajan

International Journal of Electrical and Computer Engineering (IJECE), 2016

The future of wireless system is facing the problem of spectrum scarcity. Number of users is increasing rapidly but available spectrum is limited. The Cognitive Radio (CR) network technology can enable the unlicensed users to share the frequency spectrum with the licensed users on a dynamic basis without creating any interference to primary user. Whenever secondary user finds that primary user is not transmitting and channel is free then it uses channel opportunistically. In this paper cognitive radio with predictive capability using artificial neural network has been proposed. The advantage of such cognitive user is saving of time and energy for spectrum sensing. Proposed radio will sense only that channel which is predicted to be free and channel is selected on the basis of maximum vacant time. Performance has been evaluated in the term of mean square error. The results show that this learning capability can be embedded in secondary users for better performance of future wireless technologies.

View PDFchevron_right
Learning and Adaptation in Cognitive Radios using Neural Networks
Maria Faizan

The estimation of the communication performance achievable with respect to environmental factors and configuration parameters plays a key role in the optimization process performed by a Cognitive Radio according to the original definition by Mitola [1]. In this paper we propose the use of Multilayered Feedforward Neural Networks as an effective technique for real-time characterization of the communication performance which is based on measurements carried out by the device and therefore offers some interesting learning capabilities.

View PDFchevron_right
Recent advances on artificial intelligence and learning techniques in cognitive radio networks
Nadine Abbas

EURASIP Journal on Wireless Communications and Networking, 2015

Cognitive radios are expected to play a major role towards meeting the exploding traffic demand over wireless systems. A cognitive radio node senses the environment, analyzes the outdoor parameters, and then makes decisions for dynamic time-frequency-space resource allocation and management to improve the utilization of the radio spectrum. For efficient real-time process, the cognitive radio is usually combined with artificial intelligence and machine-learning techniques so that an adaptive and intelligent allocation is achieved. This paper firstly presents the cognitive radio networks, resources, objectives, constraints, and challenges. Then, it introduces artificial intelligence and machine-learning techniques and emphasizes the role of learning in cognitive radios. Then, a survey on the state-of-the-art of machine-learning techniques in cognitive radios is presented. The literature survey is organized based on different artificial intelligence techniques such as fuzzy logic, genetic algorithms, neural networks, game theory, reinforcement learning, support vector machine, case-based reasoning, entropy, Bayesian, Markov model, multi-agent systems, and artificial bee colony algorithm. This paper also discusses the cognitive radio implementation and the learning challenges foreseen in cognitive radio applications.

View PDFchevron_right
A Survey on Cognitive Radio Network using Artificial Neural Network
Aamir Z . Shaikh

International Journal of Future Generation Communication and Networking

The emergence of Internet of Things and other applications of wireless communication has resulted in increase of air interference among various wireless devices. In upcoming time we will be connecting more and more devices wirelessly. In addition to an increase in number of devices, many devices also demand higher bandwidth. We have a limited spectrum available for communication and as the demand increases it creates more and more congestion in the available spectrum. Besides this scarcity of spectrum, it has been observed that all available frequencies in this spectrum are not efficiently utilized. Some frequency bands face congestion while others are underutilized. The solution of all these issues is Cognitive Radio. The fundamental theory of cognitive radio deals with the issues mentioned above and provides efficient utilization of available spectrum. In cognitive radio when a frequency is not utilized by primary user (Licensed user), it is allocated to secondary user (Unlicensed user) who can use the frequency until there is no primary user. For searching primary and secondary user we use spectrum sensing. Depending on the type of users and the environment, this spectrum sensing can be a time consuming task which can severely impact the QoS. To deal with this critical issue we use machine learning techniques, which predict spectrum holes in an available frequency band. This in turn reduces spectrum sensing time and power consumed in sensing. Among various Machine learning techniques, Artificial Neural Networks is one of the most popular and widely used technique. Unlike other Machine learning Techniques, Neural Network doesn't require prior knowledge of the system and in most cases it doesn't require the model to be retrained an every instance. These advantages makes it one of the most popular technique for cognitive radios. So far a lot of work has been done on implementing Artificial Neural Network models for predicting the most suitable frequency for a secondary user. In this paper a comprehensive survey has been conducted on various ANN techniques, its comparison with other machine learning techniques and discussion on various learning models to increase the decision making ability of cognitive radio's cognitive engine. ANN uses supervised learning and this paper compares it with other supervised learning techniques (like SVM) and also unsupervised learning techniques and statistical models. The paper provides detailed knowledge about what factors influence the use of ANN in cognitive engines and under certain conditions which ANN technique is most suitable.

