Modern vehicles have an increasing number of advanced features requiring network coverage in order to function properly. In order to facilitate the requirements of such features and allow more advanced applications, we consider the possibility of planning routes that take signal strength into consideration. Previous work have shown the relationship between TCP throughput/goodput and signal strength. In this thesis signal-aware route planning is presented, implemented, and validated. Crowd-sourced map and signal data (3G) from two sources is used for building a signal coverage map. The signal and map data is validated in a field experiment, where routes were travelled while measuring the signal strength. The field experiment showed gains in signal characteristics when deviating from the shortest possible path. The average signal strength increased by 11 dBm between algorithms and the shortest possible path. Lastly, routes were planned for all possible sources and destinations in a given urban area. The results of this calculation confirms the patterns found in the field experiment.

Floodplains are among the most productive and biologically diverse freshwater ecosystems on earth. The exchange of nutrients and biota that occurs within these systems during seasonal inundation is essential in maintaining floodplain and river health. Anthropogenic structures, such as weirs, channels, and dams, have altered the natural flood hydrology of floodplain systems minimizing the frequency, strength and duration of flood events. This reduction ultimately leads to the isolation of important floodplain habitat, such as off-channel lakes, from the main channel, decreasing connectivity. Although some studies have examined the productivity of off-channel floodplain lakes in relation to connectivity, most are limited to tropical or highly degraded systems. Northern floodplains are not as well understood, with most of the research limited to the spring, summer, and fall seasons, when waterbodies are free of ice. With research limited to ice free seasons, there is not a full understanding of the year-round processes that occur within these off-channel lake habitats. This knowledge is crucial as the winter season is often when conditions within these habitats are at their most extreme. Such conditions prevent many fish species from permanent settlement; however, no research has been attempted to understand fish presence within these habitats during the winter season. In tropical systems, hypoxia-tolerant species and juveniles utilize these habitats as refuge from intolerant predators, so such habitat may be used similarly in more northern systems. The purpose of this research was to understand the connectivity, limnology and suitability as fish habitat of off-channel floodplain lakes in the Saskatchewan River Delta (SRD), SK, during winter months. I determined the degree of connectivity to the main channel for 26 individual lakes within the SRD by two modern methods: remote sensing imagery, and stable isotopes (δ18O, δ2H). Both of these techniques proved effective at determining connectivity of individual lakes and showed good agreement, with lakes arranged into five connectivity categories using remote sensing imagery. Winter limnological conditions within these lakes were significantly influenced by their degree of connectivity, with lakes that were more connected having characteristics similar to that of the river, with higher levels of dissolved oxygen (DO), nitrates (NO3-NO2), pH, and lower levels of nutrients (TN,TP). Lakes that were less connected were characterized by low levels of DO and nutrients, and high levels of ammonia/ammonium (NH3-NH4), conditions that are not favourable for the survival of many fish species. Some of the more hypoxia-tolerant species found within the SRD appear, however, to use these habitats in the winter. This was supported by detection of fish presence using environmental DNA; five fish species were detected in many of the 26 lakes sampled, but only in lakes with NH3-NH4 levels below 1.77 mg/L and volumes greater than 178000 m3. Together, these analyses suggest the influence of a spring/summer flood pulse on limnology is not limited to the months following a flood event, but rather extends well into the ice-cover season. This knowledge is critical as it points to controls on key processes (e.g. nutrient cycling, provision of fish habitat) during the period when lake conditions are most severe. As a result of human induced climate change, and from increased water demands for agriculture and hydropower, the natural flood pulse is expected to further decrease in size and frequency in large river-wetlands such as the SRD. This will reduce the connection between the floodplain and the main channel, with profound impacts on the SRD ecosystem as a whole. Lakes that currently experience frequent inundation will likely have conditions characteristic of infrequently flooded lakes, with low DO and nutrients and high NH3-NH4. Lakes which currently experience infrequent inundation will likely dry up completely due to decreased water renewal.

