Traditionally, networking technology has been limited because of the networks inability to adapt resulting in sub-optimal performance. Limited in state, scope and response mechanisms, network elements consisting of nodes, protocol layers and policies have been unable to make intelligent decisions. Modern networks often operate in environments where network resources (e.g. node energy, link quality, bandwidth, etc.), application data (e.g. location of user) and user behaviors (e.g. user mobility and user request pattern) experience changes over time. These changes degrade the network performance and cause service interruption. In recent years, the words "cognitive" and "smart" have become the buzzwords and have been applied to many different networking and communication systems. Cognitive networks are being touted as the next generation network services which will perceive the current network conditions and dynamically adjust their parameters to achieve better productivity. Cognitive radios will provide the end-user intelligence needed for cognitive networks and provide dynamic spectrum access for better spectrum efficiency. We are interested in assessing the practical impact of Cognitive Networks on the Wireless Communication industry. Our goal is to propose a formal business model that will help assess the implications of this new technology in the real world and the practical feasibility of its implementation. We use the layered business model proposed by Ballon [8] which follows a multi-parameter approach by defining four levels on which business models operate and by identifying three critical design parameters on each layer. The Value Network layer identifies the important entities which come into the picture in the light of the new technology. The Functional layer addresses the issue of different architectural implementations of the Cognitive Networks. At the Financial layer, we propose a NPV model which highlights the cost/revenue implications of the technology in the real world and contrasts the different Dynamic Spectrum Access (DSA) schemes from a financial perspective. Finally, the Value Proposition layer seeks to explain the end-user flexibility and efficient spectrum management provided by the use of Cognitive radios and Cognitive networks. / Master of Science

Esta Tese estuda a implementação de um novo mecanismo de análise e atuação em Redes de Sensores Sem Fio (RSSF) com múltiplos saltos baseado em características de cognição aplicadas aos nós que compõem a rede. Para tanto, é proposto um algoritmo de detecção de variabilidade dos nós sensores, envolvendo movimentação do nó, alcance do sinal da antena do sensor, quantidade de nós que fazem parte da rede e o número de conexões possíveis com nós vizinhos. Além do algoritmo de detecção de variabilidade, propõe-se um sistema multilayer denominado Adaptive Cognitive System (ACS) com base na arquitetura de Cognitive Networks (CN), que abrange: coleta, tratamento e tomada de decisão. O tratamento se refere à parte cognitiva do sistema, contemplando a criação do Cognitive Processor Module (CPMod), que por sua vez, abrange a semântica da rede, aplicação de Lógica Fuzzy e interação com um simulador de Wireless Sensor Networks (WSN) e a tomada de decisão é realizada pelo CPMod com base no resultado de análises executadas em rounds e histórico da rede com o uso de funções de pertinência de fuzzificação e defuzzificação, regras Fuzzy e inferência sobre informações coletadas da rede. Observou-se com os testes realizados na rede, utilizando-se o algoritmo de detecção, que a variabilidade dos nós sensores afeta diretamente o desempenho da rede, devido à necessidade de reestabelecimento de links e rotas entre os nós. Através de testes realizados via software na WSN, identificou-se que com o uso do ACS houve melhora significativa no desempenho em relação ao atraso fim-a-fim, latência, quantidade de pacotes descartados e de energia consumida pelos nós na rede. O ACS demonstrou potencial para a solução de problemas relacionados com as métricas destacadas, realizando ajustes em múltiplas camadas de rede do padrão IEEE 802.15.4 para até 200 nós na rede. / This Dissertation examines the implementation of a mechanism to analyze and act on multi-hop Wireless Sensor Networks (WSN) with the use of cognitive features applied to the network nodes. For this purpose, a variation detection algorithm was proposed for monitoring sensor nodes, involving the node\'s mobility features, signal range of the sensor antenna, the number of nodes in the network and the number of possible connections to neighboring nodes. In addition to the detection algorithm, a multi-layer system is proposed, named Adaptive Cognitive System (ACS). It is based on Cognitive Networks (CN) architecture, including data gathering, information treatment and decision making. The main part of the system is the Cognitive Processor Module (CPMod), which extracts the information about the WSN. In turn the Fuzzy Logic block works in tandem with the semantic engine to feed the codes to CPMod in the decision making process. The codes are the result of analysis performed on rounds using fuzzification and defuzzification membership functions, fuzzy rules and inference over information collected from the network. It was observed in tests performed in the WSN, using the detection algorithm, that the variability in sensor nodes directly affects the network performance due to the effort spent in rerounting links and paths. Through WSN testing performed via software, it was found that using the ACS implies in significant improvement in performance over the end-to-end delay, network latency, dropped packets and amount of energy consumed by nodes on the network. The ACS potential is proven for solving problems related to the previously mentioned metrics, performing adjustments on multiple network layers standardized by IEEE 802.15.4 up to 200 nodes in the network.

