Digitized University of Waterloo Theses

Permanent URI for this collectionhttps://hdl.handle.net/10012/21195

The following collection includes theses created by UW graduate students prior to 2010 that have since been digitized. Please note that not all theses written by UW graduate students have been digitized. Current graduate students should only submit their work in the Theses collection.

Browse

Now showing 1 - 20 of 689

The declining role of the earth sciences in the planning, management and interpretation of Canada's national parks, identification, documentation, possible causes, and means of reversal
(University of Waterloo, 1996) Campbell, John Michael
This dissertation identifies a reduction which is currently occurring in the role of the earth sciences in the planning, management and interpretation of the national parks of Canada. Through the analysis of the Canadian National Parks Act, National Park Policies, Systems Plan and the Natural Resource and Visitor Activity Management Processes, as well as information on experience in the United States, New Zealand and other countries, it is shown that there is and has been a role for the earth sciences in the planning, management and interpretation of the parts. Case studies of ten management plans from nine parks indicated that the natural resource management functions at individual national parks rarely identified the earth sciences as high priority management concerns. Detailed case studies of Banff and the associated mountain parks (Yoho, Kootenay, Jasper), and Point Pelee National Park, along with interviews of park personnel. confirmed the findings that the earth sciences have never been afforded the same degree of management concern in the sample parks as have the biotic resources. The development of ecosystem-based management in response to the 1988 amendments to the National Parks Act has further reduced the role of the earth sciences, due largely, to the definition of ecological integrity as a biological concept. In contrast, interpretation messages developed in the 1970's included the earth science messages to the same degree as biotic messages. However, the increasing focus of interpretation on the communication of resource management goals has resulted in significant reduction of earth science messages. Indeed, in Point Pelee the earth sciences have been completely omitted from the message framework while in Banff the 1993 ecosystem plan makes no reference to the earth sciences resources of the park. A number of explanations are suggested for this reduction, including the definition of ecological integrity as a biological concept, the lack of earth science expertise within the park system's staff, the engineering focus of earth science management. and the failure of academic geomorpbologists to become involved in the identification of earth science measures of ecological integrity. The dissertation concludes with a number of recommendations aimed at reversing the decline of the earth sciences in the planning, management and interpretation of the national parts. Recommendations directed towards Parks Canada include developing a balance in the park staff and staffing structure through the identification of some positions as earth science specialists. This could be accomplished as part of the redefinition of the roles of wardens. Research and training partnerships were identified as other means of beginning to address the reduction. It is recommended that earth scientists take advantage of the opportunities in the developing field of natural areas management. and that they seek to identify meaningful earth science measures of ecological integrity. It is also recommended that earth scientists conducting research in the parks be encouraged to become involved in the management and planning process. The study develops and presents an interpretation program for Banff National Park which links some of the features and processes of the park to ecosystem-based management and education. Finally, the study identifies a significant new area for research and application of the earth sciences within the evolving field of parks and protected areas planning and management.
Falls among the elderly, risk factors and prevention strategies
(University of Waterloo, 1996) Fletcher, Paula C.
Morphological and biochemical variability within eastern North American populations of Asterionella Hassall, possible taxonomic implications
(University of Waterloo, 1996) McIntyre, Shona Helen
Synthetic studies related to reductively activated antitumor antibiotics
(University of Waterloo, 1996) Mithani, Salim
Raman spectral studies of solutions of formic acid and methyl formate
(University of Waterloo, 1996) Bartholomew, Richard J.
Blood flow kinetics at the onset of moderate exercise in the elderly
(University of Waterloo, 1996) Gobbi, Sebastiao
A framework for on-line partial evaluation
(University of Waterloo, 1996) Vreugdenhil, Gordon J.
The level of organizational effectiveness of the (IC) construction firm, a multivariate model-based prediction methodology
(University of Waterloo, 1996) Adas, Adnan A.
