Resumen

Abstract

This paper presents a comprehensive bibliometric and systematic review of fog computing research from its inception through 2023. By analyzing over 3,000 publications indexed in Scopus and Web of Science, the study maps the evolution of the field, identifies prolific authors, leading institutions, and dominant research clusters including IoT integration, latency optimization, security, and resource management. The analysis reveals an exponential growth in fog computing publications post-2016, with strong contributions from China, USA, and Europe, highlighting the technology's maturation and convergence with edge computing paradigms.

ACM Computing Surveys · DOI: 10.1145/3702313

Resumen

Abstract

This work performs a rigorous bibliometric analysis of deep-learning research applied to COVID-19 during 2020. The study evaluates over 500 peer-reviewed articles covering applications such as chest X-ray classification, CT scan analysis, epidemiological prediction, and drug discovery. Results show rapid proliferation of CNN-based models for diagnostic support, with significant variance in reported accuracy metrics, highlighting the need for standardized benchmarks and clinical validation frameworks for AI-assisted healthcare tools.

IEEE Access · DOI: 10.1109/ACCESS.2022.3159025

Abstract

Chapter presenting IoT system architectures for precision agriculture and poultry farm management. The proposed framework integrates low-cost sensor networks (temperature, humidity, gas, weight), edge computing nodes, and cloud dashboards to automate environmental control, feeding systems, and anomaly detection. Case studies demonstrate 23% reduction in feed waste and 18% improvement in flock survival rates through automated intervention systems guided by real-time analytics.

Springer Book · DOI: 10.1007/978-3-030-75123-4_17

Abstract

This paper proposes and evaluates novel ROADM architectures specifically designed for few-mode multicore fiber (FM-MCF) based SDM networks. The designs support flexible spatial group configurations enabling scalable node architectures. A unified QoT estimator integrating mode-dependent loss, inter-core crosstalk, and nonlinear effects is derived and validated against Monte Carlo simulations, showing excellent agreement and enabling accurate network planning for next-generation ultra-high capacity optical transport networks.

Journal of Lightwave Technology · DOI: 10.1109/JLT.2020.3009413

Abstract

This work introduces a gradual migration framework enabling network operators to evolve their optical infrastructure from conventional elastic optical networks toward spectrally-spatially flexible SDM architectures. The proposed planning methodology accounts for incremental fiber and node upgrades, capital expenditure constraints, and traffic demand growth, providing actionable migration roadmaps validated on realistic European and North American reference networks with significant capacity gains at controlled investment levels.

IEEE Communications Magazine · DOI: 10.1109/MCOM.001.1900207

Addresses the joint RMSA problem (Route, Modulation, MIMO, Spectrum Assignment) for Flex-Grid multi-core fiber networks. Integer linear programming formulations and heuristic algorithms are proposed and compared, showing that simultaneous optimization across all dimensions reduces spectrum consumption by up to 30% compared to sequential assignment strategies on typical core network topologies.

JLT · DOI: 10.1109/JLT.2018.2848199

Comprehensive scalability analysis of dynamic Flex-Grid/SDM optical networks as traffic demand grows. Simulation studies across multiple fiber and network configurations reveal that SDM networks with 7-core and 12-core fibers achieve near-linear capacity scaling with manageable complexity overhead, making them viable candidates for next decade's traffic projections.

Computer Networks · DOI: 10.1016/j.comnet.2018.06.009

This paper evaluates how the space continuity constraint (SCC) — requiring the same spatial channel across all links of a connection — affects network performance in dynamic SDM networks. Results show that relaxing the SCC through spatial switching capability can reduce blocking probability by over 40%, though at the cost of increased ROADM complexity.

Computer Communications · DOI: 10.1016/j.comcom.2018.05.013

Presents cost-aware resource allocation algorithms for spatial superchannels in MCF Flex-Grid networks. By jointly minimizing transponder count and core switching operations, the proposed heuristics achieve cost reductions of 15–25% versus baseline strategies, with only marginal increases in blocking probability, enabling more economical deployment of SDM optical infrastructure.

Optical Switching and Networking · DOI: 10.1016/j.osn.2017.10.001

Provides a detailed evaluation of ROADM architectures that enforce core-continuity constraints in Flex-Grid MCF optical networks. Different switching fabrics and waveband grouping strategies are analyzed, revealing that partial core switching support can achieve up to 80% of the performance of full spatial switching at substantially lower implementation cost.

JOCN · DOI: 10.1364/jocn.9.001041

Proposes a dynamic buffer management scheme tailored for wireless mesh networks carrying non-elastic traffic (voice, video conferencing, real-time control). The algorithm adapts buffer sizes at each mesh node based on traffic load and queue occupancy measurements, reducing packet loss by 60% compared to static buffer configurations while maintaining bounded delay for latency-sensitive applications.

Revista Politécnica EPN · ISSN 1390-0129