Esta Tese estuda a implementação de um novo mecanismo de análise e atuação em Redes de Sensores Sem Fio (RSSF) com múltiplos saltos baseado em características de cognição aplicadas aos nós que compõem a rede. Para tanto, é proposto um algoritmo de detecção de variabilidade dos nós sensores, envolvendo movimentação do nó, alcance do sinal da antena do sensor, quantidade de nós que fazem parte da rede e o número de conexões possíveis com nós vizinhos. Além do algoritmo de detecção de variabilidade, propõe-se um sistema multilayer denominado Adaptive Cognitive System (ACS) com base na arquitetura de Cognitive Networks (CN), que abrange: coleta, tratamento e tomada de decisão. O tratamento se refere à parte cognitiva do sistema, contemplando a criação do Cognitive Processor Module (CPMod), que por sua vez, abrange a semântica da rede, aplicação de Lógica Fuzzy e interação com um simulador de Wireless Sensor Networks (WSN) e a tomada de decisão é realizada pelo CPMod com base no resultado de análises executadas em rounds e histórico da rede com o uso de funções de pertinência de fuzzificação e defuzzificação, regras Fuzzy e inferência sobre informações coletadas da rede. Observou-se com os testes realizados na rede, utilizando-se o algoritmo de detecção, que a variabilidade dos nós sensores afeta diretamente o desempenho da rede, devido à necessidade de reestabelecimento de links e rotas entre os nós. Através de testes realizados via software na WSN, identificou-se que com o uso do ACS houve melhora significativa no desempenho em relação ao atraso fim-a-fim, latência, quantidade de pacotes descartados e de energia consumida pelos nós na rede. O ACS demonstrou potencial para a solução de problemas relacionados com as métricas destacadas, realizando ajustes em múltiplas camadas de rede do padrão IEEE 802.15.4 para até 200 nós na rede. / This Dissertation examines the implementation of a mechanism to analyze and act on multi-hop Wireless Sensor Networks (WSN) with the use of cognitive features applied to the network nodes. For this purpose, a variation detection algorithm was proposed for monitoring sensor nodes, involving the node\'s mobility features, signal range of the sensor antenna, the number of nodes in the network and the number of possible connections to neighboring nodes. In addition to the detection algorithm, a multi-layer system is proposed, named Adaptive Cognitive System (ACS). It is based on Cognitive Networks (CN) architecture, including data gathering, information treatment and decision making. The main part of the system is the Cognitive Processor Module (CPMod), which extracts the information about the WSN. In turn the Fuzzy Logic block works in tandem with the semantic engine to feed the codes to CPMod in the decision making process. The codes are the result of analysis performed on rounds using fuzzification and defuzzification membership functions, fuzzy rules and inference over information collected from the network. It was observed in tests performed in the WSN, using the detection algorithm, that the variability in sensor nodes directly affects the network performance due to the effort spent in rerounting links and paths. Through WSN testing performed via software, it was found that using the ACS implies in significant improvement in performance over the end-to-end delay, network latency, dropped packets and amount of energy consumed by nodes on the network. The ACS potential is proven for solving problems related to the previously mentioned metrics, performing adjustments on multiple network layers standardized by IEEE 802.15.4 up to 200 nodes in the network.