The spatial distribution and properties of rainfed orchards in semi-arid environments result from complex interactions between man and the physical and economical environment. This thesis investigates a number of these interactions in the context of the mechanisation of management practices since the 1950's. It is shown how the practice of clean sweeping (i.e. frequent shallow tillage) influences the orchard water balance and how the removal of traditional soil and water conservation structures affects the connectivity of overland flow to the river system. Although clean sweeping prevents transpiration and competition by weeds, it also constrains the root growth in the plough layer, so that the trees cannot access the water from small rain events. In addition, clean sweeping promotes accelerated soil erosion. It appears that the practice of clean sweeping limits the water availability in orchards on loamy soils with an annual rainfall in the order of 300 mm. It is demonstrated that the presence and properties of rainfed orchards are related to spatial patterns of soil characteristics and climate. The observed decline in conservation structures like terraces and check-dams leads to an increase in the connectivity of water and sediment to the river system. An alternative for these traditional techniques to retain the water and the soil is the application of cover crops. The advantage of cover crops is that they do not limit the field size. A drawback in dry areas is the competition for water and nutrients between the cover crop and the trees. Field evidence and water balance simulations suggest that cover crops are feasible in areas with an annual precipitation of 500 mm or more.

Network design problems involving survivability usually include trade-off of the potential for lost revenues and customer goodwill against the extra costs required to increase the network survivability. It also involves selection of nodes and edges from lists of potential sets to accomplish certain desirable properties. In many applications it is imperative to have built-in reliability or survivability of the network. Delays of traffic are undesirable since it affects quality of service (QoS) to clients of the network. In this dissertation we consider the construction of an optimization system for network design with survivability properties that may help in the planning of mesh topologies while maintaining a certain degree of survivability of the network. This is done by providing for at least two diverse paths between certain "special" nodes to provide protection against any single edge or node failure. This part is modelled by using mixed integer programming techniques. A software product called CPLEX then solves these models and various facilities are built into the decision support system to allow the decision maker to experiment with some topological and flow requirement changes. / Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2007

The purpose of this paper is to investigate some of the more basic properties of peripherally continuous functions, graph maps and connectivity maps.

A wireless sensor network is a network consisting of spatially distributed, sometimeautonomous sensors, communicating wirelessly to cooperatively achieve some task. For example, a wireless sensor network may be used for habitat monitoring to ascertain the environment’s temperature, pressure, humidity, etc. In order for a wireless sensor network to provide such data, one needs to ensure there is connectivity between nodes. That is, nodes can communicate to exchange information. To analyze connectivity between sensors, the radio communication range of each sensor, also called the communication footprint, needs to be known. To date, the models used to analyze a sensor’s radio communication footprint have been overly simplistic (i.e., isotropic) and thus yield results not found in practice. Footprints are highly dependent on the deployment environments, which are typically heterogeneous and non-isotropic in structure. In this work, a ‘weak-monotonicity’ (W-M) model is leveraged to represent a footprint’s non-isotropic behavior. The work also considers the heterogeneity of the environment through the use of the log-normal shadowing model. In particular, the usable percentage of the W-M footprint (the area where the power exceeds the receiver threshold) in such environments is considered through analysis and simulation. We then develop an enhanced footprint model which overlays multiple W-M patterns and use this method to represent experimental propagation data. The work also considers the use of graph theory methods to analyze the connectivity of randomly deployed networks in nonhomogeneous, non-isotropic environments.

Protected forests need to be a representative selection of the natural proportion of forest types, including distribution of productivity levels, age classes and nature types This is important for the possibility to preserve biodiversity. In addition, the protected areas has to be of sufficient size and not isolated from each other, to function as effective biodiversity preservers. The question is, how does it look in reality? The objective with this study was to get an overall picture of the current forest protection situation in Östergötland, Sweden, and how it has changed the last 60 years. Are all ecologically relevant forest habitat types represented in appropriate proportions in protected forests? To evaluate where the protected areas are located in relation to each other, a connectivity index was calculated for each patch of protected area. Together with a value for size, a value index was created and applied to all protected areas, and it turns out that the protected areas of Östergötland is not totally representative when it comes to nature types, age classes and levels of productivity.For example, there is an underrepresentation of both pine and spruce forests on high-productivity soils. However, areas with higher productivity levels have been protected over time. The age distribution seems to be skewed towards older forests in protected areas. There are some underrepresented nature types, as well as overrepresented ones in nature reserves, a small overrepresentation of unproductive impediments, and only spruce and mixed forests seems well connected in the landscape. The greatest differences in protected and unprotected forests is the productivity level, were focus should be on protecting higher productivity areas in order to succeed in preserving the biodiversity of forests as intended.