Bayesian nonparametric survival analysis
(University of Waterloo, 1997) Yuan, Lin
This thesis makes contributions to the Bayesian nonparametric approach for survival and bioassay problems. It contains creative work towards a simple and practical Bayesian analysis for right-censored failure time data using smoothed prior, and for binary and doubly-censored data using the Dirichlet process prior. One-sample survival analysis under a smoothed prior is fully studied. The posterior computations are realized via the Gibbs sampler, and illustrated by unmerical examples. Bayesian inference under non-informative priors is addressed and compared with existing results. A compromised version of Bayesian nonparametric approach is proposed which retreats from the infinite-dimensional priors and considers a more practical treatment using data-dependent priors. Links to some well-known results such as Cox's partial likelihood for proportional hazards regression and Hill's rule for prediction are established. Fiducial inference for failure time data is also discussed, which is numerically equivalent to the Bayesian approach under a non-informative and data-dependent prior. A new auxiliary variables technique is proposed which has substantially simplified the Bayesian bioassay under a Dirichlet process prior, and application is illustrated in cancer risk assessment. The problem of combining many assays is discussed in the empirical Bayes framework, and more complicated types of data such as doubly-censored data are also considered.
Huygens' principle for relativistic wave equations on Petrov type III space-times
(University of Waterloo, 1997) Sasse, Fernando Deeke
This thesis makes a contribution to the solution of Hadamard's problem for some relativistic wave equations on Petrov type III space-times. For the conformally invariant scalar wave equation we show that if any one of the spin coefficients a, /3, tr or Ricci spin coefficient ~ 11 vanishes in an appropriate null tetrad, then Huygens' principle is not satisfied on Petrov type III space-times. We also show that the corresponding problem for the non-self-adjoint scalar wave equation can be reduced to the conformally invariant scalar equation case. Finally, we prove that there are no Petrov type Ill space-times on which either the conformally invariant scalar equation, Weyl 's neutrino equation or Maxwell's equations satisfy Huygens' principle in the strict sense. In order to obtain the above results we have employed Newman-Penrose spin coefficient formalism and Penrose's two-component spinor formalism, together with their implementations available in the computer algebra system Maple, to determine the components of the tensorial relations given by the imposition of Huygens' principle. The resulting system of polynomial equations is then analysed by using a variant of Buchberger's algorithm, available in Maple.
Applications of orthonormal bases of wavelets to deconvolution
(University of Waterloo, 1997) Corbett, Norman Christopher
Convolution integral equations arise frequently in many areas of science and engineering. If the kernel of such an equation is well behaved, say integrable, then the task of solving a convolution equation is ill-posed. Indeed, if the kernel is integrable, then the Riemann-Lebesgue Lemma implies that the recovery of high frequency information pertaining to the unknown function will be difficult, if not impossible. Orthonormal wavelet bases are bases generated by translating and dilating a single function, known as the mother wavelet. One key advantage of these bases is that the mother wavelet can be selected to have fast decay in both the time and frequency domains. This property suggests that wavelet bases may be useful when attempting to solve a convolution equation. In this thesis, we investigate the applicability of orthonormal wavelet bases with regard to solving convolution equations. In particular, we concentrate on the construction of approximations to the unknown function belonging to scaling function subspaces. We also briefly consider regularization algorithms which are based on the multiresolution analysis, a structure defined by the scaling function association with the mother wavelet.