Individuals with Parkinson’s disease (PD) often show cognitive impairments in addition to motor symptoms, with the majority of PD patients converting to dementia as the disease progresses. The changes in the microstructural integrity of key nodes in resting state networks (RSNs) could be a good indicator of the cognitive effects of PD on brain regions as it progresses to dementia. To assess the association between cognitive effects and microstructural change, the microstructural integrity of the regions of interest (ROIs) in 4 resting state networks (RSN), specifically the default mode network (DMN), based on DTI were obtained in three separate groups of patients with PD. One group of patients (PD-N) were cognitively normal, while the second group of patients (PD-MCI) reflect the transitional phase of mild cognitive impairment prior to dementia, and the third group of patients (PD-D) possessed a clear diagnosis of dementia. A comparison group of healthy controls (HC) were included, matched across the three patient groups. The PD-D group showed worse microstructural integrity for the majority of the ROIs across the 4 networks. The loss of structural integrity in the PD-MCI group was more selective, with some ROIs showing similar changes to PD-D, and others showing similar changes to the PD-N group. The PD-N group fail to show any changes in the structural integrity of any ROIs, relative to HC. For future study, a combined structural / functional study should be performed to examine if there are similar changes across both measures.

Abstract Traditional communication networks consist of large sets of vendor-specific manually configurable devices. These devices are hardwired with specific control logic or algorithms used for different network functions. The resulting networks comprise distributed control plane architectures that are complex in nature, difficult to integrate and operate, and are least efficient in terms of resource usage. However, the rapid increase in data traffic requires the integrated use of diverse access technologies and autonomic network operations with increased resource efficiency. Therefore, the concepts of Software Defined Networking (SDN) are proposed that decouple the network control plane from the data-forwarding plane and logically centralize the control plane. The SDN control plane can integrate a diverse set of devices, and tune them at run-time through vendor-agnostic programmable Application Programming Interfaces (APIs). This thesis proposes software defined cognitive networking to enable intelligent use of network resources. Different radio access technologies, including cognitive radios, are integrated through a common control platform to increase the overall network performance. The architectural framework of software defined cognitive networking is presented alongside the experimental performance evaluation. Since SDN enables applications to change the network behavior and centralizes the network control plane to oversee the whole network, it is highly important to investigate SDN in terms of security. Therefore, this thesis finds the potential security vulnerabilities in SDN, studies the proposed security platforms and architectures for those vulnerabilities, and presents future directions for unresolved security vulnerabilities. Furthermore, this thesis also investigates the potential security challenges and their solutions for the enabling technologies of 5G, such as SDN, cloud technologies, and virtual network functions, and provides key insights into increasing the security of 5G networks. / Tiivistelmä Perinteiset tietoliikenneverkot pohjautuvat usein laajoille manuaalisesti konfiguroitaville valmistajakohtaisille ratkaisuille. Niissä käytetään laitekohtaista kontrollilogiikkaa tai verkon eri toiminnallisuuksien algoritmeja. Tämän johdosta verkon hajautettu kontrollitaso muodostuu monimutkaiseksi, jota on vaikea integroida ja operoida, eikä se ole kovin joustava resurssien käytön suhteen. Tietoliikenteen määrän kasvaessa tulee entistä tärkeämmäksi integroida useita verkkoteknologioita ja autonomisia verkon toiminnallisuuksia tehokkaan resurssinhallinnan saavuttamiseksi. Ohjelmisto-ohjatut verkkoratkaisut (SDN, Software Defined Networking) tarjoavat keinon hallita erikseen verkon kontrolliliikennettä eroteltuna dataliikenteestä keskitetysti. Tämä kontrollitaso voi integroida erilaisia verkkolaitteita ja ohjata niitä ajonaikaisesti valmistajariippumattoman sovellusohjelmointirajapinnan kautta. Tässä työssä on tutkittu älykästä ohjelmisto-ohjattavaa verkkoratkaisua, jonka avulla eri radioverkkoteknologiat (mukaan lukien konginitiiviradio) voidaan integroida yhteisen kontrollialustan kautta lisäämään verkon kokonaissuorituskykyä. Työssä esitetään kognitiivinen ohjelmisto-ohjattu verkon arkkitehtuuriratkaisu sekä sen suorituskyvyn arviointi mittauksiin pohjautuen. Koska ohjelmisto-ohjattu verkko pohjautuu koko verkon keskitettyyn kontrollilogiikkaan, on tietoturvan merkitys korostunut entisestään. Tässä työssä on sen vuoksi tutkittu juuri tällaisen verkkoratkaisun mahdollisia tietoturvauhkia sekä niiden torjumiseen soveltuvia ratkaisuvaihtoehtoja sekä esitetään tulevaisuuden kehityssuuntia vielä ratkaisemattomille uhkille. Lisäksi työssä on tutkittu laajemmin tulevien 5G verkkojen tietoturvauhkia ja niiden ratkaisuja, liittyen ohjelmisto-ohjattuihin verkkoratkaisuin, pilviteknologioihin ja virtualisoiduille verkkotoiminnallisuuksille. Työ tarjoaa myös näkemyksen siitä, miten verkon tietoturvaa voidaan kokonaisuudessaan lisätä 5G verkoissa.

