AUTOREGRESSIVE INTEGRATED MOVING AVERAGE-BASED PREDICTIVE MODEL FOR BASE STATION AVAILABILITY OF TELECOMMUNICATION NETWORKS IN MINNA

ABSTRACT

There is a standard of 99.999% (five ‗nines‘) availability for telecommunication hardware and software. This is to guarantee the high level of service required by the Mobile Network Operator  (MNO) for service delivery. MNOs in Nigeria and most sub-Saharan Africa countries are, however, not being able to meet up with the expected base station availability mainly due to high restoration time after the outage. In this thesis, the historical Base Transceiver Station (BTS) Availability reports of a thousand data points each for four MNOs were used. The MNOs (MNO W, MNO X, MNO Y and MNO Z in Minna) data were acquired from 1st of January 2018 to 26th September 2020. The first 73% of the data was partitioned into the Training period and the remaining 27% was set for Validation. The data is in the form of Time Series (TS) and was modelled using Autoregressive Integrated Moving Average (ARIMA) prediction. Correlation plots of the data were done and the ARIMA (p,d,q) parameters were got with the aid of the Autocorrelation Function (ACF) and  the  Partial  Autocorrelation  Function  (PACF).  The  ARIMA-Based  models  for  the MNOs are ARIMA (0,1,3), ARIMA (1,0,1), ARIMA (2,0,4) and ARIMA (0,1,1) for MNO W, MNO X, MNO Y and MNO Z, respectively. The predictive models were used to predict BTS Availability for the MNOs from 27th September 2020 to 20th December 2020. The performance of the models was evaluated with data in the validation period for Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). The MAEs for the respective MNOs are: 1.3959, 0.6602, 1.5666 and 0.6177; while their MAPE are: 0.0150, 0.0068, 0.0176 and 0. 0063. The long short-term (LSTM) model was used for comparison with the ARIMA model for the same MNOs and their MAE and MAPE are 2.8397, 0.8894, 2.8223, and 1.1245; 0.0322, 0.0092, 0.0349 and 0.0118 for MNO W, MNO X, MNO Y and MNO Z respectively. From the results, it is observed that the LSTM models have higher MAE values than the ARIMA models by 51%, 26%, 44% and 45% for MNO W, MNO X, MNO Y and MNO Z respectively. Similarly, for MAPE, the LSTM models have 53%, 26%, 50% and 47% higher values than the ARIMA models for the respective MNOs. These indicate that the ARIMA models have performed better than the LSTM models in all the MNOs. The values of the MAE and MAPE for the predictive models are very low which implies that the predicted Availability data is close to the actual values and can be used for proper planning and decision-making. MNOs can proactively schedule Predictive Maintenance (PdM) with the PdM algorithm developed in this work. Using the 95% availability threshold of this algorithm, MNO W and MNO Y have no savings in maintenance count, while MNO X and MNO Z have savings of 33 and 32 respectively.

CHAPTER ONE

1.0      INTRODUCTION

1.1      Overview

The massive influx of cellular mobile technology has made virtually all aspects of human activities dependent on the use of telecommunications services. For instance, in Nigeria, according to the Nigerian Communication Commission (NCC), the number of active lines in the Global System for Mobile Communications (GSM) rose from 148,774,015 in April 2017 to 198,961,361 in July 2020. A few decades ago, the delivery of telecommunications services like voice calls, SMS and the internet were just emerging and users of the services were not too keen about the extent of the availability of the service. Today, the narrative has changed tremendously; high network availability is usually requested from the Mobile Network Operators (MNO).

Subscribers are anticipating a significant degree of service, from which availability is viewed as the  major  evaluation  of  quality  (Mahdi  et  al.,  2018).  Availability measurement  might  be utilised as a contribution to attract customers, but perhaps more importantly, the operators may profit from its use in evaluating the overall quality of the network (Thulin, 2004).

This  research  is  on  the  Prediction  of  Base  Station  Availability  for  Telecommunication Operators in Minna. This work can be of use to the MNO in the planning process to focus on the aspects of the Network Elements (NE) maintenance. The areas in the NE that are prone to failure could be provided with redundancy. The major MNOs in Nigeria are 9mobile or EMTS (formerly known as Etisalat), Globacom Mobile, Airtel Nigeria and MTN Nigeria.

1.1.1   The Base Transceiver Station (BTS) and the Mobile Cellular Network

The BTS is the cell phone’s admission point to the network (Qing, 2017). It is accountable for carrying out radio communications between  the network and the mobile phone. It executes  speech  encoding,  encryption,  multiplexing (Time Division  Multiplexing), and modulation/demodulation of the radio signals.

The U-mobile (Um) interface is a sort of radio interface liable for the communication between the mobile station and the BTS. It makes available the link joining the Mobile Station (MS) and the GSM system. Its physical linking is accomplished through the radio waves. The Um interface is the main interface amongst all the interfaces in the GSM framework.

A BTS is supervised by a Base Station Controller (BSC) using the Base Station Control Function (BCF). The BCF is designed as a discrete unit or even combined in a Transceiver (TRX) in compact base stations. The BCF offers an Operations and Maintenance (O&M) connection to the Network Management System (NMS) and accomplishes the operational conditions of each Transmit/Receive (TRX), as well as software management and alarm gathering (Qing, 2017). The basic structure and roles of the BTS remain the same irrespective of the wireless technologies.

