Drive Test Data Research Articles

A data‐driven assessment of mobile operator service quality in Ghana

AbstractThe rapid proliferation of mobile services has increased the need for data‐driven oversight of service quality, yet deriving insights from regulator‐collected datasets remains challenging. This study demonstrates techniques to tap the rich potential of drive test measurement data for analytical regulatory and policy decision‐making. Focusing on leading operator MTN in Ghana, we analyzed 4 years of drive test data supplied by the telecom regulator for the capital city of Accra. Three key performance indicators were evaluated—coverage, call setup time, and speech quality. We assessed service quality trends through statistical summaries, data visualization, and machine learning modeling and predicted speech quality scores. Our analysis revealed deteriorating performance post‐2019 and found that the light gradient boosting machine algorithm provided the highest accuracy predictions of speech quality. Overall, this work showcases how regulators can capitalize on vast datasets using big data mining techniques to evaluate network conditions over time and geography, enhancing field measurements for oversight. Our approach and techniques provide a template for evidence‐based policy‐making to uphold consumer service quality as mobile networks evolve.

Low-density parity-check codes: tracking non-stationary channel noise using sequential variational Bayesian estimates

We present a sequential Bayesian learning method for tracking non-stationary signal-to-noise ratios in low-density parity-check (LDPC) codes by way of probabilistic graphical models. We represent the LDPC code as a cluster graph using a general purpose cluster graph construction algorithm called the layered trees running intersection property (LTRIP) algorithm. The channel noise estimator is a global gamma cluster, which we extend to allow for Bayesian tracking of non-stationary noise variation. We evaluate our proposed model on real-world 5G drive-test data. Our results show that our model can track non-stationary channel noise accurately while adding performance benefits to the LDPC code, which outperforms an LDPC code with a fixed stationary knowledge of the actual channel noise.

Physical Tunning Influence on 4G Network in Minangkabau International Airport

The growth of cellular telephone subscribers experienced a significant increase compared to other industries. With large and widespread number of subscribers, many operators have established new Base Transceiver Stations (BTS) to keep up with the increase number of subscribers. However, the addition of BTS is not a total solution to offset the increase number of subscribers. The purpose of this study is to analyze the signal quality, coverage area and network repairs by using physical tuning method. In Telkomsel operator drive test data in Minangkabau International Airport area, the RSRP parameter obtained an average sample data area around the site in the range -110 to 0, with a red indicator of 3.35%, an orange indicator of 18.96% and a yellow indicator amounted 36.3%. In this way, this research optimizes physical tuning method, including Re-azimuth which successfully directs the antenna to the badspot area, then changes the height of the antenna which makes it covered farther and tilting the Antenna which succeeds in making the coverage area more optimal. For the next bad spot area, optimization should be done in real time in the field so that the improvement can be resolved properly.

Research on Computational Intelligence Algorithm in LTE Channel Estimation

Because data traffic is growing at a rapid pace thanks to advancements in the Internet of Things, precise modelling and precisely anticipating Long-Term Evolution (LTE) Channel is critical for a variety of applications like as video streaming, effective bandwidth consumption, and power management. In this research, we propose a model based on a Computational Intelligence (CI) Algorithm that may enhance Channel Estimation based on received signal. Two Algorithms are considered. In contrast to previous work that focused solely on designing models to estimate channel using traditional Minimum Mean Square Error (MMSE) and Least Square (LS) algorithms, we used 1) GA (Genetic Algorithm) and 2) PSO (Particle Swarm Optimization Algorithm) to work on Discrete and Continuous Long-Term Evolution (LTE) drive test data. We're looking at LTE in the 5.8 GHz band in particular. By lowering the mean square error of LS and the complexity of MMSE, the design model attempts to improve channel estimation. Pilots are put at random and sent with data to gather channel information, which aids the receiver in decoding and estimating the channel using LS, MMSE, Taguchi GA, and PSO. The Bit Error Rate (BER), Signal to Noise Ratio, and Mean Square Error of a CI-based model have all been estimated. In comparison to the MMSE and LS algorithms, the proposed model BER achieves the target gain of 2.4 dB and 5.4 dB.

