This thesis presents a performance evaluation system for cognitive radio. It considers performance as a complex, multi-dimensional function. Typically such a function would take some record of actions as an argument; however, a key contribution of this work is the addition of background information to the domain of the performance function. Including this information generalizes the performance function across many radios and applications, with the additional cost of complicating the domain. Thus the presented evaluation system organizes the domain information into sets. These sets are divided into two categories, one capturing necessary information that is external to the radio and on capturing necessary information that internal to the radio. These categories highlight the fact that neither the true actions nor the true performance is directly observable at the onset of evaluation. This arises because a cognitive radio can only express its actions in terms of the available knobs and meters, which together form the radio's language. Some understanding of this language and its limitations is required to fully understand the radio's expression of its actions. This parallelism of actions and performance suggests implementing the evaluation method as a composite form of the performance function. The composite performance function is made up of two sub-functions, one of which producing action information and one of which producing performance information. Specifically, the first sub-function is used to determine general measures of the actions' influence on performance; these are labeled Measures of Effectiveness. The second sub-function uses these Measures of Effectiveness to determine application specific performance values, called Measures of Performance. This work covers both these measures in detail. Each measure is determined as the result of a neural network based interpolation. This thesis also provides an examination of artificial neural networks in the scope of performance evaluation. Once these concepts are explored, a walk-through evaluation is presented. The four phases are the Setup Phase, the Logging Phase, the Training Phase, and the Evaluation Phase. Each phase is structured to provide the information necessary to determine the final performance. These phases detail the process of evaluation and discuss the realization of concepts explored earlier. This work concludes with a comparative evaluation example that proves the worth of the presented approach. A full evaluation system is outlined by this thesis and the foundational details for the system are explored in detail. / Master of Science

In this work, we make contributions in three main areas of Cognitive Engine (CE) design for link adaptation. The three areas are CE design, CE training, and the impact of imperfect observations in the operation of the CE. First, we present a CE design for link adaptation and apply it to a system which can adapt its use of multiple antennas in addition to modulation and coding. Our design moves forward the state of the art in several ways while having a simple structure. Specifically, the CE only needs to observe the number of successes and failures associated with each set of channel conditions and communication method. From these two numbers, the CE can derive all of its functionality: estimate confidence intervals, balance exploration vs. exploitation, and utilize prior knowledge such as communication fundamentals. Finally, the CE learns the radio abilities independently of the operation objectives. Thus, if an objective changes, information regarding the radio's abilities is not lost. Second, we provide an overview of CE training, and we analytically estimate the number of trials needed to conclusively find the best performing method in a list of methods sorted by their potential performance. Furthermore, we propose the Robust Training Algorithm (RoTA) for applications where stable performance is of topmost importance. Finally, we test four key training techniques and identify and explain the three main factors that affect performance during training. Third, we assess the impact of the estimation noise on the performance of a CE. Furthermore, we derive the effect of estimation delay, in terms of the correlation between the observed SNR and the true SNR. We evaluate the effect of estimation noise and delay to the operation of the CE individually and jointly. It is found that impairments on learning make the CE more conservative in its choices leading to submaximal performance. It is found that the CE should learn using the impaired observations, if the observations are highly correlated with the actual conditions. Otherwise, it is better for the CE to learn with knowledge of the ideal conditions, if that knowledge is available. / Ph. D.