Randomized trial of telephone counselling in association with the Guide Your Parents to a Smoke-Free Future program
(University of Waterloo, 1997) Reid, Robert Donald
Objective: To evaluate the incremental benefit of telephone counselling in association with the Guide Your Patients to a Smoke-Free Future program. Design: Clinical trial with stratification (by gender & degree of nicotine dependence) and random assignment to Guide Your Patients (GYP) or Guide Your Patients+ Telephone Counselling (GYP+TC) group. Setting: Smoking Cessation Clinic at the Ottawa Heart Institute. Participants: Volunteer sample of 396 smokers(~ 15 cigarettes/day), free of major health problems, interested in quitting smoking within 30 days. Interventions: Physician advice on three occasions according to the Guide Your Patients handbook, self-help materials and 12 weeks of nicotine replacement therapy, with ( Guide Your Patients+ Telephone Counselling group), or without (Guide Your Patients group) nurse-mediated telephone counselling two, six, and 13 weeks after a target quit date. Main Outcome :Measures: Smoking status (point-prevalent abstinence, continuous abstinence, and time to relapse) at 26-week follow-up; processes of change, self-efficacy. and perceived stress at baseline, four and 12 weeks after target quit date. Results: There was no difference in the 26-week point prevalent abstinence rate (29 .6% vs. 26.9%; P-Value=.54) or continuous abstinence rate (25.6% vs. 25.4%; P Value=.96) between the Guide Your Patients and Guide Your Patients+ Telephone Counselling groups, respectively. Survival analysis showed no difference between the relapse curves for the two groups (median time to relapse= I IO vs. 92 days; P-V aJue=. lO). Survival analysis within subgroups revealed an unexpected reduction in the survival function for low nicotine-dependent males receiving telephone counselling (median time to relapse= 99 vs. 187 days; P-Value=.0l). Repeated measures ANOVA analysis of process of change data revealed significant increases in the use of consciousness raising, self-liberation, counterconditioning, stimulus control reinforcement management, and helping relationships over time, but no significant interactions between treatment condition and changes in use of processes of change. Successful quitters endorsed significantly less use of self-reevaluation and greater use of counterconditioning and helping relationships. Repeated measures ANOVA analysis of self-efficacy data revealed significant increases in total confidence and confidence in social, negative affect and habitual situations over time during treatment, but no effect of treatment condition. Successful quitters had significantly higher levels of total confidence and confidence in social. negative affect and habitual situations over time during treatment. Perceived stress during treatment was unaffected by the treatment group assignment. Successful quitters had significantly lower levels of perceived stress at baseline and four and 12 weeks after the target quit date. Logistic regression analysis revealed three significant univariate baseline predictors of cessation: level of nicotine dependence: education level: and perceived stress. The odds of being abstinent at 26-week follow-up were increased by having more than a high school education (OR: 959c CI=2.3: l.44. 3.68). The odds of being abstinent were reduced by having a Fagerstrom Tolerance Questionnaire Score 2:: 7 (OR; 95lk CI= 0.63: 0.40. 0.99) or a Perceived Stress Score 2:: 8 (OR: 95Ck CI = 0.39: 0.22. 0.69). Conclusions: Physician assistance, using the Guide Your Patients program. and incorporating nicotine replacement therapy. is enough to help many smokers. Quit rates are not improved by additional nurse-mediated telephone counselling. Telephone counselling may be counterproductive in low nicotine-dependent males. Telephone counselling did not incrementally enhance the stage appropriate use of processes of change or the development of self-efficacy. This study does not rule out the possibility that telephone counselling may benefit smokers in earlier stages of preparedness to quit. smokers receiving less intense intervention or less than optimal assistance from their physician, or smokers who self-select telephone counselling. This study also does not rule out the possibility that a different telephone intervention or altered timing of the calls could have yielded different results.
New applications of elliptic curves and function fields in cryptography
(University of Waterloo, 1997) Zuccherato, Robert
Public key cryptography based on elliptic curves over finite fields was proposed by Miller and Koblitz in 1985. Elliptic curves over finite fields have been used to implement the Di:ffie-Hellm.an key passing scheme and the ElGamal, Schnorr and NIST signature schemes. Elliptic curves have also been used over the ring Zn to implement an RSA type scheme. In the first part of this thesis however, we propose using elliptic curves over the ring ~n in a new way. In this system the information is carried in the exponent space and not in the group itself. Also security depends on the difficulty of factoring a 150 digit number in order to trapdoor the discrete logarithm problem. The continued fraction expansion and infrastructure for quadratic congruence function fields of odd characteristic have been well studied. Recently, these ideas have even been used to produce cryptosystems. Much less is known concerning the continued fraction expansion and infrastructure for quadratic function fields of even characteristic. In the second part of this thesis we will explore these ideas, and show that the situation is very similar to the odd characteristic case. This exploration will result in a method for computing the regulator for quadratic function fields of characteristic 2. We will also be able to show that cryptosystems proposed for the infrastructure of function fields of odd characteristic can be implemented in even characteristic and give a possible attack. Most importantly we will be able to show that the elliptic curve discrete logarithm problem is equivalent to a discrete logarithm problem in the infrastructure of certain quadratic function fields. This is a modification of a result by Stein for fields of odd characteristic.