In this dissertation, the performance of cooperative cognitive systems with spectrum sharing is investigated. A low-complexity and high performance node selection strategy is proposed for two different of cooperative cognitive systems models. In the first model, the secondary network is composed by one source node that communicates with one among L destinations through a direct link and also assisted by one among N AF or DF relays nodes. The selected secondary destination employs a selection combining technique for choosing the best link (direct or dual-hop link) from the secondary source. Considering an underlay spectrum sharing approach, the secondary communication is performed taking into account an interference constraint, where the overall transmit power is limited by the interference at the primary receiver as well as by the maximum transmission power available at the respective nodes. An asymptotic analysis is carried out, revealing that the diversity order of the considered system is not affected by the interference, and equals to L + N. In the second model, by its turn, the secondary network is composed by one source, N AF or DF relays, and one destination. However, it is assumed the presence of M primary receivers. A relay selection strategy is proposed with the aim of maximing the end-to-end signal-to-noise ratio and, at the same time, to satisfy the interference constraints imposed by these primary receivers. After the relay selection procedure is performed, the secondary destination chooses the best path (direct link or relaying link) by employing a selection combining scheme. An asymptotic analysis is carried out, revealing that the system diversity order equals to N + 1, and showing that it is not affected neither by the number of primary receivers nor by the interference threshold. A close-form expression and an approximation for the outage probability is derived for the DF and AF protocols, respectively. / Nesta dissertaÃÃo, o desempenho de sistemas cooperativos cognitivos com compartilhamento espectral à investigado. Uma estratÃgia de seleÃÃo de nÃs de baixa complexidade e alto desempenho à proposta para dois modelos distintos de redes cooperativas cognitivas. No primeiro modelo, a rede secundÃria à composta por um nà fonte que comunica-se com um dentre L nÃs destinos atravÃs de um link direto e atravÃs de um dentre N nÃs relays decodifica-e-encaminha (DF) ou amplifica-e-encaminha (AF). O nà destino secundÃrio selecionado emprega uma tÃcnica de combinaÃÃo por seleÃÃo para selecionar o melhor link (direto ou auxiliar) a partir da fonte secundÃria. Considerando um ambiente com compartilhamento espectral, tem-se que a comunicaÃÃo secundÃria à realizada levando em consideraÃÃo uma restriÃÃo de interferÃncia, na qual a potÃncia de transmissÃo à governada pela interferÃncia no receptor primÃrio bem como pela mÃxima potÃncia de transmissÃo dos respectivos nÃs secundÃrios. Uma anÃlise assintÃtica à realizada, revelando que a ordem de diversidade do sistema nÃo à afetada pela interferÃncia, sendo igual a L + N. Jà no segundo modelo, a rede secundÃria à composta por uma fonte, N relays DF ou AF e um nà destino, no entanto assume-se a presenÃa de M receptores primÃrios. A seleÃÃo do relay deve satisfazer as restriÃÃes de interferÃncia impostas por estes Ãltimos. ApÃs a seleÃÃo de relay ser realizada, o nà destino seleciona o melhor caminho (link direto ou link via relay) proveniente da fonte utilizando um combinador por seleÃÃo. Uma anÃlise assintÃtica à realizada, revelando que a ordem de diversidade do esquema proposto iguala a N + 1, o que mostra que a mesma nÃo à afetada nem pelo nÃmero de receptores primÃrios nem pelo limiar de interferÃncia. Uma expressÃo em forma fechada para a probabilidade de outage à obtida para ambos protocolos cooperativos. SimulaÃÃes Monte Carlo sÃo apresentadas com o intuito de validar as anÃlises propostas.