Improving the efficiency of spectrum access and utilization under the umbrella of cognitive radio (CR) is one of the most crucial research areas for nearly two decades. The results have been algorithms called cognitive radio engines which use machine learning (ML) to learn and adapt the communication's link based on the operating scenarios. While a number of algorithms for cognitive engine design have been identified, it is widely understood that significant room remains to grow the capabilities of the cognitive engines, and substantially better spectrum utilization and higher throughput can be achieved if cognitive engines are improved. This requires working through some difficult challenges and takes an innovative look at the problem. A tenet of the existing cognitive engine designs is that they are usually designed around one primary ML algorithm or framework. In this dissertation, we discover that it is entirely possible for an algorithm to perform better in one operating scenario (combination of channel conditions, available energy, and operational objectives such as max throughput, and max energy efficiency) while performing less effectively in other operating scenarios. This arises due to the unique behavior of an individual ML algorithm regardless of its operating conditions. Therefore, there is no individual algorithm or parameter sets that have superiority in performance over all other algorithms or parameter sets in all operating scenarios. Using the same algorithm at all times may present a performance that is acceptable, yet may not be the best possible performance under all operating scenarios we are faced with over time. Ideally, the system should be able to adapt its behavior by switching between various ML algorithms or adjusting the operating ML algorithm for the prevailing operating conditions and goal. In this dissertation, we introduce a novel architecture for cognitive radio engines, with the goal of better cognitive engines for improved link adaptation in order to enhance spectrum utilization. This architecture is capable of meta-reasoning and metacognition and the algorithms developed based on this architecture are called metacognitive engines (meta-CE). Meta-reasoning and metacognitive abilities provide for self-assessment, self-awareness, and inherent use and adaptation of multiple methods for link adaptation and utilization. In this work, we provide four different implementation instances of the proposed meta-CE architecture. First, a meta-CE which is equipped with a classification algorithm to find the most appropriate individual cognitive engine algorithm for each operating scenario. The meta-CE switches between the individual cognitive engine algorithms to decrease the training period of the learning algorithms and not only find the most optimal communication configuration in the fastest possible time but also provide the acceptable performance during its training period. Second, we provide different knowledge indicators for estimating the experience level of cognitive engine algorithms. We introduce a meta-CE equipped with these knowledge indicators extracted from metacognitive knowledge component. This meta-CE adjusts the exploration factors of learning algorithms to gain higher performance and decrease training time. The third implementation of meta-CE is based on the robust training algorithm (RoTA) which switches and adjusts the individual cognitive engine algorithms to guarantee a minimum performance level during the training phase. This meta-CE is also equipped with forgetting factor to deal with non-stationary channel scenarios. The last implementation of meta-CE enables the individual cognitive engine algorithms to handle delayed feedback scenarios. We analyze the impact of delayed feedback on cognitive radio engines' performances in two cases of constant and varying delay. Then we propose two meta-CEs to address the delayed feedback problem in cognitive engine algorithms. Our experimental results show that the meta-CE approach, when utilized for a CRS engine performed about 20 percent better (total throughput) than the second best performing algorithm, because of its ability to learn about its own learning and adaptation. In effect, the meta-CE is able to deliver about 70% more data than the CE with the fixed exploration rate in the 1000 decision steps. Moreover, the knowledge indicator (KI) autocorrelation plots show that the proposed KIs can predict the performance of the CEs as early as 100 time steps in advance. In non-stationary environments, the proposed RoTA based meta-CE guarantees the minimum required performance of a CRS while it’s searching for the optimal communication configurations. The RoTA based meta-CE delivers at least about 45% more data than the other algorithms in non-stationary scenarios when the channel conditions are often fluctuating. Furthermore, in delayed feedback scenarios, our results show that the proposed meta-CE algorithms are able to mitigate the adverse impact of delay in low latency scenarios and relieve the effects in high latency situations. The proposed algorithms show a minimum of 15% improvement in their performance compared to the other available delayed feedback strategies in literature. We also empirically tested the algorithms introduced in this dissertation and verified the results therein by designing an over the air (OTA) radio setup. For our experiments, we used GNU Radio and LiquidDSP as free software development toolkits that provide signal processing blocks to implement software-defined radios and signal-processing systems such as modulation, pulse-shaping, frame detection, equalization, and others. We also used two USRP N200 with WBX daughterboards, one as a transmitter and the other as a receiver. In these experiments, we monitored the packet success rate (PSR), throughput, and total data transferred as our key performance indicators (KPI). Then, we tested different proposed meta-CE algorithms in this dissertation to verify the productivity of the proposed algorithms in an OTA real-time radio setup. We showed that the experiments’ outputs support our simulations results as well.