Synthetic Generation of Realistic Signal Strength Data to Enable 5G Rogue Base Station Investigation in Vehicular Platooning

Rogue Base Stations (RBS), also known as 5G Subscription Concealed Identifier (SUCI) catchers, were initially developed to maliciously intercept subscribers’ identities. Since then, further advances have been made, not only in RBSs, but also in communication network security. The identification and prevention of RBSs in Fifth Generation (5G) networks are among the main security challenges for users and network infrastructure. The security architecture group in 3GPP clarified that the radio configuration information received from user equipment could contain fingerprints of the RBS. This information is periodically included in the measurement report generated by the user equipment to report location information and Received Signal Strength (RSS) measurements for the strongest base stations. The motivation in this work, then is to generate 5G measurement reports to provide a large and realistic dataset of radio information and RSS measurements for an autonomous vehicle driving along various sections of a road. These simulated measurement reports can then be used to develop and test new methods for identifying an RBS and taking mitigating actions. The proposed approach can generate 20 min of synthetic drive test data in 15 s, which is 80 times faster than real time.

Automatic Fault Detection and Diagnosis in Cellular Networks and Beyond 5G: Intelligent Network Management

Handling faults in a running cellular network can impair the performance and dissatisfy the end users. It is important to design an automatic self-healing procedure to not only detect the active faults, but also to diagnosis them automatically. Although fault detection has been well studied in the literature, fewer studies have targeted the more complicated task of diagnosing. Our presented method aims to tackle fault detection and diagnosis using two sets of data collected by the network: performance support system data and drive test data. Although performance support system data is collected automatically by the network, drive test data are collected manually in three mode call scenarios: short, long and idle. The short call can identify faults in a call setup, the long call is designed to identify handover failures and call interruption, and, finally, the idle mode is designed to understand the characteristics of the standard signal in the network. We have applied unsupervised learning, along with various classified algorithms, on performance support system data. Congestion and failures in TCH assignments are a few examples of the detected and diagnosed faults with our method. In addition, we present a framework to identify the need for handovers. The Silhouette coefficient is used to evaluate the quality of the unsupervised learning approach. We achieved an accuracy of 96.86% with the dynamic neural network method.

Classified 3D mapping and deep learning-aided signal power estimation architecture for the deployment of wireless communication systems

The traditional wireless communication systems deployment models require expensive and time-consuming procedures, including environment selection (rural, urban, and suburban), drive test data collection, and analysis of the raw data. These procedures mainly utilize stochastic and deterministic approaches for signal strength prediction to locate the optimum cellular tower (eNodeB) position for 4G and 5G systems. Since environment selection is limited by urban, suburban, and rural areas, they do not cover complex macro and micro variations, especially buildings and tree canopies having a higher impact on signal fading due to scattering and absorption. Therefore, they usually end up with high prediction errors. This article proposes an efficient architecture for the deployment of communication systems. The proposed method determines the effect of the environment via extracting tree and building properties by using a classified 3D map and You Only Look Once (YOLO) V5, which is one of the most efficient deep learning algorithms. According to the results, the mean average precision (mAP) 0.5% and mAP 0.95% accuracies are obtained as 0.96 and 0.45, and image color classification (ICC) findings indicate 77.6% accuracy on vegetation detection, especially for tree canopies. Thus, the obtained results significantly improved signal strength prediction with a 3.96% Mean Absolute Percentage Error (MAPE) rate, while other empirical models’ prediction errors fall in the range of 6.07–15.26%.

A Data-Driven Framework for Inter-Frequency Handover Failure Prediction and Mitigation

With 5G already deployed, challenges related to handover exacerbate due to the dense base station deployment operating on a motley of frequencies. In this paper, we present and evaluate a novel data-driven solution, to reduce inter-frequency handover failures (HOFs), hereafter referred to as TORIS (Transmit Power Tuning-based Handover Success Rate Improvement Scheme). TORIS is designed by developing and integrating two sub-solutions. First sub-solution consists of an Artificial Intelligence (AI)-based model to predict inter-frequency HOFs. In this model, we achieve higher than the state-of-the-art accuracy by leveraging two approaches. First, we devise a novel feature set by exploiting domain knowledge gathered from extensive drive test data analysis. Second, we exploit an extensive set of data augmentation techniques to address the class imbalance in training the HOF prediction model. The data augmentation techniques include Chow-Liu Bayesian Network and Generative Adversarial Network further improved by focusing the sampling only on the borderline. We also compare the performance of state-of-the-art AI models for predicting HOFs with and without augmented data. Results show that AdaBoost yields best performance for predicting HOFs. The second sub-solution is a heuristic scheme to tune the transmit (Tx) power of serving and target cells. Unlike the state-of-the-art approaches for HOF reduction that tune cell individual offset, TORIS targets the main cause of HOFs i.e., poor signal quality and propagation condition, by proactively varying the Tx power of the cells whenever a HOF is anticipated. Results show that TORIS outperforms the state-of-the-art HOF reduction solution and yields 40%-75% reduction in HOFs.

