Open Shortest Path First (OSPF) had been recognized as a de facto standard for inter-domain Internet Protocol (IP) routing. Its hierarchical architecture and the introducing of Area Border Router (ABR) had been proved to be an effective solution to the scalability problem. However, manual selection of ABR may fail to accommodate the dynamic change in the network load and therefore lead to sub-optimal solution. There are two new schemes proposed in this thesis. One scheme is Dynamic Appointment ABR for OSPF (DAA-OSPF). Other one scheme is P-median OSPF Partition Algorithm. DAA-OSPF scheme is proposed for the dynamic appointment of ABR for OSPF. Based on observed traffic demands and knowledge on link capacities, the proposed scheme will dynamically switch to a new ABR from an old one to avoid incipient link congestion and accompanying performance degradation. In response to traffic change,the proposed approach will locate an adequate ABR for the new traffic pattern and then direct traffics to the new ABR. System performance, in terms of link utilization, throughput, and delay, is expected to benefit from the proposed scheme, as demonstrated in the simulation results. This study also contributes to the issue of stub area partitioning. According to our study, we recognize that the stub area partitioning problem can be well modeled by the P-median problem in logistics [WR99]. In this study, we extend and refine existing approaches to the P-median problem and apply them to the stub area partitioning problem. The refined P-median algorithm is capable of finding out the optical locations of P ABRs in a network such that the total cost is minimized. With the refined P-median algorithm, the proposed OSPF partitioning scheme is expected to minimize the cost of OSPF hierarchical routing, as well as the cost in subsequent packet transmission.

En esta Tesis Doctoral se propone una solución para el soporte de la clase de servicio ABR que utiliza técnicas de planificación equitativa (fair queueing). La clase de servicio ABR especifica un reparto equitativo del ancho de banda disponible en la red entre las conexiones que la atraviesan. Para efectuar este reparto, la provisión de ABR se basa en un control de flujo por realimentación explícita de tasa. LOS ALGORITMOS de conmutador son los encargados de generar la señal de realimentación para el control de flujo en ABR. En esta Tesis se propone un algoritmo de conmutador original que asume planificación equitativa en los puertos de salida de los conmutadores. La estimación de tasa equitativa que efectúa este algoritmo de conmutador se hace a partir del tiempo de cabeza de cola de las células de cada conexión, el cual es una medida del ancho de banda que asigna el algoritmo de planificación equitativa a cada conexión. LOS ALGORITMOS DE PLANIFICACION EQUITATIVA REALIZAN UN REPARTO DEL ANCHO DE BANDA DISPONIBLE EN EL ENLACE QUE EQUITATIVO EN EL SENTIDO MAX-MIN: De este modo el algoritmo de conmutador propuesto genera una señal de realimentación que resulta también en una distribución del ancho de banda disponible en la red que es equitativo max-min. A partir del algoritmo de conmutador propuesto, se han incorporado dos extensiones que permiten, por un lado, que la asignación de ancho de banda en los puertos (y, por tanto, la distribución del ancho de banda de la red) pueda ponderarse / Guijarro Coloma, LA. (1998). Contribución al soporte de la clase de servicio ABR en redes ATM mediante Técnicas de planificación equitativa [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/4182

With regards to the potential severity of increased antibiotic resistance around the world it is puzzling that the global response to this issue has not been more comprehensive. In this thesis I will examine the political frames on ABR formulated by the global network ReAct in an attempt to understand why this is the case. The frames of an issue, that is how it is described politically in different ways, are crucial for agenda-setting. Moreover, framing is an important part of the work of transnational advocacy networks. Since the acknowledgement of an issue in terms of agenda-setting is an important part of a global response, the frames of transnational advocacy networks make up the focus of this thesis. My findings suggests that the existence of multiple frames on ABR to some extent helps us understand the lacking response to ABR. The construction of the frames in terms of causality, and inparticular a general vagueness in terms of responsibility, is however the main finding.

There is a higher prevalence of hearing loss in children with diagnosed neurological disorders than the general paediatric population. It is therefore essential that these children have their hearing assessed. Conventional behavioural audiometry requires participation from the child, and in a majority of this population with neurological pathology this is not always possible owing to their neurocompromised state. These children will have to undergo objective testing, such as the Auditory Brainstem Response (ABR) in order to obtain estimated hearing thresholds, as this requires no active involvement from the patient. This study therefore aims to describe the audiological ABR findings in order to determine hearing function in this group and to establish a relationship between audiological ABR findings to behavioural audiometry findings where these exist in a group of neurologically disordered children in a tertiary hospital in South Africa. Methods: A retrospective review was conducted on 40 ABR patient records of children between the ages of 5 months and 10 years diagnosed with a neurological disorder. Behavioural audiometry results were then sought for these children, where these existed. Hearing status was described for each child per ear for both objective and behavioural results, and descriptive statistics were conducted. Results: 56.25 % (n=45) of ears in this study presented with normal hearing on ABR testing. No behavioural audiometry results were obtained in 72.5 % (n=58) of ears in this study. Results correlated between ABR and behavioural testing for only 7.5% (n=8) of ears tested and in all eight of these ears the hearing result was within normal hearing limits. Twelve and a half percent (n=10) of ears were misdiagnosed on behavioural testing. More premature infants were able to be tested behaviourally when compared to other pathologies. Cerebral palsy, Down’s Syndrome, prematurity and RVD were the pathologies in which the most hearing losses were diagnosed. Conclusions: Behavioural audiometry appears a largely unreliable method of hearing testing in children diagnosed with neurological disorders as results were obtained in only 27.5 % of the study sample; however it remains the gold standard in paediatric hearing testing in order to evaluate the entire auditory system and provides information on how a child processes sound, unlike ABR testing which only provides hearing information up to the auditory brainstem. This study highlights the high prevalence of hearing problems in children with neurological disorders and therefore the importance of hearing testing in this population. Hearing thresholds should be established for subsequent remediation via objective testing. Conditioning should continue simultaneously for a behavioural audiological test battery with adaptations for the child’s developmental ability.

