We propose a novel call admission control which makes use of ratio-based traffic measurement to estimate the required bandwidth when a new call is issued. Existing approaches fail to estimate properly the required bandwidth. To alleviate the problem, we calculate the ratio between the measured mean rate and the mean rate declared by UPC parameters. The ratio and the target cell loss rate are used to estimate the required bandwidth to make decision if a new call is accepted or rejected. Because of more accurate estimation of required bandwidth, our method can provide a better control on quality of service.

ATM (Asynchronous Transfer Mode) networks were deemed the best choice for multimedia communication. The traditional mode was replaced because ATM can provide varied traffic types and QoS (quality of service). Maintaining QoS, however, requires a flexible traffic control, including call admission control and congestion control. Traditional approaches fail to estimate the required bandwidth and cell loss rate precisely. To alleviate these problems, we employ AI methods to improve the capability of estimated bandwidth and predicted cell loss rate. This thesis aims to apply neural network techniques to ATM traffic control and consists of two parts. The first part concerns a neural-based call admission control, while the second part presents an intelligent congestion control for ATM networks. In the first part, we focus on the improvement of RBF (Radial basis function) networks and the design of a neural-based call admission control. RBF networks have been widely used for modeling a function from given input-output patterns. However, two difficulties are encountered with traditional RBF networks. One is that the initial configuration of a RBF network needs to be determined by a trial-and-error method. The other is that the performance suffers from some difficulties when the desired output has abrupt changes or constant values in certain intervals. We propose a novel approach to overcome these difficulties. New kernel functions are used for hidden nodes, and the number of nodes is determined automatically by an ART-like algorithm. Parameters and weights are initialized appropriately, and then tuned and adjusted by the gradient descent method to improve the performance of the network. Then, we employ ART-RBF networks to design and implement a call admission control. Traditional approaches fail to estimate appropriately the required bandwidth, leading to a waste of bandwidth or a high cell loss rate. To alleviate the problem, we employ ART-RBF networks to estimate the required bandwidth, and thus a new connection request can then be accepted or rejected. Because of the more accurate estimation on the required bandwidth, the proposed method can provide a better control on quality of service for ATM networks. In the second part, we propose a neural-fuzzy rate-based feedback congestion control for ATM networks. Traditional methods perform congestion control by monitoring the queue length. The source rate is decreased by a fixed rate when the queue length is greater than a predefined threshold. However, it is difficult to get a suitable rate according to the degree of traffic congestion. We employ a neural-fuzzy mechanism to control the source rate. Through learning, cell loss can be predicted from the current value and the derivative of the queue length. Then an explicit rate is calculated and the source rate is controlled appropriately. In summary, we have proposed improvements on architecture and performance of neural networks, and applied neural networks to traffic control for ATM networks. We have developed some control mechanisms which, through simulations, have been shown to be more effective than traditional methods.

Thesis (M. Soc. Sc.)--University of Hong Kong, 1984.

New technologies are expected to play a major role for wireless cellular systems beyond the existing 4G paradigm. The need for several orders of magnitude increase in system capacity has led to the proliferation of low-powered cellular layers overlaid on the existing macrocell layer. This type of network consisting of different cellular layers, each with their unique characteristics including transmission power and frequency of operation among others is termed as a heterogeneous network (HetNet). The emergence of HetNets leads to several research challenges and calls for a profound rethinking of several existing approaches for mobility management and interference management among other issues.

This research study investigated the relationship to the pre-enrollment factors of admissions-to-enrollment and orientation-to-enrollment timespans to transfer student success as measured by persistence and the length of time taken to earn a baccalaureate degree. This quantitative study analyzed secondary data (N = 357) from a large, four-year, public research institution in the southeast United States. A logistic regression analysis was used to explore the relationships between the pre-enrollment timespans and persistence. The relationship between the admissions-to-enrollment timespan and persistence was not statistically significant. The orientation-to-enrollment timespan was found to have a statistically significant relationship to persistence (p < .05). This indicated that students who had increased orientation-to-enrollment timespans were more likely to persist. To further explore this relationship, a multiple logistic regression analysis was conducted to control for possible extraneous demographic, pre-enrollment, and enrollment variables. The relationship of orientation-to-enrollment timespan and persistence continued to be statistically significant. An ordered logistic technique was used to explore the relationship between the admissions- and orientation-to-enrollment timespans and time-to-degree completion. Neither timespan was found to have a significant relationship with time elapsed to complete the degree. Implications for admissions and orientation timespans were discussed in relation to transfer student transitions.