This thesis provides three main contributions with respect to the Dynamic Channel and Power Assignment (DCPA) problem. DCPA refers to the allocation of transmit power and frequency channels to links in a cognitive dynamic spectrum network so as to maximize the total number of feasible links while minimizing the aggregate transmit power. In order to provide a method to compare related, yet disparate, work, the first contribution of this thesis is a unifying optimization formulation to describe the DCPA problem. This optimization problem is based on maximizing the number of feasible links and minimizing transmit power of a set of communications links in a given communications network. Using this optimization formulation, this thesis develops its second contribution: a evaluation method for comparing DCPA algorithms. The evaluation method is applied to five DPCA algorithms representative of the DCPA literature . These five algorithms are selected to illustrate the tradeoffs between control modes (centralized versus distributed) and channel/power assignment techniques. Initial algorithm comparisons are done by analyzing channel and power assignment techniques and algorithmic complexity of five different DCPA algorithms. Through simulations, algorithm performance is evaluated by the metrics of feasibility ratio and average power per link. Results show that the centralized algorithm Minimum Power Increase Assignment (MPIA) has the overall best feasibility ratio and the lowest average power per link of the five algorithms we investigated. Through assignment by the least change in transmit power, MPIA minimizes interference and increases the number of feasible links. However, implementation of this algorithm requires calculating the inverse of near singular matrices, which could lead to inaccurate results. The third contribution of this thesis is a proposed distributed channel assignment algorithm, Least Interfering Channel and Iterative Power Assignment (LICIPA). This distributed algorithm has the best feasibility ratio and lowest average power per link of the distributed algorithms. In some cases, LICIPA achieves 90% of the feasibility ratio of MPIA, while having lower complexity and overall lower average run time. / Master of Science

This dissertation addresses the problem of channel assignment in multi-hop, multi-transceiver wireless networks. We investigate (1) how channels can be assigned throughout the network to ensure that the network is connected and (2) how the channel assignment can be adapted to suit the current traffic demands. We analyze a traffic-aware method for channel assignment that addresses both maintaining network connectivity and adapting the topology based on dynamic traffic demands. The traffic-aware approach has one component that assigns channels independently of traffic conditions and a second component that assigns channels in response to traffic conditions. The traffic-independent (TI) component is designed to allocate as few transceivers or radios as possible in order to maintain network connectivity, while limiting the aggregate interference induced by the topology. The traffic-driven (TD) component is then designed to maximize end-to-end flow rate using the resources remaining after the TI assignment is complete. By minimizing resources in the TI component, the TD component has more resources to adapt the topology to suit the traffic demands and support higher end-to-end flow rate. We investigate the fundamental tradeoff between how many resources are allocated to maintaining network connectivity versus how many resources are allocated to maximize flow rate. We show that the traffic-aware approach achieves an appropriately balanced resource allocation, maintaining a baseline network connectivity and adapting to achieve near the maximum theoretical flow rate in the scenarios evaluated. We develop a set of greedy, heuristic algorithms that address the problem of resource- minimized TI assignment, the first component of the traffic-aware assignment. We develop centralized and distributed schemes for nodes to assign channels to their transceivers. These schemes perform well as compared to the optimal approach in the evaluation. We show that both of these schemes perform within 2% of the optimum in terms of the maximum achievable flow rate. We develop a set of techniques for adapting the network's channel assignment based on traffic demands, the second component of the traffic-aware assignment. In our approach, nodes sense traffic conditions and adapt their own channel assignment independently to support a high flow rate and adapt when network demand changes. We demonstrate how our distributed TI and TD approaches complement each other in an event-driven simulation. / Ph. D.