Canada's India policy, 1947-1997, the emerging policy agenda
(University of Waterloo, 1997) Narain, Nigmendra
This thesis examines Canada's India policy between 1947 and 1997, with greater emphasis on the post-1990 period to differentiate it from previous works. As well its focus is on Canadian foreign policy perspectives of Canada-India relations. First, the thesis examines three scholarly contexts: L) niche diplomacy; 2) the business-government relationship in foreign policy; and, 3) the American factor in Canadian foreign policy. Second, the thesis thematically chronologizes formative and indicative events in Canada's India policy. Third, the thesis critically examines the literature about Canada-India relations and draws analyses and conclusions. Finally, the thesis considers the three scholarly contexts in light of the history of Canada's India policy. The thesis's major findings are: I) while the America factor impinges less on Canadian foreign policy now, America effects security issues more than others, and a greater problem for Canadians and Indians is that they view each other through American lens and examples; 2) while the Canadian government is in control and formulating Canadian foreign policy in the state's interest, business occupies both an adversarial and opportunistic relationship to the government and foreign policy; and. 3) niche diplomacy is the best approach for Canada's India policy to engage India, with more emphasis on impact, results and public diplomacy.
Surface-subsurface transport cycle of chloride induced by wetland-focused groundwater recharge
(University of Waterloo, 1997) Hayashi, Masaki
The glacial plain in the northern prairie region has numerous wetlands. These wetlands provide wildlife habitat, and particularly those located at relatively high elevation are the major source of groundwater recharge. Our field site in southern Saskatchewan is located in a typical northern prairie landscape. One of the main hydro logic connections between the recharge wetland and the surrounding farm land is the large snowmelt runoff resulting from low infiltration capacity of the frozen soil. In the spring time, a large volume of water is transferred from the slope to the wetland and infiltrates to form a groundwater mound below the wetland. This, in turn, induces the subsurface flow from the wetland to the slope. Most of the infiltration in the wetland is consumed by evapotranspiration in the surrounding slope; however, a small portion recharges the 25-m deep aquifer. Little net infiltration occurs into the slope because the average flow direction in the vadose zone is upward; therefore, groundwater recharge is focused below the wetland. The large infiltration leaches chloride from the sediments below the wetland. Most of the chloride is transported to the slope by subsurface flow, and accumulates in the root zone during evapotranspiration. The accumulated chloride is leached again by snowmelt runoff and is transported to the wetland via overland flow; therefore, chloride is cycled between the wetland and the slope. A small portion of chloride leached under the wetland is transported down to the aquifer and eventually leaves the catchment. A numerical model that incorporates both surface and subsurface transport processes observed in the field shows that the output of chloride to the aquifer becomes balanced with the atmospheric input on the order of 103 years. The model also shows that the average chloride concentration in the groundwater entering the aquifer is roughly equal to that in shallow groundwater under the wetland. The product of this concentration and groundwater recharge must be equal to the atmospheric input which is monitored at many locations in North America. Therefore, we can estimate groundwater recharge from the concentration of chloride in shallow groundwater under recharge wetlands. This method is best suited to estimate the average recharge over an entire catchment, which is difficult to estimate from sparse measurements of the hydraulic gradient and hydraulic conductivity. The water and mass transfer characteristics of a wetland will be dramatically altered, if an artificial stress is applied to the system. For example, the drainage of the wetland will stop the divergent subsurface flow. As a result, the dissolved species will accumulate in the wetland, and the groundwater recharge will be reduced. Therefore, the impact of drainage and other land-use practices must be carefully evaluated before implementation.
