No abstract

RESUMENLas necesidades detectadas en los estudiantes de nuevo ingreso de las Escuelas Técnicas Superiores de Ingenieros de Telecomunicación han llevado a plantear iniciativas de mentoría formal en sus grados. Estas carencias están relacionadas con una escasa orientación previa a la Universidad, gran heterogeneidad del alumnado, alto índice de fracaso académico en el primer curso y falta de información acerca de la inserción laboral. Este artículo pretende dar a conocer el programa MENTUm cuya finalidad es que estudiantes de últimos cursos ofrezcan apoyo e incentiven el aprendizaje e integración de estudiantes de primero con el objetivo de detectar las dificultades personales y académicas que presentan y desarrollar competencias instrumentales. La metodología de investigación empleada consiste en un Estudio de Caso desde un enfoque mixto, empleando técnicas tanto cuantitativas como cualitativas con la finalidad de alcanzar la mayor comprensión del programa en una de sus cuatro dimensiones: dar respuesta a las necesidades del alumnado de nuevo ingreso. Para la recogida de información, se han empleado tres técnicas de recogida de datos: cuestionarios, observaciones y entrevistas. Los resultados muestran que las dificultades personales tienen que ver con la falta de concentración, planificación y organización del tiempo, mientras que las dificultades académicas tienen que ver más con la falta de conocimientos previos ante los contenidos de las asignaturas. Como conclusiones hemos constatado que los estudiantes son conscientes desde su ingreso a la Universidad de la importancia de desarrollar competencias instrumentales de aprendizaje autónomo y planificación a medida que transcurre el año académico.ABSTRACT The needs detected in the new students of the Higher Technical Schools of Telecommunication Engineers have led to propose initiatives of formal mentoring in their grades. These shortcomings are related to a poor orientation prior to the University, great heterogeneity of the students, high rate of academic failure in the first year and lack of information about the labor insertion. This article describes the MENTUm program, whose purpose is for senior students to support and encourage the learning and integration of first-year students with the aim of detecting the personal and academic difficulties they present and developing some instrumental skills. The research methodology used consists of a Case Study from a mixed approach, using both quantitative and qualitative techniques in order to achieve a greater understanding of the program in one of its four dimensions: respond to the needs of new students. For the collection of information, three data collection techniques have been used: questionnaires, observations and interviews. The results show that personal difficulties have to do with the lack of concentration, planning and organization of time, while academic difficulties have more to do with the lack of prior knowledge regarding the contents of the subjects. As conclusions, we have verified that the students are aware from their beginning at the University of the importance of developing instrumental skills of autonomous learning and planning as the academic year progresses.

Network Slicing (NS) is a key enabler of the 5G network ecosystem due to its potential to provide distinct services over the same physical infrastructure. However, the necessity to optimally orchestrate resources for heterogeneous demands is crucial when dealing with resource constraints and Quality-of-Service (QoS) requirements. We consider a radio access network scenario providing NS over multiple base stations (BS) with limited resources, and we design an efficient resource orchestration technique, based on reinforcement learning, which optimizes resource utilization among different services while satisfying the constraints and complying with Service Level Agreement (SLA) and QoS requirements. The proposed technique makes use of the Trust Region Method to formulate the orchestration objective function and satisfy the constraints and is then optimized via Kronecker Factored Approximate Curvature (K-FAC). Extensive simulations demonstrate that the proposed technique outperforms other Reinforcement Learning (RL) algorithms, reaching 99% of QoS and SLA satisfaction while assuring bandwidth constraints.

Space division multiplexing (SDM) and band division multiplexing (BDM) are considered promising technologies to increase the capacity of optical transport networks. The progressive shortage of available dark fibers and the immaturity of multicore and multimode fibers for multichannel transmission induce network operators to postpone the process of capacity enhancement through SDM. Therefore, capacity increase revolves around BDM by lighting up at least the L-band of the already installed optical fiber infrastructure, which is a practical solution in the short to middle term. However, L-band activation requires the upgrade of network components such as erbium-doped fiber amplifiers (EDFAs). To manage the imposed cost while leveraging the L-band, a network can be partially rather than fully migrated in a single step by upgrading just a subset of the fibers and thus a subset of EDFAs to operate in the C+L-bands. In this paper, the focus is set on determining which fibers in the network should be upgraded to exploit the L-band, subject to a constraint on the maximum number of EDFAs to be upgraded, and analyzing its impact on network performance when facing dynamic traffic in terms of the blocking ratio. To this end, three heuristic algorithms, each pursuing a different objective, and two of them based on an integer linear programming (ILP) formulation, are proposed for the network planning to identify which fibers to upgrade. Simulation results demonstrate that, thanks to the use of these heuristics, the upgrade of a partial set of links to the C+L line system is a viable solution for network operators to circumvent the huge cost associated with migrating the full network. For instance, we demonstrate that a strategic partial upgrade using the proposed methods, subject to upgrading a maximum of 60% of the EDFAs, can significantly boost the supported traffic load in the examined topologies, ranging from 175% to 322%, when compared to the non-upgraded network.

Abstract-An advanced medium access control protocol is presented demonstrating dynamic bandwidth allocation for long-reach gigabit-capable passive optical networks (GPONs). The protocol enables the optical line terminal to overlap the idle time slots in each packet transmission cycle with a virtual polling cycle to increase the effective transmission bandwidth. Contrasting the new scheme with developed algorithms, network modeling has exhibited significant improvement in channel throughput, mean packet delay, and packet loss rate in the presence of class-of-service and service-level differentiation. In particular, the displayed 34% increase in the overall channel throughput and 30 times reduction in mean packet delay for service-level 1 and service-level 2 optical network units (ONUs) at accustomed 50% ONU load constitutes the highest extended-reach GPON performance reported up to date.Index Terms-Class of service (CoS), dynamic bandwidth allocation (DBA), fiber-to-the-home (FTTH), gigabit-capable passive optical network (GPON), long-reach passive optical network (long-reach PON), quality of service (QoS), service level agreement (SLA).