This dissertation introduces the concept of an artificial society based on the use of an action based social language combined with the behavior-based approach to the construction of multi-agent systems to address the problem of developing decentralized, self-organizing networks that dynamically fit into their environment. In the course of accomplishing this, social language is defined as an efficient method for communicating coordination information among cognitive radios inspired by natural societies. This communication method connects the radios within a network in a way that allows the network to learn in a distributed holistic manner. The behavior-based approach to developing multi-agent systems from the field of robotics provides the framework for developing these learning networks. In this approach several behaviors are used to address the multiple objectives of a cognitive radio society and then combined to achieve emergent properties and behaviors. This work presents a prototype cognitive radio society. This society is implemented, using low complexity hardware, and evaluated. The work does not focus on the development of optimized techniques, but rather the complementary design of techniques and agents to create dynamic, decentralized self-organizing networks / Ph. D.

In recent years, cognitive radios and cognitive wireless networks have been introduced as a new paradigm for enabling much higher spectrum utilization, providing more reliable and personal radio services, reducing harmful interference, and facilitating the interoperability or convergence of different wireless communication networks. Cognitive radios are goal-oriented, autonomously learn from experience and adapt to changing operating conditions. Cognitive radios have the potential to drive the next generation of radio devices and wireless communication system design and to enable a variety of niche applications in demanding environments, such as spectrum-sharing networks, public safety, natural disasters, civil emergencies, and military operations. This research first introduces an innovative approach to developing cognitive radios based on the Radio Environment Map (REM). The REM can be viewed as an integrated database that provides multi-domain environmental information and prior knowledge for cognitive radios, such as the geographical features, available services and networks, spectral regulations, locations and activities of neighboring radios, policies of the users and/or service providers, and past experience. The REM, serving as a vehicle of network support to cognitive radios, can be exploited by the cognitive engine for most cognitive functionalities, such as situation awareness, reasoning, learning, planning, and decision support. This research examines the role of the REM in cognitive radio development from a network point of view, and focuses on addressing three specific issues about the REM: how to design and populate the REM; how to exploit the REM with the cognitive engine algorithms; and how to evaluate the performance of the cognitive radios. Applications of the REM to wireless local area networks (WLAN) and wireless regional area networks (WRAN) are investigated, especially from the perspectives of interference management and radio resource management, which illustrate the significance of cognitive radios to the evolution of wireless communications and the revolution in spectral regulation. Network architecture for REM-enabled cognitive radios and framework for REM-enabled situation-aware cognitive engine learning algorithms have been proposed and formalized. As an example, the REM, including the data model and basic application programmer interfaces (API) to the cognitive engine, has been developed for cognitive WRAN systems. Furthermore, REM-enabled cognitive cooperative learning (REM-CCL) and REM-enabled case- and knowledge-based learning algorithms (REM-CKL) have been proposed and validated with link-level or network-level simulations and a WRAN base station cognitive engine testbed. Simulation results demonstrate that the WRAN CE can adapt orders of magnitude faster when using the REM-CKL than when using the genetic algorithms and achieve near-optimal global utility by leveraging the REM-CKL and a local search. Simulation results also suggest that exploiting the Global REM information can considerably improve the performance of both primary and secondary users and mitigate the hidden node (or hidden receiver) problem. REM dissemination schemes and the resulting overhead have been investigated and analyzed under various network scenarios. By extending the optimized link state routing protocol, the overhead of REM dissemination in wireless ad hoc networks via multipoint relays can be significantly reduced by orders of magnitude as compared to plain flooding. Performance metrics for various cognitive radio applications are also proposed. REM-based scenario-driven testing (REM-SDT) has been proposed and employed to evaluate the performances of the cognitive engine and cognitive wireless networks. This research shows that REM is a viable, cost-efficient approach to developing cognitive radios and cognitive wireless networks with significant potential in various applications. Future research recommendations are provided in the conclusion. / Ph. D.