Analysis of analog sampled data circuits
(University of Waterloo, 1997) Raahemi, Bijan
Complex systems consist of a core of digital signal processing (DSP) buffered from external environments by analog interface circuitry. "Analog sampled data" circuits can be predominantly found in the data conversion portions of these mixedmode integrated circuits. Delta-Sigma modulator is an example. Analog sampled data circuits also perform signal processing tasks, and are used in many filtering applications. Switched-capacitor, and switched-current circuits are the examples of these applications. Sampled data circuits are dual time systems that contain a rapidly varying clock and a slowly varying input signal. As a result, most of the simulation effort is dedicated to the transient analysis at switching instants, where information is usually not needed. The important required information is the response of the circuit at the end of the clock period, when it reaches the steady state. Searching for a method that can provide the solution of the circuit at discrete instants of time led to the "sampled data simulation" technique (l]. This thesis looks at different aspects of analog sampled data systems. It extends the idea of sampled data simulation to provide an accurate and efficient method of computing the time domain sensitivity of linear circuits. The method is applied to sensitivity analysis of an important class of sampled data systems, Delta Sigma modulators. We also provide efficient methods for analysis of switched networks, including the group delay and group delay sensitivity of periodically switched linear networks in general, and harmonic distortion of switched-current circuits in particular. Sources of distortion in switched-current circuits are examined, and both upper and lower bounds are derived on total harmonic distortion of a current memory cell.
Enhancing the dissemination of population-based smoking cessation programs, a study of selected communication variables
(University of Waterloo, 1997) McDonald, Paul Wesley
Attempts to reduce the prevalence of smoking through quit smoking program have, to date, been largely unsuccessful. A major reason for this may be that quit smoking programs designed for broad population have not attracted large numbers of smokers to participate in them. A systematic review of the literature was conducted to identify potential communication variables that might enhance recruitment for population based quit smoking programs. Thirty-three publications reporting the results of 40 recruitment campaigns were located. The median recruitment rate across all campaigns was 2.0 per cent of smokers. Logistic regression was used to examine the effect of six variables on recruitment rate: the type of program sponsor, the type of program offered, the cost of the program, presence or absence of an incentive, whether messages were segmented by stage of change, and the type of channel used to send messages. The only significant predictor of recruitment rate was channel type. Studies that used pro-active recruitment strategies (telephone and interpersonal communication) were 66.5 times more effective than those using passive recruitment strategies (mass media, direct mail). An attempt to examine whether segmenting messages by stage of change enhanced recruitment was inconclusive. In a follow-up study, 14,369 smokers and ex-smokers aged 18 and over from Windsor, Canada were randomly assigned to one of seven recruitment conditions to test the effects of using different channel types (mass media, direct mail, telephone) and messages (segmented by stage or change or not) on recruitment rate (% of target population who enrolled in the quit smoking program offered), recruitment efficiency (% of target population who received the messages and enrolled), and cost efficiency (mean cost per enrollee). Messages segmented by stage of change and delivered by telemarketing produced the highest recruitment rate (8.8% of smokers; 7.2% of smokers plus ex-smokers). Messages segmented by stage of change improved the odds of recruitment relative to generic messages for smokers (OR = 1.40; 95% CI: 1.02 - 1.92) and smokers plus ex-smokers (OR = 1.53; 95% CI: 1.16-2.01). Using two channels to deliver messages to smokers (OR = 1.30; 95% CI: 0.99 - 1.72) or smokers plus ex-smokers (OR = 1.15; 95% CI:.90 - 1.46) did not improve recruitment relative to the use of a single channel. The effect of channel was highly significant. For smokers, staged messages delivered by telephone were more successful than those using mass media (OR = 4.72; 95% CI: 2.90 - 7.67) or mail (OR = 4.49; 95% CI: 2.83-7.11). A similar pattern of results emerged with smokers plus ex-smokers. The greatest increase in recruitment rates and recruitment efficiency occurred among smokers in precontemplation and former smokers in the action and maintenance stages. Staged messages by telephone also produced the highest recruitment efficiency rate (26.9%), and were also the most cost efficient recruitment strategy ($32.57 CDN per enrollee). The present results suggest that researchers and practitioners interested in population wide tobacco control should pay more attention to recruitment. Relative to the most common recruitment strategies (i.e. mass media or direct mail), using telemarketing to send messages specifically designed to appeal to smokers across the continuum of change has the potential to increase participation in population based quit smoking programs by more than 400 per cent. Future studies are required to replicate and extend these findings.
