Spectrum Occupancy Measurement of Cellular Spectrum and Smart Network Sharing in Pakistan

Abstract

In wireless communication, the radio spectrum is a very rare and precious resource that has currently become a major problem to efficiently exploit the underutilized band of the static allocated licensed band. Recently, the cognitive radio (CR) has emerged as a promising technology to overcome the spectrum crisis, in which, the licensed band can be utilized by the unlicensed user until and unless it does not affect the transmission of the licensed band. In this paper, the spectrum occupancy of three bands i.e. GSM 900, 1800 and 2100 bands have been measured through spectrum analyzer in the indoor and outdoor environment. The measured results of all the three bands have been calculated through MATLAB against the power spectral density versus frequency plots. Results have shown that the majority of the licensed band is underutilized. Therefore, the CR can play a pivotal role to efficiently utilize the unused spectrum and to overcome the cellular wireless spectrum crisis in Pakistan. The second part of this paper deals with the emerging concept of network sharing among mobile operators and its impacts on cost. Network sharing become the standard among mobile operators worldwide and so in Pakistan. Capital (CAPEX) as well as operational (OPEX) expenditure and rapid advancement in technology encouraged all operators to go for sharing business models. In Pakistan, all four mobile operators Jazz, Telenor, Zong, and Ufone are actively adopting this model to maintain EBITDA (earnings before interest, taxes, depreciation, and amortization). Mainly there are two types of network sharing, Passive infrastructure sharing, and active resource sharing. Passive network sharing is widely used in Pakistan among operators.

Machine Learning Based Analysis of Cellular Spectrum

“…As studied in literature [1,13], researchers analyzed spectrum occupancy using conventional methods but current technology is moving for autonomous examination therefore conventional method is not competent solution to analyze occupancy of spectrum even cannot process continuously for 24 hours.…”

“…In case study [1,13], authors evaluated the spectrum occupancy model using traditional statistical or probabilistic tools but they have some limitations because these tools require assumption to drive their theories either value is a random variable or a randomly process.…”

“…In this work, we analyze spectrum occupancy in three transmission bands that are GSM 900, 1800, and 2100 in order to determine unused spectrum within the three bands in indoor and outdoor locations using the Rohde & Schwarz FSH6 spectrum analyzer. Using MATLAB, the measured results of all three bands were determined against the power spectral density versus frequency plots and show that there exists a lot amount of unused spectrum in a licensed band [13]. To find spectrum occupancy using MATLAB is not also an efficient solution because using MATLAB one Copyright © 2021 MECS I.J.…”