This dissertation provides theory and analysis on the impact resource management has on software-defined radio platforms by investigating the inherent trade-off between spectrum and processing effciencies with their relation to both the power consumed by the host processor and the complexity of the algorithm which it can support. The analysis demonstrates that considerable resource savings can be gained without compromising the resulting quality of service to the user, concentrating specifically on physical-layer signal processing elements commonly found in software definitions of single- and multi-carrier communications signals. Novel synchronization techniques and estimators for unknown physical layer reference parameters are introduced which complement the energy-quality scalability of software-defined receivers. A new framing structure is proposed for single-carrier systems which enables fast synchronization of short packet bursts, applicable for use in dynamic spectrum access. The frame is embedded with information describing its own structure, permitting the receiver to automatically modify its software configuration, promoting full waveformfl‚exibility for adapting to quickly changing wireless channels. The synchronizer's acquisition time is reduced by exploiting cyclostationary properties in the preamble of transmitted framing structure, and the results are validated over the air in a wireless multi-path laboratory environment. Multi-carrier analysis is concentrated on synchronizing orthogonal frequency-division multiplexing (OFDM) using offset quadrature amplitude modulation (OFDM/OQAM) which is shown to have significant spectral compactness advantages over traditional OFDM. Demodulation of OFDM/OQAM is accomplished using computationally effcient polyphase analysis filterbanks, enabled by a novel approximate square-root Nyquist filter design based on the near-optimum Kaiser-Bessel window. Furthermore, recovery of sample timing and carrier frequency offsets are shown to be possible entirely in the frequency domain, enabling demodulation in the presence of strong interference signals while promoting heterogeneous signal coexistence in dynamic spectrum environments. Resource management is accomplished through the introduction of a self-monitoring framework which permits system-level feedback to the radio at run time. The architecture permits the radio to monitor its own processor usage, demonstrating considerable savings in computation bandwidths on the tested platform. Resource management is assisted by supervised intelligent heuristic-based learning algorithms which use software-level feedback of the radio's active resource consumption to optimize energy and processing effciencies in dynamic spectrum environments. In particular, a case database-enabled cognitive engine is proposed which abstracts from the radio application by using specific knowledge of previous experience rather than relying on general knowledge within a specific problem domain. / Ph. D.

Cognitive radio (CR) technology introduces a revolutionary wireless communication mechanism in terminals and network segments, so that they are able to learn their environment and adapt intelligently to the most appropriate way of providing the service for the user's exact need. By supporting multi-band, mode-mode cognitive applications, the cognitive radio addresses an interactive way of managing the spectrum that harmonizes technology, market and regulation. This dissertation gives a complete story of building a cognitive radio. It goes through concept clarification, architecture definition, functional block building, system integration, and finally to the implementation of a fully-functional cognitive radio node prototype that can be directly packaged for application use. This dissertation starts with a comprehensive review of CR research from its origin to today. Several fundamental research issues are then addressed to let the reader know what makes CR a challenging and interesting research area. Then the CR system solution is introduced with the details of its hierarchical functional architecture called the Egg Model, modular software system called the cognitive engine, and the kernel machine learning mechanism called the cognition cycle. Next, this dissertation discusses the design of specific functional building blocks which incorporate environment awareness, solution making, and adaptation. These building blocks are designed to focus on the radio domain that mainly concerns the radio environment and the radio platform. Awareness of the radio environment is achieved by extracting the key environmental features and applying statistical pattern recognition methods including artificial neural networks and k-nearest neighbor clustering. Solutions for the radio behavior are made according to the recognized environment and the previous knowledge through case based reasoning, and further adapted or optimized through genetic algorithm solution search. New experiences are gained through the practice of the new solution, and thus the CR's knowledge evolves for future use; therefore, the CR's performance continues improving with this reinforcement learning approach. To deploy the solved solution in terms of the radio's parameters, a platform independent radio interface is designed. With this general radio interface, the algorithms in the cognitive engine software system can be applied to various radio hardware platforms. To support and verify designed cognitive algorithms and cognitive functionalities, a complete reconfigurable SDR platform, called the CWT2 waveform framework, is designed in this dissertation. In this waveform framework, a hierarchical configuration and control system is constructed to support flexible, real-time waveform reconfigurability. Integrating all the building blocks described above allows a complete CR node system. Based on this general CR node structure, a fully-functional Public Safety Cognitive Radio (PSCR) node is prototyped to provide the universal interoperability for public safety communications. Although the complete PSCR node software system has been packaged to an official release including installation guide and user/developer manuals, the process of building a cognitive radio from concept to a functional prototype is not the end of the CR research; on-going and future research issues are addressed in the last chapter of the dissertation. / Ph. D.

