In a wireless network, owing to the time-varying nature of wireless channels, different mobile users typically experience peaks and troughs in channel quality at different times. This diversity in channel quality is known as multiuser diversity. With the aid of rate adaptation, multiuser diversity can be exploited by allowing the mobile user with the best channel to use the channel resource. However, in order to achieve this in most practical systems, the mobile users in the network must feed back their channel state information (CSI) to the transmitting user. In large networks, this feedback overhead can outweigh the multiuser diversity gain. In this thesis dissertation, a centralised wireless medium access control (MAC) scheme, namely Multiuser Diversity with Capture (MDC), is discussed as a solution to obviate the overhead problem. MDC explicitly employs the capture effect in radio receivers to reduce network overhead by allowing multiple mobile stations (MSs) with channels better than a nominal response threshold to simultaneously compete for the wireless channel. Owing to the capture effect, the base station (BS) can determine which MS has the best channel. In comparison with the Medium Access Diversity (MAD) scheme in the literature, the proposed MDC possesses the strong merit that the feedback overhead is independent of the number of MSs in the network. Several aspects of the MDC scheme are investigated in detail. An application of the MDC scheme based on the physical layer and parts of the MAC layer of the IEEE 802.11a standard is considered. A general analytical framework for the goodput performance of MDC is derived. Using this framework, the exact closed form solution for the expected goodput of MDC with rate adaptation over Rayleigh fading channels is calculated. The fairness performance of MDC in networks where some MSs experience better average channel conditions than others is also addressed. MSs with low average channel states tend to use the channel less often in MDC than MSs with high average channel states. This issue is tackled with Fairer Multiuser Diversity with Capture (FMDC), a variant of the MDC scheme designed to share the channel resource more equitably across all of the MSs in the network. In FMDC, instead of using the network-wide response threshold to decide whether to compete for the channel, each MS only competes for the channel when their channel state is greater than a threshold factor multiplied by their average channel state. Finally, the problem of adaptive optimisation of the response threshold for MDC and the threshold factor for FMDC is also considered. In the proposed solution, the response threshold and the threshold factor are adapted heuristically according to the estimated goodput performance of the system. The adaptive heuristic has importance in practical systems because the BS usually does not know the characteristics of the time varying channels of the MSs in the network.

In this dissertation, we explore the issues related to opportunistic and cooperative communications in a multiuser environment. In the first part of the dissertation, we consider opportunistic scheduling for delay limited systems. Multiuser communication over fading channels is a challenging problem due to fast varying channel conditions. On the other hand, it provides opportunities to exploit the varying nature of the channel and maximize the throughput by scheduling the user (or users) with good channel. This gain is termed as multiuser diversity. The larger the number of users, the greater is the multiuser diversity gain. However, there is an inherent scheduling delay in exploiting multiuser diversity. The objective of this work is to design the scheduling schemes which use multiuser diversity to minimize the system transmit energy. We analyze the schemes in large system limit and characterize the energy--delay tradeoff. We show that delay tolerance in data transmission helps us to exploit multiuser diversity and results in an energy efficient use of the system resources. We assume a general multiuser environment but the proposed scheduling schemes are specifically suitable for the wireless sensor network applications where saving of transmit energyat the cost of delay in transmission is extremely useful to increase the life of battery for the sensor node. In the first part of the thesis, we propose scheduling schemes withthe objective of minimizing transmit energy for a given fixed tolerable transmission delay. The fixed delay is termed as hard deadline. A group of users with channels better than a transmission threshold are scheduled for transmission simultaneously using superposition coding. The transmission thresholds depend onthe fading statistics of the underlying channel and hard deadline of the data to be scheduled. As deadline is approached, the thresholds decrease monotonically to reflect the scheduling priority for theuser. We analyze the proposed schedulers in the large system limit. We compute the optimized transmission thresholds for the proposed scheduling schemes. We analyze the proposed schemes for practically relevant scenarios when the randomly arriving packets have individual, non--identical deadlines. We analyze the case when loss tolerance of the application is exploited to further decrease the system energy. The transmitted energy is not a convex function oftransmission thresholds. Therefore, we propose heuristic optimization procedures to compute the transmission thresholds and evaluate the performance of the schemes. Finally, we study the effect of outer cell interference on the proposed scheduling schemes. The second part of the thesis investigates the problem of cooperative communication between the nodes which relay the data of other sources multiplex with their own data towards a common destination, i.e. a relay node performs as a relay and data source at the same time. This problem setting is very useful in case of some wireless sensor network (WSN) applications where all the nodes relay sensed data towards a common destination sink node. The capacity region of a relay region is still an open problem. We use deterministic network model to study the problem. We characterizethe capacity region for a cooperative deterministic network with single source, multiple relays and single destination. We also characterize the capacity region when communicating nodes have correlated information to be sent to the destination. / Cross Layer Optimization of Wireless Sensor Networks

