Networks of communicating agents require distributed algorithms for a variety of tasks in the field of network analysis and control. For applications such as swarms of autonomous vehicles, ad hoc and wireless sensor networks, and such military and civilian applications as exploring and patrolling a robust autonomous system that uses a distributed algorithm for self-partitioning can be significantly helpful. A single team of autonomous vehicles in a field may need to self-dissemble into multiple teams, conducive to completing multiple control tasks. Moreover, because communicating agents are subject to changes, namely, addition or failure of an agent or link, a distributed or decentralized algorithm is favorable over having a central agent. A framework to help with the study of self-partitioning of such multi agent systems that have most basic mobility model not only saves our time in conception but also gives us a cost effective prototype without negotiating the physical realization of the proposed idea. In this thesis I present my work on the implementation of a flexible and distributed stochastic partitioning algorithm on the Lego® Mindstorms’ NXT on a graphical programming platform using National Instruments’ LabVIEW™ forming a team of communicating agents via NXT-Bee radio module. We single out mobility, communication and self-partition as the core elements of the work. The goal is to randomly explore a precinct for reference sites. Agents who have discovered the reference sites announce their target acquisition to form a network formed based upon the distance of each agent with the other wherein the self-partitioning begins to find an optimal partition. Further, to illustrate the work, an experimental test-bench of five Lego NXT robots is presented.

A recently suggested algorithm for recursive partitioning of statistical models (Zeileis, Hothorn and Hornik, 2005), such as models estimated by maximum likelihood or least squares, is evaluated in practice. The general algorithm is applied to linear regression, logisitic regression and survival regression and applied to economical and medical regression problems. Furthermore, its performance with respect to prediction quality and model complexity is compared in a benchmark study with a large collection of other tree-based algorithms showing that the algorithm yields interpretable trees, competitive with previously suggested approaches. / Series: Research Report Series / Department of Statistics and Mathematics

The presence of volatile methylsiloxanes (VMS) in the environment has raised concerns among environmental chemists and regulators about their persistence and the risks they may pose to the environment. This thesis explores the partitioning and persistence of VMS in aquatic environments. In Paper I, we reported new measurements of the organic carbon/water (KOC) and dissolved organic carbon/water (KDOC) partition ratios of three cyclic volatile methylsiloxanes (cVMS) and of three polychlorinated biphenyls (PCBs), which were used as reference chemicals. We combined new measurements with existing data to construct polyparameter linear free energy relationships (PP-LFER) that describe the KOC and KDOC of diverse sets of chemicals. The findings suggest that cVMS do not conform to single-parameter regressions that relate the chemicals’ KOC to their octanol/water partition ratio (KOW). PP-LFERs can accurately describe the KOC and KDOC of cVMS but only if cVMS are included in their training sets. In Paper II, we studied the effect of salinity on the KOC and KDOC of three cVMS, two linear volatile methylsiloxanes (lVMS) and three PCBs. We also evaluated the predictive power of the PP-LFERs constructed in Paper I by testing them on the newly measured KOC values of lVMS. The KOC and KDOC increased with increasing salinities similarly to those of the PCBs. PP-LFERs that were trained with datasets that included siloxanes could predict the KOC and KDOC of other siloxanes more accurately than PP-LFERs without siloxanes in the training set. In Paper III, we evaluated the effect of temperature on the KOC of VMS and we compared our measurements of the enthalpy of sorption to organic carbon (ΔHOC) to existing measurements of the enthalpy of phase change between octanol and water (ΔHOW). Due to the scarcity of ΔHOC data in the literature it is common practice in modeling calculations to use ΔHOW instead when correcting for temperature changes. The KOC of cVMS increased with decreasing temperatures. Moreover, our results indicate that ΔHOC and ΔHOW may be intrinsically different and hence replacing ΔHOC with ΔHOW in modeling calculations could lead to substantial errors, especially for VMS. In Paper IV, we explored the environmental fate of VMS in aquatic environments using multimedia models. In particular, we assessed the differences that may occur in calculations of persistence due to (i) the reported KOC measurements of VMS differing by one log unit (ii) the influence of salinity on KOC, and (iii) the differences in the reported ΔHOC and ΔHOW measurements of VMS. The calculated residence times for decamethylcyclopentasiloxane (D5) in a site-specific scenario for a Norwegian fjord receiving siloxanes in wastewater ranged from 200 to 1000 days, and demonstrated that the selection of KOC values can result in substantially different calculated persistence. Future partitioning measurements of VMS in the real environment and mass-balance modeling studies in aquatic environments combined with field measurements could help us to deepen our understanding about their persistence and to assess the risks VMS may pose to the environment. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 1: Manuscript. Paper 3: Submitted.</p>

