Elisangela Aguiar's research while affiliated with Federal University of Pará and other places

... Objective quality assessment metrics are useful in many image and computer vision applications such as denoising, restoration, printing, archiving. Objec- 20 tive quality metrics are critical for designing imaging systems such as digital cameras [11], benchmarking image and video processing algorithms [5] [12][13][14][15][16], and monitoring and optimizing transmission systems [17][18][19][20][21]. For various video streaming services such as IPTV and video on demand, VQA also provides the cost reduction benefits for both the service provider and the consumer [14] [22][23][24]. ...

... Nevertheless, this approach is not applied in wireless mesh networks. Fortunately, another tool called Hybrid Quality of Experience (HyQoE) can predict for real-time video streaming applications [140]. It takes into account six parameters percents losses in I frame, P frame and B frame, general loss, complexity, and motion. ...

... The performed experiments revealed that HyQoE produces scores equal to the subjective assessment (MOS) in 70% of the scenarios, compared with a PSQA figure of 24.11%. An almost identical approach is proposed by Cerqueira et al. [84] and used by Rosário et al. [85], using multiple artificial neural networks (MANN) instead of RNN. In MANN (also known as modular artificial neural networks), one uses a series of independent neural networks, with each neural network serving as a module and operating on a separate subtask. ...

... Some QoS parameters, such as the capacity of assuring service for some traffic types, by means of some available technologies are defined in Abelém et al. (2007). The main characteristics of QoS in a WMN are listed, which are adopted as metrics to measure how good the service is provided by the network. ...

... Based on the QoE, they also introduce video applications with some future perspectives. The perception of videos related to QoE, is measured plainly by the opinion of users, and differs widely from the principle of QoS [64]. In this context, QoS metrics fail to catch subjective aspects of video content related to human visual system as expected for humancentric environments. ...

... QoE takes into consideration the end-user subjectivity, which depends on QoS and other factors; consequently, subjective and objective quality assessment methods are needed to model the impact of both technical and non-technical factors, as analysed in [33]. Several works [34][35][36] faced with the definition of mapping QoS onto QoE. In [35,36], the authors discuss learning approaches for both online and offline mapping. ...

... While SCOR can be used to model variety of QoS routing algorithms in SDN, the selected routing algorithms were modelled to illustrate the expressive power and completeness of SCOR. A small set of these models were selected to evaluate various aspects of SCOR in- Widest Path [10] Chapter 7 3 Bandwidth-Guaranteed [16] Chapter 6 4 Bandwidth-Constrained [10] Chapter 6 5 Minimum-Loss Path [17] Chapter 6 6 Minimum-Fixed Delay Path [17] Appendix-A2 7 ...

... Nele existem agentes usados para coletar informações da rede e enviar os dados para os repositórios. Vale ressaltar que esse algoritmo é dependente do protocolo de roteamento Ad-hoc On-Demand Distance Vector (AODV) para sua operação, ou seja, é impossível o uso deste framework em uma rede que use OLSR [Aguiar et al. 2007] ou outro protocolo de roteamento, reduzindo desta forma a flexibilidade do sistema. Jardosh [Jardosh et al 2008] projetou o SCUBA, um framework para visualização interativa de problemas em WMNs de grande escala. ...

... Many network management scenarios rely on an efficient and precise measurement of the delay on a link, called Link Delay Measurement (LDM) in the following. Some examples are managing quality-of-service, detecting security attacks, and supporting many other time-sensitive applications, as described, among others, by Cordeiro et al. [1] which uses LDM to provide quality-of-service for Mesh Networks, or by Al Sadi et al. [2], in which LDMs are used together with asymmetric flow measurement in order to detect a possible Denial-of-Service attack, or by Giridhar et al. [3], in which LDM is used to perform search efficiently in a clustered network. In newer terms, 5G aims at providing network slices tailored to specific vertical applications, among which the Ultra Reliable Low Latency for time sensitive services [4]. ...