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / Nesse trabalho, propÃe-se um esquema de seleÃÃo eficiente e de baixa complexidade para redes cooperativas multiusuÃrio multi-relay compostas de um nà fonte, L nÃs destinos e N nÃs relays. O esquema proposto primeiro seleciona o melhor destino baseado na qualidade de canal dos links diretos e entÃo seleciona o melhor relay que provà o melhor caminho da fonte para o destino selecionado. Considerando-se os protocolos de cooperaÃÃo decodifica-e-encaminha e amplifica-e-encaminha, o desempenho do sistema à investigado. ExpressÃes em forma fechada para a probabilidade de bloqueio sÃo obtidas e validadas por simulaÃÃes de Monte Carlo. ComparaÃÃes com o esquema de seleÃÃo Ãtimo sÃo realizadas e demonstram que o desempenho do esquema de seleÃÃo proposto à bem prÃximo ao do esquema Ãtimo, com a vantagem de o primeiro possuir uma complexidade menor que o Ãltimo. AlÃm disso, em nossa anÃlise, a fonte pode ser equipada com uma Ãnica antena ou com M mÃltiplas antenas. Uma anÃlise assintÃtica à realizada e revela que, independentemente da estratÃgia de cooperaÃÃo empregada, a ordem de diversidade à de L+N para o caso da fonte com uma Ãnica antena, enquanto que para o caso multiantena a diversidade à igual a ML+N. Os efeitos do nÃmero de nÃs relays e destinos no desempenho do sistema e sua influÃncia na posiÃÃo Ãtima do relay sÃo examinados. AlÃm disso, um compromisso entre desempenho e eficiÃncia espectral à observado para o caso em que mÃltiplas antenas sÃo empregadas. / On this work, it is proposed an efficient low-complexity selection scheme for multiuser multi-relay downlink cooperative networks comprised of one source node, L destination nodes, and N relay nodes. The proposed scheme first selects the best destination node based on the channel quality of the direct links and then selects the best relay that yields the best path from the source to the selected destination. Assuming both decode-and-forward and amplify-and-forward relaying strategies, the performance of the considered system is investigated. Closed-form expressions for the outage probability are obtained and validated by means of Monte Carlo simulations. Comparisons with the optimal selection scheme are performed and shows that the performance of the proposed scheme is very close to that of the optimal selection scheme, with the proposed scheme having the advantage of lower complexity than the optimal scheme. Furthermore, in our analysis, the source node may be equipped with either a single antenna or M multiple antennas. An asymptotic analysis is carried out, and it reveals that, regardless of the relaying strategy employed, the diversity order reduces to L+N for the single-antenna source case, whereas it is equal to ML+N for the multiple-antenna source case. The effects of the number of relay and destination nodes on the system performance and its influence on the best relay position are examined. In addition, a trade-off concerning the system performance and spectral efficiency is observed when multiple antennas are employed at the source node.

<p>This thesis is concerned with scheduling the use of resources, or allocating resources, so as to meet future demands for the entities produced by the resources. We consider applications in mobile communications such as scheduling users' transmissions so that the amount of transmitted information is maximized, and scenarios in the manufacturing industry where the task is to distribute work among production units so as to minimize the number of missed orders.</p><p>The allocation decisions are complicated by a lack of information concerning the future demand and possibly also about the capacities of the available resources. We therefore resort to using probability theory and the maximum entropy principle as a means for making rational decisions under uncertainty.</p><p>By using probabilities interpreted as a reasonable degree of belief, we find optimum decision rules for the manufacturing problem, bidding under uncertainty in a certain type of auctions, scheduling users in communications with uncertain channel qualities and uncertain arrival rates, quantization of channel information, partitioning bandwidth between interfering and non-interfering areas in cellular networks, hand-overs and admission control. Moreover, a new method for making optimum approximate Bayesian inference is introduced.</p><p>We further discuss reasonable optimization criteria for the mentioned applications, and provide an introduction to the topic of probability theory as an extension to two-valued logic. It is argued that this view unifies a wide range of resource-allocation problems, and we discuss various directions for further research.</p>

This thesis is concerned with scheduling the use of resources, or allocating resources, so as to meet future demands for the entities produced by the resources. We consider applications in mobile communications such as scheduling users' transmissions so that the amount of transmitted information is maximized, and scenarios in the manufacturing industry where the task is to distribute work among production units so as to minimize the number of missed orders. The allocation decisions are complicated by a lack of information concerning the future demand and possibly also about the capacities of the available resources. We therefore resort to using probability theory and the maximum entropy principle as a means for making rational decisions under uncertainty. By using probabilities interpreted as a reasonable degree of belief, we find optimum decision rules for the manufacturing problem, bidding under uncertainty in a certain type of auctions, scheduling users in communications with uncertain channel qualities and uncertain arrival rates, quantization of channel information, partitioning bandwidth between interfering and non-interfering areas in cellular networks, hand-overs and admission control. Moreover, a new method for making optimum approximate Bayesian inference is introduced. We further discuss reasonable optimization criteria for the mentioned applications, and provide an introduction to the topic of probability theory as an extension to two-valued logic. It is argued that this view unifies a wide range of resource-allocation problems, and we discuss various directions for further research.