Int. J. Advanced Networking and Applications Volume: 15 Issue: 03 Pages: 5931– 5939 (2023) ISSN: 0975-0290 5931 Hybrid Empirical and Machine Learning Approach to Efficient Path Loss Predictive Modelling: An Overview Ituabhor Odesanya * Federal University Lokoja/Department of Physics, Lokoja, Nigeria; Email: ituabhor.odesanya@ fulokoja.edu.ng Orcid ID: https://orcid.org/0000-0002-5901-3370 Joseph Isabona Department of Physics, Federal University Lokoja, Lokoja, Kogi State, Nigeria Email: joseph.isabona@fulokoja.edu.ng Orcid ID: https://orcid.org/0000-0002-2606-4315 Emughedi Oghu Department of Computer Science, Federal University Lokoja, Nigeria Email: emughedi.oghu-pg@fulokoja.edu.ng Okiemute Roberts Omasheye Department of Physics, Delta State College of Education, Mosogar 331101, Nigeria Email: okiemuteomasheye@yahoo.com -----------------------------------------------------------------ABSTRACT--------------------------------------------------------------- In the field of wireless communication and network planning, accurate path loss predictive modelling plays a vital role in understanding the behavior of signal propagation in diverse environments. Traditional empirical models have been widely used for path loss estimation, but they often lack the flexibility to adapt to complex scenarios. On the other hand, machine learning techniques have shown great potential in various domains, including wireless communication. This paper aims to present a hybrid empirical and machine learning approach for efficient path loss predictive modelling. By combining the strengths of empirical models and machine learning algorithms, we can enhance the accuracy and adaptability of path loss predictions. The following sections provide an overview of path loss modelling, explore traditional empirical techniques, discuss the application of machine learning approaches, and outline the methodology for the hybrid approach, along with evaluation and analysis. Finally, we conclude with a summary of findings and suggest future directions for research in this field. Keywords: Network planning, Accurate predictive modelling, Signal propagation, Empirical models, Machine learning models ------------------------------------------------------------------------------------------------------------------------------------------------- Date of Submission: October 16, 2023 Date of Acceptance: October 27, 2023 -------------------------------------------------------------------------------------------------------------------------------------------------- 1. Introduction Path loss predictive modelling is a crucial tool in wireless communication systems to estimate the attenuation of the signal as it propagates through the environment. Accurate path loss models are essential for optimizing network performance and planning efficient wireless networks [1-10]. As the demand for wireless communication continues to grow, there is a need for more accurate and efficient path loss predictive models. Traditional empirical models have limitations and may not capture the complex and dynamic nature of real-world environments [11-20]. This has led to the emergence of machine learning techniques that can provide more accurate predictions. Accurate path loss predictive modelling is crucial in wireless communication as it helps in optimizing network planning, coverage prediction, and resource allocation. It allows network engineers to understand signal propagation characteristics, estimate signal strength, and anticipate coverage limitations in different environments [21-30] The objective of this paper is to provide an overview of a hybrid approach that combines both empirical and machine learning techniques for path loss predictive modelling. By leveraging the strengths of both approaches, we can achieve more accurate and efficient predictions. This article will explore traditional empirical models, the limitations they face, and how machine learning algorithms can overcome these limitations.