High levels of soil nitrogen (N) can interfere with N2 fixation of lentil (Lens culinaris) and have variable effects on growth, yield and maturity of this indeterminate crop in Saskatchewan. In a series of field and greenhouse experiments during 2006 to 2008, response of the above-ground biomass (DW), plant N, N2 fixation, yield and days to maturity (DTM) of lentil to different N sources, time of N availability, and also to two no tillage (NT) durations were studied. First, eight cultivars of lentil were grown under three fertility treatments of granular rhizobium inoculant, 50 kg N fertilizer ha-1 and a non-treated control in three environment-years at Saskatoon and Indian Head, SK. The fertility treatments, plant N status and N2 fixation did not alter lentil DTM, but weather did. On average, lentil matured 101 and 84 days after seeding with sufficient rain and with drought, respectively. Growth and yield of the lentil were identical in the inoculant and the N fertilizer treatments. The N fertilizer treatment occasionally restricted N2 fixation, but N shortage was compensated via more N uptake from soil. The greatest N accumulation of lentil occurred during podding to maturity and benefitted pod N content. By maturity, pod, stem and leaf had 60, 24 and 14% of total dry matter and 78, 9 and 13% of total plant N, respectively. Leaf N concentration, which closely resembled soil and plant N status, was reasonably predicted by SPAD chlorophyll meter observations after pod set. Yield of five lentil cultivars was tested for the effects of 25-years (LN) versus 5-years (SN) of no tillage in the Black Soil Zone at Indian Head, SK in 2006, 2007 and 2008. In the same location, CDC Sedley was grown with four N fertilizer rates at the both LN and SN. Under terminal drought in 2006, average DW, N content and yield of the lentil cultivars in SN were greater than in LN, likely because of inhibited N2 fixation by the amplified soil N in the LN. In this year, 60 kg N fertilizer ha-1 reduced the yield difference of CDC Sedley in SN and LN. Lentil yield was identical or tended to be greater in LN than in SN with more rain in 2007 and 2008 that prolonged N mineralization and N uptake. In the greenhouse study, applying N fertilizer from flowering until podding and until maturity increased DW, N content and yield, and delayed maturity of lentil compared to lentil relying on N2 fixation. Later flowering of one cultivar or greater N2 fixation in one soil medium diminished the variation of inoculated lentil with the post-flowering N treatments, suggesting N fixation could supply lentil N requirement. Large-seeded cultivars produced greater yield than the small-seeded cultivars across environments in the fertility treatment study. Cultivar CDC Milestone produced comparable yield to high-yielding cultivars CDC Plato and CDC Greenland, but matured earlier. This cultivar showed promising results under both cool-wet and drought conditions. In contrast, CDC Sedley had lower on N2 fixation and HI values across the experiments. In the Black Soil Zone, CDC Milestone and CDC Robin performance was improved by improved HI and N2 fixation. Overall, results of this thesis do not support the application of N fertilizer for inducing early maturity in lentil. Soil inoculation with commercial strains is suggested for Saskatchewan cropping systems. Applying N fertilizer is not required, unless soil test results suggest otherwise. In places like Indian Head, SK, cultivars with greater N2 fixation and higher HI can better fit the short growing season, cool temperature and high soil N content.

The objective of this study was to determine the dominant biogeochemical controls on mercury partitioning in super-permafrost groundwater at Truelove Lowland, Devon Island, Nunavut. Mercury partitioning in snow, ephemeral standing water, and super-permafrost groundwater was investigated. <p>Results indicate that partitioning differs between matrices, and that particulate mercury is spatially and temporally dynamic in Truelove Lowland groundwater. Particulate mercury in groundwater was 73 % of total mercury, while snow had only 22 % particulate mercury. Particulate mercury in groundwater rose by over 20 % from Julian day 181 to 189, and decreased slightly on Julian day 191. No single geochemical parameter was a good predictor of particulate mercury concentrations. To expand upon the findings of the field study a laboratory microcosm study was conducted to determine whether certain biogeochemical processes influence mercury partitioning in super-permafrost groundwater. Particulate mercury in the dissimilatory iron reducing bacteria inhibited microcosm was 61 % of total mercury, approximately 18 % lower than in all other treatments. Iron (III) concentrations had a positive correlation with particulate mercury while chloride concentrations had a negative correlation with particulate mercury. Sulfate reducing bacteria were not found to influence mercury partitioning.