View PDFchevron_right
Comparative studies of artificial intelligence techniques in the context of cognitive radio
Badr Benmammar, Fethi Tarik Bendimerad

… Conference on Multimedia …, 2012

Cognitive radio (CR) is considered as the key enabling technology of Next Generation Wireless Systems (NGWS). In this context, CR enables users to dynamically access and to fairly share the spectrum with other users. Our paper describes some artificial intelligence techniques like artificial neural network, metahueristic algorithms and hidden Markov model, these techniques are proposed to provide the cognition capability in a cognitive engine.

View PDFchevron_right

Related topics

  • Distributed Computing
  • Cognitive radio
  • Frequency Modulation
  • Radio Access Technology
  • Electrical and Electronic Engine...

    • Cited by

    Utility-Aware Cognitive Network Selections in Wireless Infrastructures
    Vera A Stavroulaki

    Wireless Personal Communications, 2012

    Operators of wireless infrastructures should maintain their users "always-bestconnected". This concept means that applications should be offered to users at the best possible Quality of Service (QoS) level, taking into account profile, context and policy information. The profiles provide the user requirements and preferences, the terminal capabilities, and the application requirements. The policies provide the objectives, constraints imposed by various stakeholders, for instance the network operator (NO). The context of operation designates relevant applications, available networks and their QoS capabilities. The "alwaysbest-connectivity" concept can be achieved by directing user terminals to the most appropriate networks of the heterogeneous infrastructure of the NO. In this respect, advanced terminal management functionality is required. This paper presents management mechanisms for utility-based cognitive network selections. The utility is used for expressing the user desire for a QoS level. Cognition mechanisms are applied for learning the QoS capabilities of candidate networks, and therefore increasing the reliability and seamlessness of the network selections. Extensive results are provided, which show the behaviour of the scheme in terms of network selections made, and computational effort required for the acquisition of the knowledge.

    View PDFchevron_right
    Cognitive radio: survey on communication protocols, spectrum decision issues, and future research directions
    Edmundo Monteiro

    Wireless Networks, 2012

    View PDFchevron_right
    An overview of learning mechanisms for cognitive systems
    Yiouli Kritikou

    EURASIP Journal on Wireless Communications and Networking, 2012

    Cognitive systems were first introduced by Mitola and in the last decade they have proved to be beneficial in selfmanagement functionalities of future generation networks. The advantages and the way that networks gain benefits from cognitive systems is analysed in this article. Moreover, since such systems are closely related to machine learning, the focus of this article is also placed on machine learning techniques applied both in the network and the user devices side. In particular, celebrating 10 years of cognitive systems, this survey-oriented article presents an extended state-of-the-art of machine learning applied to cognitive systems as coming from the recent research and an overview of three different learning capabilities of both the network and the user device.

    View PDFchevron_right
    Overview of Machine Learning Approaches for Wireless Communication
    melike günay

    2019

    Machine learning is one of the most popular research areas, and it is commonly used in wireless communications and networks. Security and fast communication are among of the key requirements for next generation wireless networks. Machine learning techniques are getting more important day-by-day since the types, amount, and structure of data is continuously changing. Recent developments in smart phones and other devices like drones, wearable devices, machines with sensors need reliable communication within internet of things (IoT) systems. For this purpose, artificial intelligence can increase the security and reliability and manage the data that is generated by the wireless systems. In this chapter, the authors investigate several machine learning techniques for wireless communications including deep learning, which represents a branch of artificial neural networks.

    View PDFchevron_right
    Machine Learning Application in the Hybrid Optical Wireless Networks
    DEEPA NAIK

    Advances in Intelligent Systems and Computing

    Complex problems invariably involve big data. Machine learning (ML) deals with this big data. Machine learning does classification, regression and decision making better than human beings. Large data produced by communication networks can be utilized for deriving efficient optimized solutions. Network performance improves by usage of ML techniques for network design and management. In this article, we have summarized advanced ML techniques for network domain. Further, a survey of ML techniques in several network design problems done, and their performance is discussed.

    View PDFchevron_right
    580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved