The integration of heterogeneous access networks is becoming a possible feature of 4G wireless networks. It is challenging to deliver the multimedia services over such integrated networks because of the discrepancy in the bandwidth of different networks. This thesis presents an adaptive approach that combines source rate adaptation and buffering to achieve high quality VBR video streaming with less quality variation over an integrated two-tier network. Statistical information of the residence time in each network or localization information are utilized to anticipate the handoff occurrence. The performance of this approach is analyzed under the CBR case using a Markov reward model. Simulation under the CBR and VBR cases is conducted for different types of network models. The results are compared with a dynamic programming algorithm as well as other naive or intuitive algorithms, and proved to be promising.

The integration of heterogeneous access networks is becoming a possible feature of 4G wireless networks. It is challenging to deliver the multimedia services over such integrated networks because of the discrepancy in the bandwidth of different networks. This thesis presents an adaptive approach that combines source rate adaptation and buffering to achieve high quality VBR video streaming with less quality variation over an integrated two-tier network. Statistical information of the residence time in each network or localization information are utilized to anticipate the handoff occurrence. The performance of this approach is analyzed under the CBR case using a Markov reward model. Simulation under the CBR and VBR cases is conducted for different types of network models. The results are compared with a dynamic programming algorithm as well as other naive or intuitive algorithms, and proved to be promising.

In the emerging wireless Internet environment involving m-commerce and other mobile applications, an increasing number of users are likely to adopt mobile transactions. These transactions are likely to have very diverse requirements and some of them may require significant amount of network resources and/or bounded delays. Most quality-of-service research in wireless networks has hitherto focused on call or connection-level QoS. Many mobile transactions are expected to be distinct from the previously investigated applications in their criticality, level of resource required, and in their group characteristics. Examples of such transactions are ones involving a financial value. These unique requirements of mobile transactions necessitate introduction of new metrics for quality-of-service. To measure QoS effectiveness of mobile transactions, two new metrics, namely transaction completion probability and transaction response time are introduced in this research. Moreover, it is well known that wireless networks are constrained for bandwidth. Mobile transactions are expected to require varying degree of bandwidth which makes the resource allocation only at connection level very inefficient. This research proposes a new framework to support QoS requirements of mobile transactions by allocating bandwidth at connection and transaction levels. The proposed framework helps in achieving a balance between transaction completion probability and the response time. Simulation and analytical modeling are used to evaluate the QoS metrics under varying network and traffic scenarios and to validate the effectiveness of the new framework. The results show that the balanced transaction and connection level resource allocation can improve the probability of transactions completion and resource utilization but at the cost of slightly increased response time.

In order to improve the Quality of Service in wireless networks it is crucial to design and optimize the communication algorithms based on the underlying Physical and Link Layers. In this thesis we show that if instead of the link abstraction used in traditional wireless networking we rely on the much broader definition of a link, used in the context of cooperative communication, we can improve the performance of relay transmission systems operating over the wireless medium. From a networking perspective there are a whole host of layering and cross-layer design issues that enable one to propose optimal cooperative algorithms for wireless communication. Most of the research in this area has been concentrated on the physical layer issues. In this thesis, we consider the interaction of the physical layer cooperative link with the higher layers, in particular the Medium Access Control Layer, and show that by appropriate protocol design we can improve the performance of wireless networks by using cooperation. Enabling cooperation among nodes in an optimal manner can lead to significant increase in the throughput for multi-hop wireless networks. We study and design cooperative protocols that lead to this throughput increase and quantify the appropriate level of cooperation among the users which leads to improving QoS.

A new promising approach for wireless network optimization is from a cross-layer perspective. This thesis focuses on exploiting channel state information (CSI) from the physical layer for optimal transmission scheduling at the medium access control (MAC) layer. The first part of the thesis considers exploiting CSI via a distributed channel-aware MAC protocol. The MAC protocol is analysed using a centralized design approach and a non-cooperative game theoretic approach. Structural results are obtained and provably convergent stochastic approximation algorithms that can estimate the optimal transmission policies are proposed. Especially, in the game theoretic MAC formulation, it is proved that the best response transmission policies are threshold in the channel state and there exists a Nash equilibrium at which every user deploys a threshold transmission policy. This threshold result leads to a particularly efficient stochastic-approximation-based adaptive learning algorithm and a simple distributed implementation of the MAC protocol. Simulations show that the channel-aware MAC protocols result in system throughputs that increase with the number of users. The thesis also considers opportunistic transmission scheduling from the perspective of a single user using Markov Decision Process (MDP) approaches. Both channel state in-formation and channel memory are exploited for opportunistic transmission. First, a finite horizon MDP transmission scheduling problem is considered. The finite horizon formulation is suitable for short-term delay constraints. It is proved for the finite horizon opportunistic transmission scheduling problem that the optimal transmission policy is threshold in the buffer occupancy state and the transmission time. This two-dimensional threshold structure substantially reduces the computational complexity required to compute and implement the optimal policy. Second, the opportunistic transmission scheduling problem is formulated as an infinite horizon average cost MDP with a constraint on the average waiting cost. An advantage of the infinite horizon formulation is that the optimal policy is stationary. Using the Lagrange dynamic programming theory and the super modularity method, it is proved that the stationary optimal transmission scheduling policy is a randomized mixture of two policies that are threshold in the buffer occupancy state. A stochastic approximation algorithm and a Q-learning based algorithm that can adaptively estimate the optimal transmission scheduling policies are then proposed.

We study a wireless full-duplex network with imperfect interference cancellation and solve the routing and power allocation problem in this network. We use a model that focuses on the effects of full-duplex by including residual self-interference and one hop interference while other interfering signals are considered negligible in comparison. We first solve the optimal power allocation for a fixed route. We then propose a priority-first search algorithm to find the joint route and power allocation to maximize throughput. The algorithm proposed has a non decomposable priority metric, but is efficiently evaluated by our solution for a fixed route. We analyze the performance of our solution in a more realistic model by deriving bounds between optimal solutions in both models. Through simulations we show that, even with imperfect interference cancellation, full-duplex achieves a higher throughput than half-duplex or direct transmission for moderate transmission power.

1. CONTEXTO1.1. MANETMANET (Mobile and Ad hoc NETworks) (Redes móviles sin cables) son redes formadas por nodos móviles. Se comunican sin cables i lo hacen de manera 'ad hoc'. En este tipo de redes, los protocolos de enrutamiento tienen que ser diferentes de los utilizados en redes fijas.Hoy en día, existen protocolos de enrutamiento capaces de operar en este tipo de redes. No obstante, son completamente inseguras y confían en que los nodos no actuarán de manera malintencionada. En una red donde no se puede contar con la presencia de servidores centrales, se necesita que los nodos puedan comunicarse sin el riesgo de que otros nodos se hagan pasar por aquellos con quien quieren comunicarse. En una red donde todo el mundo es anónimo conceptos como identidad y confianza deben ser redefinidos.1.2. AODVAd Hoc On-Demand Vector Routing (AODV) es un protocolo de enrutamiento reactivo para redes MANET. Esto significa que AODV no hace nada hasta que un nodo necesita transmitir un paquete a otro nodo para el cual no tiene ruta. AODV sólo mantiene rutas entre nodos que necesitan comunicarse. Sus mensajes no contienen información de toda la ruta, solo contienen información sobre el origen i el destino. Por lo tanto los mensajes de AODV tienen tamaño constante independientemente del numero de nodos de la ruta. Utiliza números de secuencia para especificar lo reciente que es una ruta (en relación con otra), lo cual garantiza ausencia de 'loops' (bucles).En AODV, un nodo realiza un descubrimiento de ruta haciendo un 'flooding' de la red con un mensaje llamado 'Route Request' (RREQ). Una vez llega a un nodo que conoce la ruta pedida responde con un 'Route Reply' (RREP) que viaja de vuelta al originador del RREQ. Después de esto, todos los nodos de la ruta descubierta conocen las rutas hacia los dos extremos de la ruta.2. CONTRIBUTIONS2.1. SAODVSAODV (Secure Ad hoc On-Demand Distance Vector) es una extensión de AODV que protege el mecanismo de descubrimiento de ruta. Proporciona funcionalidades de seguridad como ahora integridad i autenticación.Se utilizan firmas digitales para autenticar los campos de los mensajes que no son modificados en ruta y cadenas de hash para proteger el 'hop count' (que es el único campo que se modifica en ruta).2.2. SAKMSAKM (Simple Ad hoc Key Management) proporciona un sistema de gestión de llaves que hace posible para cada nodo obtener las llaves públicas de los otros nodos de la red. Además, permite que cada nodo pueda verificar la relación entre la identidad de un nodo y la llave pública de otro.Esto se consigue a través del uso de direcciones estadísticamente únicas y criptográficamente verificables.2.2.1. Verificación pospuestaEl método 'verificación pospuesta' permite tener rutas pendientes de verificación. Estas serán verificadas cuando el procesador disponga de tiempo para ello y (en cualquier caso) antes de que esas rutas deban ser utilizadas para transmitir paquetes.2.3. Detección de atajosCuando un protocolo de enrutamiento para redes MANET realiza un descubrimiento de ruta, no descubre la ruta más corta sino la ruta a través de la cual el mensaje de petición de ruta viajó más rápidamente. Además, debido a que los nodos son móviles, la ruta que era la más corta en el momento del descubrimiento puede dejar de ser-lo en breve. Esto causa un retraso de transmisión mucho mayor de lo necesario y provoca muchas más colisiones de paquetes.Para evitar esto, los nodos podrían realizar descubrimientos de atajos periódicos para las rutas que están siendo utilizadas. Este mismo mecanismo puede ser utilizado para 'recuperar' rutas que se han roto. / 1. BACKGROUND1.1. MANETMANET (Mobile and Ad hoc NETworks) are networks formed by nodes that are mobile. They use wireless communication to speak among them and they do it in an ad hoc manner. In this kind of networks, routing protocols have to be different than from the ones used for fixed networks. In addition, nodes use the air to communicate, so a lot of nodes might hear what a node transmits and there are messages that are lost due to collisions.Nowadays, routing in such scenario has been achieved. Nevertheless, if it has to be broadly used, it is necessary to be able to do it in a secure way. In a network where the existance of central servers cannot be expected, it is needed that nodes will be able to communicate without the risk of malicious nodes impersonating the entities they want to communicate with. In a network where everybody is anonymous, identity and trust need to be redefined.1.2. AODVAd Hoc On-Demand Vector Routing (AODV) protocol is a reactive routing protocol for ad hoc and mobile networks. That means that AODV does nothing until a node needs to transmit a packet to a node for which it does not know a route. In addition, it only maintains routes between nodes which need to communicate. Its routing messages do not contain information about the whole route path, but only about the source and the destination. Therefore, routing messages have a constant size, independently of the number of hops of the route. It uses destination sequence numbers to specify how fresh a route is (in relation to another), which is used to grant loop freedom.In AODV, a node does route discovery by flooding the network with a 'Route Request' message (RREQ). Once it reaches a node that knows the requested route, it replies with a 'Route Reply' message (RREP) that travels back to the originator of the RREQ. After this, all the nodes of the discovered path have routes to both ends of the path. 2. CONTRIBUTIONS2.1. SAODVThe Secure Ad hoc On-Demand Distance Vector (SAODV) is an extension of the AODV routing protocol that can be used to protect the route discovery mechanism providing security features like integrity and authentication.Two mechanisms are used to secure the AODV messages: digital signatures to authenticate the non-mutable fields of the messages, and hash chains to secure the hop count information (the only mutable information in the messages).The information relative to the hash chains and the signatures is transmitted with the AODV message as an extension message.2.2. SAKMSimple Ad hoc Key Management (SAKM) provides a key management system that makes it possible for each ad hoc node to obtain public keys from the other nodes of the network. Further, each ad hoc node is capable of securely verifying the association between the identity of a given ad hoc node and the public key of that node.This is achieved by using statistically unique and cryptographically verifiable address.2.2.1. Delayed VerificationDelayed verification allows to have route entries and route entry deletions in the routing table that are pending of verification. They will be verified whenever the node has spared processor time or before these entries should be used to forward data packages.2.3. Short Cut DetectionWhen a routing protocol for MANET networks does a route discovery, it does not discover the shortest route but the route through which the route request flood traveled faster. In addition, since nodes are moving, a route that was the shortest one at discovery time might stop being so in quite a short period of time. This causes, not only a much bigger end-to-end delay, but also more collisions and a faster power consumption.In order to avoid all the performance loss due to these problems, nodes could periodically discover shortcuts to the active routes that can be used with any destination vector routing protocol. The same mechanism can be used also as a bidirectional route recovery mechanism.

The next-generation wireless communications have been envisioned to be supported by heterogeneous networks using various wireless access technologies. The popular cellular networks and wireless local area networks (WLANs) present perfectly complementary characteristics in terms of service capacity, mobility support, and quality-of-service (QoS) provisioning. The cellular/WLAN interworking is thus an effective way to promote the evolution of wireless networks. As an essential aspect of the interworking, resource allocation is vital for efficient utilization of the overall resources. Specially, multi-service provisioning can be enhanced with cellular/WLAN interworking by taking advantage of the complementary network strength and an overlay structure. Call assignment/reassignment strategies and admission control policies are effective resource allocation mechanisms for the cellular/WLAN integrated network. Initially, the incoming calls are distributed to the overlay cell or WLAN according to call assignment strategies, which are enhanced with admission control policies in the target network. Further, call reassignment can be enabled to dynamically transfer the traffic load between the overlay cell and WLAN via vertical handoff. By these means, the multi-service traffic load can be properly shared between the interworked systems. In this thesis, we investigate the load sharing problem for this heterogeneous wireless overlay network. Three load sharing schemes with different call assignment/reassignment strategies and admission control policies are proposed and analyzed. Effective analytical models are developed to evaluate the QoS performance and determine the call admission and assignment parameters. First, an admission control scheme with service-differentiated call assignment is studied to gain insights on the effects of load sharing on interworking effectiveness. Then, the admission scheme is extended by using randomized call assignment to enable distributed implementation. Also, we analyze the impact of user mobility and data traffic variability. Further, an enhanced call assignment strategy is developed to exploit the heavy-tailedness of data call size. Last, the study is extended to a multi-service scenario. The overall resource utilization and QoS satisfaction are improved substantially by taking into account the multi-service traffic characteristics, such as the delay-sensitivity of voice traffic, elasticity and heavy-tailedness of data traffic, and rate-adaptiveness of video streaming traffic.

This thesis explores the optimal node placement for linear Gaussian multiple relay networks of an arbitrary size and with one source-destination pair. Consider the the low attenuation regime (path loss exponent less than 3/2). Under the condition that the minimum achievable rate from source to destination is maintained, we derive upper bounds of node placement with the incoherent and coherent coding schemes, and examine the optimal power assignment related to the node placement with the coherent coding scheme. We prove that the farthest distance between two adjacent nodes is bounded even for an infinite total number of relay nodes, and closed-form formulas of the bounds are derived for both the coding schemes. Furthermore, the distance from the source to the destination is of the same order as the total number of nodes, given the path loss exponent greater than one half under the incoherent coding scheme and the path loss exponent greater than 1 with coherent relaying with interference subtraction coding scheme. Conditioned on a conjecture based on the simulation results, we also provide heuristic upper bounds, which are a little tighter than the strictly proved bounds. The bounds provided in this thesis can serve as a helpful guideline for the relay extension problem in practical network implementation.

The next-generation wireless networks are expected to integrate diverse network architectures and various wireless access technologies to provide a robust solution for ubiquitous broadband wireless access, such as wireless local area networks (WLANs), Ultra-Wideband (UWB), and millimeter-wave (mmWave) based wireless personal area networks (WPANs), etc. To enhance the spectral efficiency and link reliability, smart antenna systems have been proposed as a promising candidate for future broadband access networks. To effectively exploit the increased capabilities of the emerging wireless networks, the different network characteristics and the underlying physical layer features need to be considered in the medium access control (MAC) design, which plays a critical role in providing efficient and fair resource sharing among multiple users. In this thesis, we comprehensively investigate the MAC design in both single- and multi-hop broadband wireless networks, with and without infrastructure support. We first develop mathematical models to identify the performance bottlenecks and constraints in the design and operation of existing MAC. We then use a cross-layer approach to mitigate the identified bottleneck problems. Finally, by evaluating the performance of the proposed protocols with analytical models and extensive simulations, we determine the optimal protocol parameters to maximize the network performance. In specific, a generic analytical framework is developed for capacity study of an IEEE 802.11 WLAN in support of non-persistent asymmetric traffic flows. The analysis can be applied for effective admission control to guarantee the quality of service (QoS) performance of multimedia applications. As the access point (AP) becomes the bottleneck in an infrastructure based WLAN, we explore the multiple-input multiple-output (MIMO) capability in the future IEEE 802.11n WLANs and propose a MIMO-aware multi-user (MU) MAC. By exploiting the multi-user degree of freedom in a MIMO system to allow the AP to communicate with multiple users in the downlink simultaneously, the proposed MU MAC can minimize the AP-bottleneck effect and significantly improve the network capacity. Other enhanced MAC mechanisms, e.g., frame aggregation and bidirectional transmissions, are also studied. Furthermore, different from a narrowband system where simultaneous transmissions by nearby neighbors collide with each other, wideband system can support multiple concurrent transmissions if the multi-user interference can be properly managed. Taking advantage of the salient features of UWB and mmWave communications, we propose an exclusive region (ER) based MAC protocol to exploit the spatial multiplexing gain of centralized UWB and mmWave based wireless networks. Moreover, instead of studying the asymptotic capacity bounds of arbitrary networks which may be too loose to be useful in realistic networks, we derive the expected capacity or transport capacity of UWB and mmWave based networks with random topology. The analysis reveals the main factors affecting the network (transport) capacity, and how to determine the best protocol parameters to maximize the network capacity. In addition, due to limited transmission range, multi-hop relay is necessary to extend the communication coverage of UWB networks. A simple, scalable, and distributed UWB MAC protocol is crucial for efficiently utilizing the large bandwidth of UWB channels and enabling numerous new applications cost-effectively. To address this issue, we further design a distributed asynchronous ER based MAC for multi-hop UWB networks and derive the optimal ER size towards the maximum network throughput. The proposed MAC can significantly improve both network throughput and fairness performance, while the throughput and fairness are usually treated as a tradeoff in other MAC protocols.