This dissertation examines trends in access for Black students at public institutions in Texas, specifically, the University of Texas at Austin (UT-Austin), in the context of automatic admission laws and race-based admission policies. Thus, the current study explores how Black students matriculate through the educational pipeline from high school graduation through college enrollment. For this study, I utilize Critical Race Theory as the framework to analyze data for Texas high school students from 2000 through 2010. I employed a quantitative methodological approach, which includes both descriptive analyses and logistic regression. Black people are facing seemingly bleak educational outcomes throughout the education pipeline -- they are less likely to graduate from high school, be college ready, apply to a competitive four-year institution, be admitted, enroll, persist and graduate. Automatic admission laws and affirmative action have been implemented to address these issues, especially for selective institutions. Increased access to Texas' most selective public institution, UT-Austin, may lead to higher persistence and graduation rates for Black students compared to other state colleges and universities. However, Black students remain the most underrepresented group at UT-Austin in spite of these laws and policies. This study is unique because no other research has examined how automatic admission laws and affirmative action operate simultaneously. Moreover, this study fills in some significant gaps in the literature as it relates to Black students in higher education. Findings from this study suggest that while there is progress toward access for Black students at UT-Austin, there is still room for growth. Despite perceptions that Black students do not want to attend UT-Austin, this study found that Black students were more likely to apply to the University compared to White students when holding other factors constant. However, the research findings indicate that White students still have an admissions advantage over Black students even when accounting for SAT scores, family background and the type of high school a student attended. Furthermore, Black students are less likely to enroll at UT-Austin compared to their White counterparts. Finally, the results from this study also suggest that affirmative action has not had the same impact for Black students under the top 10% plan as it has when the policy was used by itself. / text

published_or_final_version / Economics and Finance / Doctoral / Doctor of Philosophy

This study examines the differential patterns of access to higher education of students from rural areas in transition from a planned to a market economy. In respect to college access, the research argues that market reforms have reproduced the advantages for students from the cadre’s and the professional’s families while simultaneously creating new opportunities for the children of the new arising economic elite. Yet, it has performed less for traditional peasant families whose children still fail to gain access to college in proportions higher than the size of the population. Based on the literature, this research places a special emphasis on how economic and cultural resources become the main influence on rural students? college access. The process dimension -- how families from different social backgrounds within rural society involve themselves in the schooling of their children and how this contributes to inequality of college access within rural society, are investigated. This research unpacks this process by examining the school involvement experiences of parents in Zong, a county located in the province of Anhui. Parental involvement is conceptualized in terms of how economic and cultural resources are converted to social capital as part of family strategies within the increasingly stratified social context of rural China. The research identifies the consequences of activating different types of social networks within family and community, and also between family and school to facilitate this process by gaining advantages in access to college. Household interviews and field notes were used as the main methods of data collection with a range of parents and teachers involved in this ethnographic study. The data analysis suggests that state, schools and teachers provide few formal and routine channels for rural parents to become involved in schooling. This raises the importance of family strategic initiatives to employ interpersonal social networks (guanxi) within family, community and between school and family. Parents from cadres and professional backgrounds are capable of maintaining these social networks that are useful for their children’s chances of entering higher education. Their counterparts from the new economic elites? backgrounds have developed the means to capitalize upon their families economic and cultural resources by converting them into social capital that creates advantages in college access for their children. Peasants, however, rely heavily on teachers and relatives in education and are substantially marginalized from those important interpersonal social networks of capital conversion. Although this research found the structure constrains interpersonal social network of peasant families, it also highlights the agency of parents from different families. For example, in some cases it found, that peasants actively use their kinships to create chances for school involvement to potentially improve the chances of their children’s college access. This research is one of the first empirical studies to inquire about the mechanism of capital conversion in affecting higher education opportunities in the post-socialist era, which will help to re-evaluate the influence of market reforms over rural education system in China. / published_or_final_version / Education / Doctoral / Doctor of Philosophy

Not available / text

published_or_final_version / abstract / toc / Education / Doctoral / Doctor of Philosophy