AN ADAPTIVE LOCAL LINEAR REGRESSION METHOD FOR MOBILE SIGNAL STRENGTH WITH APPLICATION TO RESPONSE SURFACE METHODOLOGY

  • Itua O. AKHIDENO Benson Idahosa University, Nigeria
  • O. Eguasa
Keywords: Circumscribed central composite design, Local linear regression model, Parametric regression model, Received signal strength, Response surface methodology

Abstract

The received signal strength is vital in telecommunication network or technology as it is affected by varying environmental factors such as temperature (T0C), relative humidity (H%), air quality index (m) and distance from base station (m). In this paper, we seek to find a regression model via a circumscribed central composite design that can adequately represent the functional relationship between the received signal strength and the respective factors applied to response surface methodology with the goal to obtain settings of these factors that would simultaneously optimize the received signal strength (Long Term Evolution (LTE), Third Generation network (3G) and second Generation network (2G)) technologies. The frequently utilized regression model is the parametric regression model (second-order model), though superior but lack credibility in terms of model misspecification and as a result, the optimum setting of the factors are miscalculated. In addressing the pitfall of the parametric regression model (PRM), we introduce a flexible adaptive local linear regression model that can capture local trend in the data, which ordinarily a misspecified PRM could not address. In the application, two regression models were used and the results show that the adaptive local linear regression model outperformed the parametric counterpart in terms of goodness-of-fit statistics, residual plot and optimization of the received signal strength.

References

Anderson-Cook C. M., and Prewitt K. (2005). Some guidelines for using nonparametric models for modeling data from response surface designs. Journal of Modern Applied Statistical Models, 4, 106-119.

Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. and Escaleira, L. A. (2008). Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta, 76, 965-977.

Castillo D. E. (2007). Process Optimization: A Statistical Method. New York: Springer International Series in Operations Research and Management Science.

Choudhary, S., Sharma, A. and Srivastava, K. (2021). Modeling and Optimization of Mobile Signal Strength in Challenging Atmospheric Conditions. Research Square, DOI:https://doi.org/10.21203/rs.3.rs-295295/01
Published
2022-11-02
How to Cite
AKHIDENOI. O., & EguasaO. (2022). AN ADAPTIVE LOCAL LINEAR REGRESSION METHOD FOR MOBILE SIGNAL STRENGTH WITH APPLICATION TO RESPONSE SURFACE METHODOLOGY. FUDMA JOURNAL OF SCIENCES, 6(5), 41 - 49. https://doi.org/10.33003/fjs-2022-0605-1095