Space charge distribution measurements in polymers by an improved pulsed electro-acoustic method
(University of Waterloo, 1997) Liu, Yue
This study presents a novel method of processing measured signals from the Pulsed ElectroAcoustic(PEA) method in order to obtain space charge distributions with improved resolution and accuracy. The proposed method (ie. the simulation model method) enables surface charges to be distinguished from space charge distributions. Therefore, space charge distributions can be processed separately from surface charges without surface charges being mixed into the obtained space charge distributions. In addition, a nonlinear detection and optimization process in the simulation model is proposed as an option for even better charge distribution results.. A brief review of solid breakdown mechanisms and space charge effects illustrates the importance of space charge distribution measurements. A literature review of the available space charge distribution measuring techniques shows the advantages of the PEA method for space charge distribution measurements.. Furthermore, the deconvolution procedure(DCON) and the direct method(DM), the two previous methods used to obtain space charge distributions from the PEA measured signals, are critically reviewed and reanalyzed in order to gain a better appreciation of the proposed simulation model method. The obtained results of four sets of synthetic signals by the DM, the OCON, and the point matching simulation model(PMSM) methods; and the obtained results of two sets of experimental signals by the PMSM method demonstrate the advantages associated with the use of the simulation model method. Also presented are experimental studies designed and conducted to determine the dynamic changes of space charge distributions under HVOC uniform field voltage applications for sheet specimens made of LOPE, HOPE, UHMWPE, PP and PMMA. The experimental results are then summarized and discussed. Moreover, in this thesis, the PEA method principle is further extended to nonuniform fields (ie. needle plane electrode configurations), where space charge effects are considered crucial.
Variation transmission in multi-stage industrial processes
(University of Waterloo, 1997) Agrawal, Rekha
The subject of variation analysis is of interest in manufacturing processes where items are being produced in large quantity and pass through many operations or stages before they are completed. After the final operation, they must meet certain specifications. The issue is to discover how variation in the product characteristics at the final stage of the process can be reduced. With that goal in mind, it is useful to understand how the variation is conveyed through the process. Multivariate normality is assumed as the underlying model for the measured product. Methods are given for analysing variance transmission under this model, both when a general multivariate normal holds, and in a more restricted case, when a first order autoregressive structure is appropriate. Inevitably, there will be measurement error in the data collected on the process. It is shown that this measurement error can severely hinder attempts to characterize the process, and should be incorporated explicitly in an analysis. A naive estimation method is introduced and shown to work well. It may be less expensive, in some instances, to collect large amounts of sample data after each stage, and then track only a few items through the process. Methods are given of incorporating cross-sectional data into the analysis. Also discussed is how to do this when the problem is compounded by measurement error. Finally, some consideration is given to the issue of multivariate data.
A parametric solution for local and global optimization
(University of Waterloo, 1997) Ding, Baoyan
The goal of this thesis is to present a method which when applied to certain nonconvex quadratic programming problems will locate the global minimum, all isolated local minima and some of the non-isolated local minima. The method proceeds by formulating a (multi) parametric QP or LP in terms of the data of the given non-convex quadratic programming problem. Based on the solution of the parametric QP or LP, a minimization problem is formulated. This problem is unconstrained and piece-wise quadratic. A key result is that the isolated local minimizers (including the global minimizer) of the original non-convex problem are in one to one correspondence with those of the derived unconstrained problem. As an application, the method is applied to the problem of determining if a given symmetric matrix is copositive on a given polyhedral cone. We show that the copositivity problem in which the matrix has exactly one negative value can be solved in polynomial time. The results established for non-convex quadratic programming problems are generalized to the non-convex problems in which the objective function in nonquadratic and the constraints are nonlinear.

Browse

Browsing Digitized University of Waterloo Theses by Issue Date