EUVIMEDEUVIMED entdecken
Crossref

A Data-Driven Approach for MCS Prediction in Next-Generation 5G Communication Systems

Dr. Tariq Siddiqui Dr. Tariq Siddiqui, Dr. Ateek Mansoori Dr. Ateek Mansoori

International Journal of Creative and Open Research in Engineering and Management · 25.05.2026 · 02 (05) · 1-9

DOI: 10.55041/ijcope.v2i5.730 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 25.05.2026. Verfasst von Dr. Tariq Siddiqui Dr. Tariq Siddiqui und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In next-generation wireless systems such as 5G, Beyond 5G (B5G), and emerging 6G networks, efficient link adaptation is essential for achieving high spectral efficiency, ultra-reliable low-latency communication (URLLC), and improved Quality of Service (QoS) [1], [2]. Modulation and Coding Scheme (MCS) selection plays a central role in this process; however, conventional rule-based and CQI-driven approaches are limited in handling complex, nonlinear, and rapidly varying channel conditions [3]. To address these limitations, the study proposes an advanced machine learning (ML)-based framework for accurate and adaptive MCS prediction in OFDM-based wireless systems. The proposed framework evaluates multiple ML and deep learning models, including Support Vector Machines (SVM), Random Forest (RF), Gradient Boosting, Artificial Neural Networks (ANN), and an optimized Deep Neural Network (DNN). It further incorporates Long Short-Term Memory (LSTM) networks to capture temporal channel variations and improve prediction accuracy [4], [5]. The models are trained and tested on a large-scale dataset generated from realistic non-standalone 5G simulations, integrating physical layer indicators such as SINR, CQI, and RSSI with environmental features derived from ray-tracing propagation models [6]. Additionally, preprocessing techniques such as normalization, feature selection, and imbalance handling are applied to enhance model generalization and reduce overfitting [7]. Experimental results demonstrate that the hybrid DNN-LSTM model achieves superior performance, with accuracy exceeding 99% and strong robustness across diverse channel scenarios. Ensemble methods like Gradient Boosting also show competitive results, outperforming traditional approaches [8]. Overall, the findings confirm that ML-driven MCS prediction significantly improves link adaptation efficiency, enabling intelligent and proactive decision-making in dynamic wireless environments and supporting the development of AI-driven, self-optimizing networks aligned with future 6G and Zero-Touch Network Management (ZTNM) paradigms [9]. Index Terms - 5G, B5G, 6G, modulation and coding scheme (MCS), machine learning, deep learning, LSTM, OFDM, link adaptation, predictive modeling, intelligent networks, ZTNM.

Zitieren

APA

Siddiqui, D. T. S. D. T., & Mansoori, D. A. M. D. A. (2026). A Data-Driven Approach for MCS Prediction in Next-Generation 5G Communication Systems. International Journal of Creative and Open Research in Engineering and Management, 02(05), 1-9. https://doi.org/10.55041/ijcope.v2i5.730

Vancouver

Siddiqui DTSDT, Mansoori DAMDA. A Data-Driven Approach for MCS Prediction in Next-Generation 5G Communication Systems. International Journal of Creative and Open Research in Engineering and Management. 2026;02(05):1-9. doi:10.55041/ijcope.v2i5.730

BibTeX
@article{Siddiqui2026,
  author = {Siddiqui, Dr. Tariq Siddiqui Dr. Tariq and Mansoori, Dr. Ateek Mansoori Dr. Ateek},
  title = {A Data-Driven Approach for MCS Prediction in Next-Generation 5G Communication Systems},
  journal = {International Journal of Creative and Open Research in Engineering and Management},
  year = {2026},
  volume = {02},
  number = {05},
  pages = {1--9},
  doi = {10.55041/ijcope.v2i5.730}
}
Kostenlos starten →