Satellites are some of the most difficult links to exploit in a Quality of Service (QoS) sensitive network, largely due to their high latency, variable-bandwidth and low-bandwidth nature. Central management of shared links has been shown to provide efficiency gains and enhanced QoS by effectively allocating resources according to reservations and dynamic resource availability. In a modern network, segregated by secure gateways and tunnels such as provided by IPsec, central management appears impossible to implement due to the barriers created between a global Dynamic Bandwidth Resource Allocation (DBRA) system and the mediators controlling the individual flows. This thesis explores and evaluates various through-IPsec communications techniques aimed at providing a satellite-to-network control channel, while maintaining data security for all communications involved.

Protein transduction domains (PTDs), such as TAT from human immunodeficiency virus (HIV) or VP22 from herpes-simplex-virus-1, have been shown to deliver a myriad of molecules, including synthetic small molecules, peptides and proteins in vivo and in vitro. The protein transduction processes mediated by TAT or VP22 are highly efficient and occur in many types of cells with low toxicity. The anti-tumor proteins to be investigated are abrin A chain (ABR-A) and Apoptin. ABR-A is the toxophoric subunit of plant toxin abrin from the seeds of Abrus precatoriusa. ABR-A is a potent inhibitor of translation, but not toxic to cells due to its lack of the cell-binding B chain. Apoptin is a protein derived from chicken anemia virus and has been proved to be selectively cytotoxic to various tumor cells but not to normal cells. The tumor-specific activity of Apoptin is correlated with its nuclear localization in tumor. In this study, we employed VP22 PTDs to promote the entry of natural toxins, such as ABR-A or Apoptin, into tumor cells, thereby to enhance their anti-tumor effects. We generated and characterized green fluorescent protein (GFP)-, hemagglutini (HA)-, and VP22-fused expression constructs for ABR-A and Apoptin, to evaluate the gene delivery effect of ABR-A/Apoptin genes in non-transformed NIH3T3 cells and tumor cells, including Hela and A375 melanoma cells. Gene delivery of ABR-A led to growth inhibition by 50~70% in transformed and non-transformed cells. In contrast, Apoptin gene delivery exhibited cytotoxicity only in tumor cells. The cytotoxicity of ABR-A and Apoptin gene delivery was enhanced when fused with VP-22. Furthermore, the depletion of APAP1 reduced the cytotoxic effect of Apoptin gene delivery. In the future, the anti-tumor effect of these novel PTD-toxin vectors will be explored in cell culture as well as animal model. We hope these studies will open a new avenue for cancer therapy.

No description available.

Objective: The appropriate application of audiological measures during paediatric hearing aid (HA) fitting ensures the fitting is effective and provides speech audibility across the frequency range. Audiological assessment may include both behavioural and objective measures, such as auditory brainstem response (ABR) and auditory steady-state response (ASSR). ABR and ASSR measures however do not have a 1:1 correlation with behavioural measures, and correction values need to be applied to estimate behavioural thresholds prior to HA fitting. No study has previously described how South African audiologists are utilising ABR and ASSR results during paediatric HA fitting. This study aimed to describe the current South African audiological clinical practice for paediatric HA fitting, with specific reference to the application of ABR and ASSR measures. Design: The study employed a quantitative, non-experimental, descriptive, cross-sectional survey research design. Study sample: Thirty-four personal interviews with audiologists were completed, seven within the private health sector and 27 within the public health sector. Results: Results indicated that limited departmental protocols exist and adherence to available protocols was questioned. There was a lack of consensus regarding the application of correction values to ABR and ASSR measures for HA fitting and the values utilised often differed significantly from recommended guidelines. There appeared to be an over-reliance on electrophysiological measures for paediatric audiological assessment, as well as a lack of adherence to recommended age-appropriate assessment guidelines. Conclusion: Findings suggest the need for promoting improved clinical practice and knowledge within the area of paediatric audiology in South Africa. The need for the development of nationally-agreed guidelines was highlighted.

Auditory brainstem response (ABR) evaluation has been one of the most reliable methods for evaluating hearing loss. Clinically available methods for ABR tests require averaging for a large number of sweeps (~1000-2000) in order to obtain a meaningful ABR signal, which is time consuming.  This study proposes a faster new method for ABR filtering based on wavelet-Kalman filter that is able to produce a meaningful ABR signal with less than 500 sweeps. The method is validated against ABR data acquired from 7 normal hearing subjects with different stimulus intensity levels, the lowest being 30 dB NHL. The proposed method was able to filter and produce a readable ABR signal using 400 sweeps; other ABR signal criteria were also presented to validate the performance of the proposed method.

No description available.