High levels of soil nitrogen (N) can interfere with N2 fixation of lentil (Lens culinaris) and have variable effects on growth, yield and maturity of this indeterminate crop in Saskatchewan. In a series of field and greenhouse experiments during 2006 to 2008, response of the above-ground biomass (DW), plant N, N2 fixation, yield and days to maturity (DTM) of lentil to different N sources, time of N availability, and also to two no tillage (NT) durations were studied. First, eight cultivars of lentil were grown under three fertility treatments of granular rhizobium inoculant, 50 kg N fertilizer ha-1 and a non-treated control in three environment-years at Saskatoon and Indian Head, SK. The fertility treatments, plant N status and N2 fixation did not alter lentil DTM, but weather did. On average, lentil matured 101 and 84 days after seeding with sufficient rain and with drought, respectively. Growth and yield of the lentil were identical in the inoculant and the N fertilizer treatments. The N fertilizer treatment occasionally restricted N2 fixation, but N shortage was compensated via more N uptake from soil. The greatest N accumulation of lentil occurred during podding to maturity and benefitted pod N content. By maturity, pod, stem and leaf had 60, 24 and 14% of total dry matter and 78, 9 and 13% of total plant N, respectively. Leaf N concentration, which closely resembled soil and plant N status, was reasonably predicted by SPAD chlorophyll meter observations after pod set. Yield of five lentil cultivars was tested for the effects of 25-years (LN) versus 5-years (SN) of no tillage in the Black Soil Zone at Indian Head, SK in 2006, 2007 and 2008. In the same location, CDC Sedley was grown with four N fertilizer rates at the both LN and SN. Under terminal drought in 2006, average DW, N content and yield of the lentil cultivars in SN were greater than in LN, likely because of inhibited N2 fixation by the amplified soil N in the LN. In this year, 60 kg N fertilizer ha-1 reduced the yield difference of CDC Sedley in SN and LN. Lentil yield was identical or tended to be greater in LN than in SN with more rain in 2007 and 2008 that prolonged N mineralization and N uptake. In the greenhouse study, applying N fertilizer from flowering until podding and until maturity increased DW, N content and yield, and delayed maturity of lentil compared to lentil relying on N2 fixation. Later flowering of one cultivar or greater N2 fixation in one soil medium diminished the variation of inoculated lentil with the post-flowering N treatments, suggesting N fixation could supply lentil N requirement. Large-seeded cultivars produced greater yield than the small-seeded cultivars across environments in the fertility treatment study. Cultivar CDC Milestone produced comparable yield to high-yielding cultivars CDC Plato and CDC Greenland, but matured earlier. This cultivar showed promising results under both cool-wet and drought conditions. In contrast, CDC Sedley had lower on N2 fixation and HI values across the experiments. In the Black Soil Zone, CDC Milestone and CDC Robin performance was improved by improved HI and N2 fixation. Overall, results of this thesis do not support the application of N fertilizer for inducing early maturity in lentil. Soil inoculation with commercial strains is suggested for Saskatchewan cropping systems. Applying N fertilizer is not required, unless soil test results suggest otherwise. In places like Indian Head, SK, cultivars with greater N2 fixation and higher HI can better fit the short growing season, cool temperature and high soil N content.

Hardware/software partitioning is a process that occurs frequently in embedded system design. It is the procedure of determining whether a part of a system should be implemented in software or hardware. This dissertation is a study of hardware/software partitioning and the use of scheduling algorithms to improve the performance of dynamically reconfigurable computing devices. Reconfigurable computing devices are devices that are adaptable at the logic level to solve specific problems [Tes05]. One example of a reconfigurable computing device is the field programmable gate array (FPGA). The emergence of dynamically reconfigurable FPGAs made it possible to configure FPGAs at runtime. Most current approaches use a simple on demand configuration scheduling algorithm for the FPGA configurations. The on demand configuration scheduling algorithm reconfigures the FPGA at runtime, whenever a configuration is needed and is found not to be configured. The problem with this approach of dynamic reconfiguration is the reconfiguration time overhead, which is the time it takes to reconfigure the FPGA with a new configuration at runtime. Configuration caches and partial configuration have been proposed as possible solutions to this problem, but these techniques suffer from various limitations. The emergence of dynamically reconfigurable FPGAs also made it possible to perform dynamic hardware/software partitioning (DHSP), which is the procedure of determining at runtime whether a computation should be performed using its software or hardware implementation. The drawback of performing DHSP using configurations that are generated at runtime is that the profiling and the dynamic generation of configurations require profiling tool and synthesis tool access at runtime. This study proposes that configuration scheduling algorithms, which perform DHSP using statically generated configurations, can be developed to combine the advantages and reduce the major disadvantages of current approaches. A case study is used to compare and evaluate the tradeoffs between the currently existing approach for dynamic reconfiguration and the DHSP configuration scheduling algorithm based approach proposed in the study. A simulation model is developed to examine the performance of the various configuration scheduling algorithms. First, the difference in the execution time between the different approaches is analyzed. Afterwards, other important design criteria such as power consumption, energy consumption, area requirements and unit cost are analyzed and estimated. Also, business and marketing considerations such as time to market and development cost are considered. The study illustrates how different types of DHSP configuration scheduling algorithms can be implemented and how their performance can be evaluated using a variety of software applications. It is also shown how to evaluate when which of the approaches would be more advantageous by determining the tradeoffs that exist between them. Also the underlying factors that affect when which design alternative is more advantageous are determined and analyzed. The study shows that configuration scheduling algorithms, which perform DHSP using statically generated configurations, can be developed to combine the advantages and reduce some major disadvantages of current approaches. It is shown that there are situations where DHSP configuration scheduling algorithms can be more advantageous than the other approaches.

