Archipelagic regions often experience limited telecommunications infrastructure, resulting in unequal Internet access and persistent digital connectivity challenges. Satellite Internet has emerged as a potential solution for expanding broadband access in geographically isolated areas. This study explores user adaptation to satellite Internet technology in the Riau Islands Province, Indonesia. A qualitative case study approach was employed using in-depth interviews, field observations, and documentation involving four active Starlink users selected through purposive sampling. The findings indicate that participants adopted satellite Internet because conventional terrestrial networks could not provide reliable connectivity. Following adoption, users gradually integrated the technology into their daily activities while adapting to weather-related disruptions, antenna positioning, and relatively high subscription costs. Four interrelated dimensions of post-adoption adaptation emerged from the analysis: technical, social, economic, and functional adaptation. The study contributes by proposing an exploratory user-centered framework that explains post-adoption adaptation to satellite Internet in an archipelagic context and extends the Diffusion of Innovation perspective beyond initial adoption. The findings also provide practical insights for service providers and policymakers seeking to improve digital connectivity and digital inclusion in geographically underserved regions.
Host-based intrusion detection systems (HIDS) and endpoint detection and response platforms increasingly rely on high-volume host telemetry such as system-call and API-call sequences. Accurate trace classification is useful, but operational response also requires localization of the suspicious fragment and a readable evidence narrative. This paper presents an analyst-oriented HIDS workflow with three stages: window-based detection using 3-gram logistic regression, 1D CNN, BiLSTM with attention pooling, and Transformer encoder models; suspicious-window localization using maximum-probability window selection and attention attribution; and evidence-grounded forensic narrative generation from the same localized evidence. The main ADFA-LD evaluation uses 1,263 training traces and 316 test traces; the best detector, 3-gram logistic regression with W=100, achieves precision 0.958, recall 0.913, F1 0.935, and AUROC 0.980. To address dataset-size and cross-telemetry concerns, this study adds two bounded additional evaluations: an eight-family Windows API-call sequence experiment on Mal-API-2019 and a large NetFlow stress test on NF-ToN-IoT-v3 with 10,000, 20,000, 50,000, and 100,000 training records. The NetFlow experiment shows that deep models become more stable as data volume increases, but the linear baseline remains strongest at 100,000 records (F1=0.888), while the Transformer improves to F1=0.856. These results support a careful conclusion: larger datasets help deep models, but they do not automatically make a Transformer outperform a strong shallow baseline when the signal is dominated by compact local motifs or well-separated tabular flow features.
Unmanned aerial vehicles (UAVs) and advanced air mobility (AAM) offer potential alternatives for improving cargo and passenger transport in regions where conventional transport is constrained by distance, terrain, and fragmented geography. This study systematically reviews the use of UAV and AAM systems for connectivity in remote, rural, and inter-island settings. Following the PRISMA 2020 guidelines, searches were conducted in Scopus and Google Scholar through Publish or Perish. Of the 3,561 records identified, 74 peer-reviewed journal articles published between 2016 and 2026 met the eligibility criteria. The evidence was synthesised narratively and supported by descriptive frequency analysis and bibliometric mapping using VOSviewer. Unlike earlier reviews that primarily examine urban air mobility, last-mile delivery, or medical-drone applications, this review jointly evaluates cargo and passenger transport, supporting infrastructure, and operational challenges from the perspective of remote and archipelagic connectivity. Cargo and goods transport accounts for 92% of the included studies, whereas passenger transport represents only approximately 1%. Multirotor platforms are the most frequently identified aircraft type, while weather, regulatory and airspace constraints, and limited supporting infrastructure are the leading operational challenges. The findings show that passenger-oriented and inter-island applications remain substantially underexplored and require field-based validation, particularly in archipelagic regions.
Private LTE Band 3 Frequency Division Duplex (FDD) systems require dedicated uplink and downlink antennas to improve channel isolation and communication reliability. However, experimentally validated split-frequency Yagi-Uda antennas for Software Defined Radio (SDR)-based base stations remain limited. This study presents the design, simulation, fabrication, and experimental validation of two four-element Yagi-Uda antennas operating at 1.840 GHz for downlink (DL) and 1.750 GHz for uplink (UL). Each antenna consists of one reflector, one driven element, and two directors fabricated from 6 mm diameter aluminum rods. The antenna dimensions were optimized using the Trust Region Framework in CST Studio Suite 2024 to achieve optimal impedance matching. Simulation results produced S11 values of −35.33 dB (DL) and −30.76 dB (UL), VSWR values of 1.035 and 1.059, and forward gains of 9.10 dBi and 8.38 dBi, respectively. Measurements using a NanoVNA over the 1.4–2.1 GHz frequency range yielded S11 values of −18.98 dB (DL) and −26.92 dB (UL), with corresponding VSWR values of 1.253 and 1.094. The measured resonant frequencies closely agreed with the simulation results, with a maximum deviation of only 1 MHz. The novelty of this work lies in the experimental validation of a low-cost, split-frequency, four-element aluminum-rod Yagi-Uda antenna pair that operates without an external impedance-matching network. These results demonstrate the suitability of the proposed antennas for SDR-based Private LTE Band 3 base station applications.
The growing demand for high-efficiency power conversion has accelerated adoption of wide-bandgap (WBG) devices in DC-DC converters. While SiC and GaN MOSFETs outperform conventional silicon, direct quantitative comparisons under identical buck-converter conditions remain an area of active investigation. This simulation-based study investigates power losses and estimated efficiency of SiC C3M0075120D (1200 V, 75 m ) and GaN GS66508B (650 V, 50 m ) in a 200 V-to-100 V, 50 kHz DC-DC buck converter under identical loads (5 A and 10 A). Using LTspice simulations with manufacturer-supplied models and Double Pulse Test methodology for loss separation, the study explicitly distinguishes switching and conduction losses under fixed operating conditions, providing validated simulation-based quantification for these specific devices. The results indicate that for the GaN device achieves lower modeled total losses than the SiC device at 10 A (6.85 W vs. 17.05 W) and attains higher simulated estimated converter efficiency (99.31% vs. 98.25% at full load). For the two specific devices and simulated operating conditions examined, switching loss remains the dominant loss mechanism at 50 kHz. A frequency sensitivity analysis from 50 to 500 kHz reveals that the GaN device maintains approximately 60% lower losses across the frequency range, with absolute savings increasing from 10.2 W to 91.2 W. These findings apply specifically to the investigated devices and operating conditions, and the comparison involves devices with different voltage ratings. The methodology demonstrates the utility of LTspice as a pre-prototyping tool for WBG converter development, although experimental validation remains necessary to confirm estimated efficiency figures.