As wireless technologies evolve, novel innovations and concepts are required to dynamically and automatically alter various radio parameters in accordance with the radio environment. These innovations open the door for cognitive radio (CR), a new concept in telecommunications. CR makes its decisions using an inference engine, which can learn and adapt to changes in radio conditions. Fuzzy logic (FL) is the proposed decision-making algorithm for controlling the CR's inference engine. Fuzzy logic is well-suited for vague environments in which incomplete and heterogeneous information is present. In our proposed approach, FL is used to alter various radio parameters according to experience gained from different environmental conditions. FL requires a set of decision-making rules, which can vary according to radio conditions, but anomalies rise among these rules, causing degradation in the CR's performance. In such cases, the CR requires a method for eliminating such anomalies. In our model, we used a method based on the Dempster-Shafer (DS) theory of belief to accomplish this task. Through extensive simulation results and vast case studies, the use of the DS theory indeed improved the CR's decision-making capability. Using FL and the DS theory of belief is considered a vital module in the automation of various radio parameters for coping with the dynamic wireless environment. To demonstrate the FL inference engine, we propose a CR version of WiMAX, which we call CogMAX, to control different radio resources. Some of the physical parameters that can be altered for better results and performance are the physical layer parameters such as channel estimation technique, the number of subcarriers used for channel estimation, the modulation technique, and the code rate. / Ph. D.

Third Generation (3G) wireless networks have been well studied and optimized with traditional radio resource management techniques, but still there is room for improvement. Cognitive radio technology can bring significantcant network improvements by providing awareness to the surrounding radio environment, exploiting previous network knowledge and optimizing the use of resources using machine learning and artificial intelligence techniques. Cognitive radio can also co-exist with legacy equipment thus acting as a bridge among heterogeneous communication systems. In this work, an approach for applying cognition in wireless networks is presented. Also, two machine learning techniques are used to create a hybrid cognitive engine. Furthermore, the concept of cognitive radio resource management along with some of the network applications are discussed. To evaluate the proposed approach cognition is applied to three typical wireless network problems: improving coverage, handover management and determining recurring policy events. A cognitive engine, that uses case-based reasoning and a decision tree algorithm is developed. The engine learns the coverage of a cell solely from observations, predicts when a handover is necessary and determines policy patterns, solely from environment observations. / Ph. D.

This thesis addresses the problem of building a Cognitive Radio that has the ability to interact with human users in a better way by making use of Quality of Experience (QoE) as its basis and marshalling its resources optimally around the user. Salient activities of this thesis include: Analysis of CR leads to the definition of its basic functional blocks such as cognition, learning and adaptation of radio behaviour in a multi-disciplinary manner. CR tracts signal processing for radio and sensors, cognitive and behavioural psychology for user intelligence, machine learning and AI for decision systems and optimization etc. Therefore it provides a rich, fertile area to make lateral connections across diverse helds. This thesis proposes a broad definition for CR (called as Unifed Cognitive Radio) inspired by key foundation works described in literature. Besides, it also describes its functionality and its ecosystem. Taking cue from the definition of UCR, this thesis proposes architectural frame-works for various sub-systems. Also their design and implementation is achieved with the aid of a comprehensive tested setup and is tested using realistic scenarios. Builds a set of intelligent decision systems that help to achieve the set goal. This involves various design decisions with a set of diverse algorithms from the world of signal processing, machine learning and articial intelligence. Transitioning disparate small functional entities (mostly built around experiments) into an integrated system that works in real-world environment is the key aspect of this thesis. It is definitely a challenging task. Therefore, starting from deterring the architectural reference frameworks for realizing various sub-systems of UCR to an evaluation based on integrated scenario, this being an important final step constitutes a sign cant amount of work. Analysis and implementation of the integrated system to meet the desired end functionality - QoE centricity of the CR system to satisfy the needs of the end user better is the contribution of this thesis