This dissertation considers the design, evaluation and optimization of algorithms/ techniques/ system parameters for distributed wireless networks specifically ad-hoc and cognitive wireless networks. In the first part of the dissertation, we consider ad-hoc networks using opportunistic carrier sense multiple access (CSMA) protocols. The key challenge in optimizing the performance of such systems is to find a good compromise among three interdependent quantities: the density and channel quality of the scheduled transmitters, and the resulting interference seen at receivers. We propose two new channel-aware slotted CSMA protocols and study the tradeoffs they achieve amongst these quantities. In particular, we show that when properly optimized these protocols offer substantial improvements relative to regular CSMA -- particularly when the density of nodes is moderate to high. Moreover, we show that a simple quantile based opportunistic CSMA protocol can achieve robust performance gains without requiring careful parameter optimization. In the second part of the dissertation, we study a cognitive wireless network where licensed (primary) users and unlicensed 'cognitive' (secondary) users coexist on shared spectrum. In this context, many system design parameters affect the joint performance, e.g., outage and capacity, seen by the two user types. We explore the performance dependencies between primary and secondary users from a spatial reuse perspective, in particular, in terms of the outage probability, node density and joint network capacity. From the design perspective the key system parameters determining the joint transmission capacity, and tradeoffs, are the detection radius (detection signal to interference and noise power ratio (SINR) threshold) and decoding SINR threshold. We show how the joint network capacity region can be optimized by varying these parameters. In the third part of the dissertation, we consider a cognitive network in a heterogeneous environment, including indoor and outdoor transmissions. We characterize the joint network capacity region under three different spectrum (white space) detection techniques which have different degrees of radio frequency (RF) - environment awareness. We show that cognitive devices relying only on the classical signal energy detection method perform poorly due to limitations on detecting primary transmitters in environments with indoor shadowing. This can be circumvented through direct use (e.g., database access) of location information on primary transmitters, or better yet, on that of primary receivers. We also show that if cognitive devices have positioning information, then the secondary network's capacity increases monotonically with increased indoor shadowing in the environment. This dissertation extends the recent efforts in using stochastic geometric models to capture large scale performance characteristics of wireless systems. It demonstrates the usefulness of these models towards understanding the impact of physical /medium access (MAC) layer parameters and how they might be optimized. / text

The heterogeneity and complexity of modern communication networks demands coupling network nodes with intelligence to perceive and adapt to different network conditions autonomously. Cognitive Networking is an emerging networking research area that aims to achieve this goal by applying distributed reasoning and learning across the protocol stack and throughout the network. Various cognitive node and cognitive network architectures with different levels of maturity have been proposed in the literature. All of them adopt the idea of coupling network devices with sensors to sense network conditions, artificial intelligence algorithms to solve problems, and a reconfigurable platform to apply solutions. However, little further research has investigated suitable reasoning and learning algorithms. In this dissertation, we take cognitive network research a step further by investigating the reasoning component of cognitive networks. In a deviation from previous suggestions, we suggest the use of a single flexible distributed reasoning algorithm for cognitive networks. We first propose an architecture for a cognitive node in a cognitive network that is general enough to apply to future networking challenges. We then introduce and justify our choice of the island genetic algorithm (iGA) as the distributed reasoning algorithm. Having introduced our cognitive node architecture, we then focus on the applicability of the island genetic algorithm as a single reasoning algorithm for cognitive networks. Our approach is to apply the island genetic algorithm to different single and cross layer communication and networking problems and to evaluate its performance through simulation. A proof of concept cognitive network is implemented to understand the implementation challenges and assess the island genetic algorithm performance in a real network environment. We apply the island genetic algorithm to three problems: channel allocation, joint power and channel allocation, and flow routing. The channel allocation problem is a major challenge for dynamic spectrum access which, in turn, has been the focal application for cognitive radios and cognitive networks. The other problems are examples of hard cross layer problems. We first apply the standard island genetic algorithm to a channel allocation problem formulated for the dynamic spectrum cognitive network environment. We also describe the details for implementing a cognitive network prototype using the universal software radio peripheral integrated with our extended implementation of the GNU radio software package and our island genetic algorithm implementation for the dynamic spectrum channel allocation problem. We then develop a localized variation of the island genetic algorithm, denoted LiGA, that allows the standard island genetic algorithm to scale and apply it to the joint power and channel allocation problem. In this context, we also investigate the importance of power control for cognitive networks and study the effect of non-cooperative behavior on the performance of the LiGA. The localized variation of the island genetic algorithm, LiGA, is powerful in solving node-centric problems and problems that requires only limited knowledge about network status. However, not every communication and networking problems can be solved efficiently in localized fashion. Thus, we propose a generalized version of the LiGA, namely the K-hop island genetic algorithm, as our final distributed reasoning algorithm proposal for cognitive networks. The K-hop island genetic algorithm is a promising algorithm to solve a large class of communication and networking problems with controllable cooperation and migration scope that allows for a tradeoff between performance and cost. We apply it to a flow routing problem that includes both power control and channel allocation. For all problems simulation results are provided to quantify the performance of the island genetic algorithm variation. In most cases, simulation and experimental results reveal promising performance for the island genetic algorithm. We conclude our work with a discussion of the shortcomings of island genetic algorithms without guidance from a learning mechanism and propose the incorporation of two learning processes into the cognitive node architecture to solve slow convergence and manual configuration problems. We suggest the cultural algorithm framework and reinforcement learning techniques as candidate leaning techniques for implementing the learning processes. However, further investigation and implementation is left as future work. / Ph. D.