ALGORITHM FOR ENHANCING VOICE QUALITY IN 5G STANDALONE VIA MOSORIENTED DYNAMIC LINK ADAPTATION

DOI: 10.31673/2786-8362.2025.021693

Abstract

This paper proposes a MOS-driven, risk-aware dynamic MCS
adaptation algorithm for VoNR in 5G Standalone networks. The algorithm enhances voice delivery
reliability by combining SNR-uncertainty look-ahead, CVaR-based control of HARQ retransmission tails,
and an external VoNR feedback loop for Δ-offset regulation. Unlike OLLA and HARQ-aware/unaware
baselines, which optimize BLER without delay prediction, the algorithm evaluates candidate MCS values
through analytical BLER curves, IR-combining gain, late-loss probability, and E-model MOS estimation
under mobility conditions. Simulation results using CRN-blocked experiments confirm stable performance
across BLER∈0.01-0.10: MOS≈4.26-4.27, >98.5% on-time RTP delivery, and Retx≥2≈2.0-2.3%, with no
notable goodput degradation. Benchmark schemes (OLLA, Random) generate ≈22–25% deep HARQ
chains and MOS<4.0, demonstrating the advantage of MOS-oriented decision policy over BLER-only
control. The results show that the proposed algorithmic approach provides deterministic QoE under
Doppler-induced CQI drift and is suitable for real-time VoNR deployment in native 5G SA environments.
Keywords: VoNR, 5G Standalone, MCS adaptation, HARQ retransmissions, MOS, QoS/QoE,
BLER, dynamic link adaptation, risk-aware scheduling, real-time voice services

References
1. 3GPP TS 38.300 version 17.2.0 Release 17. NR; NR and NG-RAN Overall Description.
Effective from 2023-05-18. Official edition. FRANCE: 650 Route des Lucioles, F-06921 Sophia
Antipolis Cedex, 2023. 152 p.
2. Ветошко І.П., Кравчук С.О. Пріоритизація голосового трафіку в 5G: роль планувальників
у забезпеченні QoS. ХІХ Міжнародна науково-технічна конференція «Перспективи
телекомунікацій 2025», Київ, 14–18 квітня 2025, с. 189–192.
3. 3GPP TS 38.331 version 17.6.0 Release 17. NR; Radio Resource Control (RRC) Protocol
specification. Effective from 2024-02-11. Official edition. FRANCE: 650 Route des Lucioles, F06921 Sophia Antipolis Cedex, 2024. 421 p.
4. Vetoshko I.P., Kravchuk S.O. Opportunities to Improve the Quality of Voice Services in 5G
Networks // 2023 IEEE International Conference on Information and Telecommunication
Technologies and Radio Electronics (UkrMiCo), ISBN: 979-8-3503-4848-4, 13-18 November 2023,
Kyiv, Ukraine. https://doi.org/10.1109/UkrMiCo61577.2023.10380376.
5. 3GPP TS 23.501 version 16.7.0 Release 16. System architecture for the 5G System (5GS).
Effective from 2021-01-21. FRANCE: Sophia Antipolis Cedex, 2021. 451 p.
6. 3GPP TR 38.802 V14.2.0 Release 14. Study on new radio (NR) access technology. Effective
from 2017-09-28. Official edition. FRANCE: Sophia Antipolis Cedex, 2017. 142 p.
7. Vetoshko I.P., Kravchuk S.O. Possibilities of improving the voice services quality in 5G
networks // Information and Telecommunication Sciences. – 2023. – Vol.14, No 2. – P. 9-16,
https://doi.org/10.20535/2411-2976.22023.9-16
8. Kwon J., Park J., Rhee W. A QoS-aware adaptive resource allocation scheme using CQI and
MCS for LTE-A networks. Wireless Personal Communications. – 2014. – Vol. 76. – P. 193–207.
9. Vora P. Voice over 5G (Vo5G): Technical Challenges and Performance Evaluation. IEEE
Communications Magazine. 2021. Vol. 59, No. 10. P. 16–22.
10. Vetoshko I. P., Kravchuk S. O. Integration of machine learning-based prediction and dynamic
QoS optimization into adaptive VONR scheduling in 5G standalone networks: a simulation-based
approach. Scientific notes of Taurida National V.I. Vernadsky University. Series: Technical Sciences.
2025. Vol. 1, no. 3. P. 43–56. URL: https://doi.org/10.32782/2663-5941/2025.3.1/07