One question relating to partially ordered sets (posets) is that of partitioning or dividing the poset's elements into the fewest number of chains that span the poset. In 1950, Dilworth established that the width of the poset - the size of the largest set composed only of incomparable elements - is the minimum number of chains needed to partition that poset. Such a bound in on-line partitioning has been harder to establish, and work has evalutated classes of posets based on their width. This paper reviews the theorems that established val(2)=5 and illustrates them with examples. It also covers some of the work on establishing bounds for on-line partitioning with the Greedy Algorithm. The paper concludes by contributing a bound on incomparable elements in graded, (t+t)-free, finite width posets.

Hardware/software partitioning is a process that occurs frequently in embedded system design. It is the procedure of determining whether a part of a system should be implemented in software or hardware. This dissertation is a study of hardware/software partitioning and the use of scheduling algorithms to improve the performance of dynamically reconfigurable computing devices. Reconfigurable computing devices are devices that are adaptable at the logic level to solve specific problems [Tes05]. One example of a reconfigurable computing device is the field programmable gate array (FPGA). The emergence of dynamically reconfigurable FPGAs made it possible to configure FPGAs at runtime. Most current approaches use a simple on demand configuration scheduling algorithm for the FPGA configurations. The on demand configuration scheduling algorithm reconfigures the FPGA at runtime, whenever a configuration is needed and is found not to be configured. The problem with this approach of dynamic reconfiguration is the reconfiguration time overhead, which is the time it takes to reconfigure the FPGA with a new configuration at runtime. Configuration caches and partial configuration have been proposed as possible solutions to this problem, but these techniques suffer from various limitations. The emergence of dynamically reconfigurable FPGAs also made it possible to perform dynamic hardware/software partitioning (DHSP), which is the procedure of determining at runtime whether a computation should be performed using its software or hardware implementation. The drawback of performing DHSP using configurations that are generated at runtime is that the profiling and the dynamic generation of configurations require profiling tool and synthesis tool access at runtime. This study proposes that configuration scheduling algorithms, which perform DHSP using statically generated configurations, can be developed to combine the advantages and reduce the major disadvantages of current approaches. A case study is used to compare and evaluate the tradeoffs between the currently existing approach for dynamic reconfiguration and the DHSP configuration scheduling algorithm based approach proposed in the study. A simulation model is developed to examine the performance of the various configuration scheduling algorithms. First, the difference in the execution time between the different approaches is analyzed. Afterwards, other important design criteria such as power consumption, energy consumption, area requirements and unit cost are analyzed and estimated. Also, business and marketing considerations such as time to market and development cost are considered. The study illustrates how different types of DHSP configuration scheduling algorithms can be implemented and how their performance can be evaluated using a variety of software applications. It is also shown how to evaluate when which of the approaches would be more advantageous by determining the tradeoffs that exist between them. Also the underlying factors that affect when which design alternative is more advantageous are determined and analyzed. The study shows that configuration scheduling algorithms, which perform DHSP using statically generated configurations, can be developed to combine the advantages and reduce some major disadvantages of current approaches. It is shown that there are situations where DHSP configuration scheduling algorithms can be more advantageous than the other approaches.

The objective of this study was to determine the dominant biogeochemical controls on mercury partitioning in super-permafrost groundwater at Truelove Lowland, Devon Island, Nunavut. Mercury partitioning in snow, ephemeral standing water, and super-permafrost groundwater was investigated. <p>Results indicate that partitioning differs between matrices, and that particulate mercury is spatially and temporally dynamic in Truelove Lowland groundwater. Particulate mercury in groundwater was 73 % of total mercury, while snow had only 22 % particulate mercury. Particulate mercury in groundwater rose by over 20 % from Julian day 181 to 189, and decreased slightly on Julian day 191. No single geochemical parameter was a good predictor of particulate mercury concentrations. To expand upon the findings of the field study a laboratory microcosm study was conducted to determine whether certain biogeochemical processes influence mercury partitioning in super-permafrost groundwater. Particulate mercury in the dissimilatory iron reducing bacteria inhibited microcosm was 61 % of total mercury, approximately 18 % lower than in all other treatments. Iron (III) concentrations had a positive correlation with particulate mercury while chloride concentrations had a negative correlation with particulate mercury. Sulfate reducing bacteria were not found to influence mercury partitioning.