This dissertation adds to the many arguments already made for the value of art (cultural artifact) in teaching and learning. The special approach developed here concludes with the articulation of Salience Strategy. The argument firstly questions the value of seeing intelligence as a problem-solving faculty. It continues by examining consciousness, memory and the imagination as both the ground and substance of intellection. It argues that, amongst other things, interconnectedness, reiterative pathways and networks are central to the operation of consciousness and therefore, are central to its epiphenomenal attributes like intelligence. As education should strive for greater intellectual functioning so it should, therefore, strive to harness the paradigms of interconnectedness, reiterative pathways and networks. The art object, (device, gesture, statement), it is proposed, is valuable when deployed as hubs in networks of ideas allowing learners to form patterns of unexpected and creative linkages enhancing both memory, curiosity and a capacity for imaginative and associative thinking. Learning becomes movement through a landscape of complex objects and outgrowths. Two salience itineraries are explored in this dissertation. The first in relation to concepts overheard during learner conversations over the duration of a school week, and a second, exploiting my own work as an artist, selected work by the British artist Richard Long, and some of the issues raised in the theoretical discussion of consciousness and networks.

Thesis advisor: Elizabeth A. Kensinger / Although healthy young and older adults remember emotional information better than neutral, emotion does not confer the same benefit upon memory for those experiencing memory impairments due to Alzheimer's disease (AD). It is poorly understood at what stage of processing these deficits occur--are they due to declines in memory storage and retrieval processes, or to a decline in earlier stages of attention allocation, which then impact memory storage and retrieval? It remains an open question how attention and memory processes may interact in aging and age-related disease. The goal of this research was to examine the effects of aging on the neural mechanisms underlying selective memory for emotional information in visual scenes, and to compare memory between healthy older adults and patients with very early AD pathophysiological changes. Experiment 1 examined young and older adults' encoding-related neural activation associated with selective memory for emotional items within visual scenes and with successful memory for emotional items and the scene background. There were few regions showing significant interactions between age and memory for positive and negative scenes. In contrast, Experiment 2 showed that aging significantly affected the neural networks underlying selective emotional item memory and successful memory for emotional items and backgrounds. The results indicate that older adults require greater connectivity among prefrontal regions than young adults to encode all elements of a scene, rather than just encoding the emotional item. Experiment 3 showed that despite poorer memory overall, patients showing very early AD pathophysiological changes have relatively well preserved memory, especially for positive information. Dividing older adults' attention during encoding did not significantly alter their pattern of selective emotional item memory, suggesting that encoding of emotional items may be an easier or relatively automatic task compared to encoding of the background. In conclusion, there are significant age-related changes in the underlying neural networks, but not activation patterns, for selective memory for positive and negative scenes. Patients with early AD pathophysiological changes have impaired memory overall, however they may be able to recruit a similar neural network of prefrontal regions as healthy older adults for encoding of scenes with positive information. / Thesis (PhD) — Boston College, 2011. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Psychology.

Robot swarms are often considered suitable for tasks that are large-scale and long-term. Large-scale missions force the robots to spread spatially. In these type of tasks, actively maintaining connectivity allows the swarm to coordinate. Similarly, long-term nature of the task requires robots to work for a long time. This is affected by the limited energy level of the robot. However current studies normally focus only on connectivity or energy awareness. Therefore, in this work, we propose an approach to tackle the problem of maintaining global connectivity (swarm-level property) considering finite battery life (individual property). We are specifically focusing on growing the communication network and keeping it alive for a long period. We construct a logical tree over the connectivity graph. The logical tree is constructed by using a subset of robots from the swarm. The tree is grown by adding robots as necessary. The tree is also periodically reconfigured to cope with dynamic robot motion. This enables the swarm to grow the tree efficiently. In addition, robots exchange their roles based on their available energy levels. This allows robots with low energy levels to navigate to dedicated charging stations for recharging thus allowing the swarm to maintain the communication network. We evaluate our approach in a wide set of experiments with a realistic robot simulator named ARGoS.