3.5 GHz Outdoor Radio Signal Strength Prediction With Machine Learning Based on Low-Cost Geographic Features

In the fifth-generation (5G) and beyond 5G (B5G) networks, received signal strength (RSS) prediction has always been a fundamental task in network planning and optimization phases. The existing machine learning (ML)-based RSS prediction models seldom testify to the model validity with reliable drive test data in 5G networks and tend to neglect the low-cost extraction in feature contriving. This article presents an ML scheme based on the geographic feature to predict RSS. We elaborately select four features closely related to RSS from the easily acquired geographic dataset and design low-cost methods for computing them. Experiments are executed in the large-scale outdoor scenario at 3.5 GHz for a 5G network where the real RSS data are collected by the field measurement in the urban city. We adopt four state-of-the-art ML algorithms for the proposed scheme and compare the algorithm accuracy with a Stanford University interim (SUI) model, an ECC-33 model, and a ray tracing method. Experiment results show the validity of the extracted features. Besides, the proposed scheme performs well in its model accuracy and computational efficiency compared with the existing methods. Furthermore, it is promising to build a universal, accurate, and efficient ML-based RSS prediction scheme based on massive field data in 5G or B5G networks.

Web based 4G Drive Tests Post Processing Software for Educational and Enterprise Purpose

In the world of digital communications, we usually use terms as processing and network optimization. Because mobile operators continuously want to provide the best network experience and quality of service to their subscribers, they continuously seek to improve their network performance both in Coverage, Quality and performance. When mobile technology such as LTE is offered to the public, operators need to be certain that it satisfies their customers. The technology then needs to be tested and optimized if it does not meet the purpose. In this context, one will propose a tool for the processing of drive test data. The proposed tool can be used for enterprises and academic purposes. In fact, online educational courses have brought great dynamics in the field of education and research. It has revolutionized the way learning is being done and perceived in our society compare to traditional learning method. During this period of Covid-19 e-learning is developing faster than before and new solutions to improve it are proposed. For many sub-Saharan African countries, it is very scarce to find telecommunication lecturers with a PhD grade and this has an impact on quality of the training provided to students in telecommunications engineering fields. Another problem is the lack of means to buy and install telecommunication equipments for laboratory experimentations where students can learn and improve their skills during their training program in a university or college. To overcome these problems, a possible solution could be the design of efficient web based tool like the one proposed in this paper and implement labs and e-labs in other to enable students to have a better understanding of mobile network,  drive test data post processing, network quality of service analysis and network optimization. The tool proposed will provide detail analytical results using drive test data. Also optimization techniques and solutions after processing Drive Test data obtained from a live LTE network shall be proposed. For this study, DT data from a mobile operator will be analyzed in details using the poposed tool, and how this analysis results can be used to identify poor performance and quality of service areas in the network. Then following optimization techniques proposed, how defined solutions and strategies can be applied to improve performance of these areas based on current network configurations.

Intelligent Optimization of Base Station Array Orientations via Scenario-Specific Modeling

Fifth-generation (5G) wireless communications confront explosive growth in mobile data demand and massive intensive user equipment (UE) connections. The optimization of base station (BS) array orientations in large-scale networks can provide high potential performance gain to meet these requirements. However, traditional schemes are highly dependent on experience and difficult in implementation due to their demand on repeated drive tests and large amounts of data samples. In this paper, via exploiting UE locations and channel direction information, we propose an intelligent network optimization framework to maximize the long-term network rate performance. For the augmentation of limited drive test data, a deep Gaussian process regression (DGPR) model is designed to construct a scenario-specific channel modeling. In addition, via a domain-knowledge driven fusion of convolutional neural network (CNN) and multi-layer perceptron (MLP), we propose a multi-branch deep neural network (DNN) to accurately map BS array orientations and concise channel measurements to the network performance. Finally, based on the scenario-specific modeling, an efficient gradient search approach is proposed to optimize BS array orientations via neural network (NN) backpropagation. Both simulations and field tests in 5G experimental networks validate the effectiveness and efficiency of our proposed framework, especially with limited drive test data.