According to Mahdi et al. (2018), a mobile cellular network is described as a communication infrastructure comprising Network Elements (NEs) that allow Mobile Stations or User Equipment (UE) to access network services through radio channels. Mobile cellular networks usually span large geographical service area which is subdivided into smaller service areas known as cells. Each cell has a fixed access point called a BTS (NodeB for 3G, eNodeB for 4G and gNodeB for 5G) which resides within the cell for wireless communication with UE. Figure 1.1 shows a typical communications network.

For illustrative purposes, the author has shown from fieldwork, some of the equipment that make up the telecommunications network as depicted in plates I to III in Appendix B.

1.1.2   Brief History and Evolution of Mobile Telecommunication in Nigeria

In the year, 2001, Global System for Mobile Communication (GSM) was rolled out in Nigeria with the deployment of the second-generation technology (2G). It came with both voice and SMS services and the data rate was 9.6kbps. The 3G was first launched in Japan using the Wide Code Division Multiple Access (WCDMA) technology in 2001 (Meraj and Kumar, 2015). 3G was deployed in Nigeria much later. In 2008, 4G was launched with the Long-Term Evolution (LTE) technology and it has a data rate of 150Mbps. Figure 1.2 illustrates the evolution of mobile technology from 2G in 1980 to 4G in 2008.

As the technology evolved, the data rate increased as well. There was an improvement of the data transmission rate of 117.2Kbps of the General Packet Radio Service (GPRS) to 384Kbps of the Enhanced Data rates for GSM Evolution (EDGE). With further evolution as depicted in Figure 1.2, a transmission data rate of 2Mbps of the Universal Mobile Telecommunication System (UMTS) was obtained in 3G technology. The 4G LTE has a data transmission rate of 150Mbps.

1.2      Statement of the Research Problem

Telecommunication hardware (HW) and software(SW) are specifically intended to support the Availability requirement of five or six ‗nines‘ (Hilt, 2019; Netes, 2018; Akinsanmi and Adebusuyi,  2016;  Thulin,  2004).  Five  or  six  ‗nines‘ mean  99.999%  or  99.9999% respectively. Availability of 99.999% for instance corresponds to an outage duration of 5 minutes, 15 seconds in a year. MNOs in Nigeria and most Sub-Saharan Africa countries are struggling  to  attain  BTS  Availability  of  two  ‗nines‘ (99.0%).  This  is  very  far  from expectations. There is a need for MNOs to be equipped with the means to proactively schedule predictive maintenance (PdM) or planned preventive maintenance (PPM) on NEs and BTS. This will mitigate or appreciably reduce outage probability and improve BTS availability.

1.3      Aim and Objectives of the Study

This research aims at predicting the availability of Base Stations for Telecommunication Networks in Minna.

To achieve the aim stated above, the following are the objectives:

I.      To acquire and process the Base Station Availability data of four MNOs: MNO W, MNO X, MNO Y and MNO Z in Minna from 1st January 2018 to 26th September 2020.

II.      To  use  the  Data  in  Objective  I  to  develop  Autoregressive  Integrated  Moving Average-Based (ARIMA-Based) Predictive Models of Base Station Availability for the stated MNOs.

III.      To evaluate the performance of the Models using MAE and MAPE, and to compare with the Long Short-Term Memory (LSTM) Model.

1.4      Scope of the Study

This study is limited to focus on BTS Availability for telecommunications network operators in Minna. The population of Availability data (measured in percentage) for  the research is a thousand each for four MNOs named as MNO W, MNO X, MNO Y and MNO Z in Minna metropolis spanning the period from 1st January 2018 – 26th September 2020. The work discusses Availability as a Time Series (TS) data and as such, Time Series Analysis (TSA) shall be discussed. Also, in the study, the concepts of the following: Availability, BTS, predictive models like the Autoregressive Integrated Moving Average (ARIMA) and the LSTM are discussed. Factors affecting BTS Availability are also highlighted in this study.

1.5      Justification for the Study

MNOs use BS Availability for attracting customers. It is used as a measure of quality (Mahdi et al. 2018; Thulin, 2004). Hilt et al. (2016) studied the Availability prediction of telecommunication application servers on cloud. The result showed that legacy telecommunication availability could be attained on cloud-based core networks. This work could have been enhanced if real time TS data was used. Both Hilt et al. (2016) and Mahdi et al. (2018) used Reliability Block Diagram (RBD) method. Mahdi et al. (2018) focussed on Availability measurement and the work lacks predictive ability. Fan et al. (2016) improved the base station Availability by improving the maintenance of the back-up battery group. This research however did not take a complete view of other factors that could impair BS Availability. MNOs in Minna and most Sub-Saharan Africa do not provide the required standard of Availability of 99.999% (5 minutes, 15 seconds of downtime within a year) to their subscribers. In the light of the drawbacks and merits of the previous research, this work leveraged on the amenability of the ARIMA model on TS data to design a predictive model for PdM. This predictive model will be utilised by the policy makers of the MNO to schedule PPM and PdM for a proactive maintenance. This improves the Base Station Availability and reduces operation expenses.

1.6      Thesis Outline

This thesis consists of five chapters as follows: Chapter one is the Introduction. Chapter Two gives the Literature Review and introduces various theoretical background. Chapter Three is the Research Methodology. Chapter Four presents the Results and Discussions. Chapter Five contains the Conclusion, Recommendation and Contributions of the Research to Knowledge.

₦8,000.00 – DOWNLOAD FULL PROJECT DOWNLOAD FULL PROJECT Added to cart

---

---

---

---

Related Project Topics :