@article{4809,
  author = {Nguyen Minh Tuan},
  title = {Multidimensional Evaluation of Telecommunication Network Quality: Construction and Empirical Analysis of the Total Network Quality Index (TNQI)},
  journal = {Progress in Signals and Telecommunication Engineering},
  year = {2026},
  volume = {15},
  number = {2},
  doi = {https://doi.org/10.6025/pste/2026/15/2/88-113},
  url = {https://www.dline.info/pste/fulltext/v15n2/pstev15n2_3.pdf},
  abstract = {Modern telecommunication networks are characterised by multidimensional, dynamically varying
performance conditions that no single key performance indicator (KPI) can adequately capture. This study
develops and empirically validates the Telecommunication Network Quality Index (TNQI), a composite
measure that integrates three heterogeneous network performance indicators signal strength (dBm), latency
(ms), and packet loss (%) into a unified 0-100 quality scale. Using a dataset of 3,000 temporally ordered
network observations collected between 1 January 2023 and 5 May 2023 across five network technologies
(4G LTE, 5G NR, Fibre Optic, Microwave, and Satellite), the study employs directional min-max normalisation,
equal weight aggregation, Pearson and Spearman correlation analysis, weighted reliability scoring, and
quartile-based classification. The correlation analysis reveals near zero associations among the three KPIs
(|r|ï‚£ 0.019), confirming that signal quality, network responsiveness, and packet delivery reliability represent
statistically independent dimensions of communication performance. The overall mean TNQI was 50.39 (SD
= 16.66; range: 2.12-94.41), indicating moderate average communication quality with substantial
heterogeneity. Network-type comparisons yielded relatively modest differences, with Satellite recording
the highest mean TNQI (52.19) and 5G NR the lowest (49.34). A complementary weighted network reliability
score (30% signal, 35% latency, 35% packet loss) and an inverse network degradation index (NDI = 100 -
TNQI) provide operational monitoring perspectives. Quartile-based classification partitions observations
into Critical, Poor, Good, and Excellent states. The findings demonstrate that multidimensional integration
through TNQI provides a more comprehensive and interpretable representation of network quality than any
individual KPI, establishing a reproducible analytical foundation for network monitoring, technology
comparison, temporal degradation tracking, and future early warning applications.},
}