An Open Mobile Communications Drive Test Data Set and Its Use for Machine Learning

The capability to provide guarantees for network metrics, such as latency, data rate, and reliability will be an important factor for widespread adoption of next generation mobile networks and hence, such metrics play a central role in standards for new wireless communication technologies. However, due to the inherently stochastic nature of mobile communications, any guarantees can only be of statistical nature and are highly dependent on the actual physical environment. To analyze the stochastic behavior, this paper presents a tool chain for measurement, collection, evaluation, and prediction of controlled mobile communications drive test data. We also publish the underlying data set of measurements covering two years’ worth of highway traffic on a 25long section comprising 267198 data points. We statistically evaluate the data set and validate it with a corresponding data set from another source. Applying machine learning to the data set illustrates possible use cases: Feed-forward neural networks to predict the data rate in five application scenarios, LIME to explain the behavior of the model, and an autoencoder to describe the interaction of five signal strength parameters. The data set and the tool chain show how machine learning can be applied to wireless networks and provide fellow researchers with the means to make further experiments.

Performance Analysis of Mobile Broadband Networks With 5G Trends and Beyond: Urban Areas Scope in Malaysia

The performance of Mobile Broadband (MBB) services of Fourth Generation (4G) and Third Generation (3G) mobile networks over urban morphology is studied in Malaysia based on experimental measurements of drive test data. The aim of this study is to provide a roadmap for service providers to establish a realistic plan for future Fifth Generation (5G) networks. This work is a continuation of our previous work for the scope of rural areas in Malaysia. The MBB measurement data have been gathered through drive tests conducted in the urban areas of four states throughout Malaysia (namely, Klang Valley/Selangor, Johor, Sarawak and Sabah) to characterise and analyse MBB performances. The gathered data are from the cities, highways and federal roads of the chosen states, and encompasses three main Mobile Network Operators (MNOs). Data has been collected in a time span of 2 months, from January to February, using the Samsung Galaxy S6 smartphone handsets. Four MBB Key Performance Indicators (KPIs) are considered in this study (coverage, latency, satisfaction and speed) for two MBB services (web browsing and video streaming). The measurement data for characterising the performance of each MBB service has been collected using a dedicated smartphone handset. YouTube videos with 720p and 1080p resolutions have been sequentially streamed to assess the performance of MBB video-streaming services. Three distinct websites (Google, Instagram and mStar) have been accessed to evaluate the performance of MBB web-browsing services. The experimental methodology of this study integrates several diversified elements including four different urban states, four distinct KPIs, three main MNOs, two MBB services and two radio networks (4G and 3G), which are both accessible by the smartphones when available to mimic real-world scenarios. The results of this study reveal that the performance of 4G radio networks is generally superior to that of 3G. For instance, 4G networks achieved a vMOS score of more than 3 for both MBB video-streaming and web-browsing services, while 3G networks scored less than 3 across all four study areas. The analysed experimental results confirmed that compared to 3G networks, 4G technology presents an enhancement factor of up to 1.6 and 4.2 in download speed when streaming a video and browsing a web page, respectively. The study outcomes can contribute to the efficient planning of non-standalone (NSA) 5G networks in Malaysia where 5G networks will be aided by existing 4G infrastructures. Analysing the 4G coverage performance is the first step towards deciding the deployment rate of NSA 5G in Malaysia.

LTE Network Area Coverage on FDD and TDD Technology

FDD (Frequency Division Duplexing) and TDD (Time Division Duplexing) Technology I is a duplex technology on the 4G LTE network. FDD technology in Indonesia is deployed in the 900 MHz, 1800 MHz and 2100 MHz frequency bands. Meanwhile, TDD technology uses frequency of 2300 MHz. In order to see the performance from the side of coverage, a measurement of the signal strength is conducted by direct measurements in field so that the appropriate technology and frequency band can be determined for the measurement area. The measurements have been done by two samples of different frequencies, namely 1800 MHz for FDD and 2300 MHz for TDD. Measurement of coverage is conducted on the same existing BTS and different BTS for both FDD and TDD modes. The parameter used in this measurement is the RSRQ value as one of the basic parameters of KPI (Key Performance Indicator). The results of drivetest data obtained the farthest coverage area of 789 m, namely the FDD (Frequency Division Duplexing) technology with frequency of 1800 MHz and TDD with frequency of 2300 MHz for 633 m.

LTE Network Area Coverage on FDD and TDD Technology

FDD (Frequency Division Duplexing) and TDD (Time Division Duplexing) Technology I is a duplex technology on the 4G LTE network. FDD technology in Indonesia is deployed in the 900 MHz, 1800 MHz and 2100 MHz frequency bands. Meanwhile, TDD technology uses frequency of 2300 MHz. In order to see the performance from the side of coverage, a measurement of the signal strength is conducted by direct measurements in field so that the appropriate technology and frequency band can be determined for the measurement area. The measurements have been done by two samples of different frequencies, namely 1800 MHz for FDD and 2300 MHz for TDD. Measurement of coverage is conducted on the same existing BTS and different BTS for both FDD and TDD modes. The parameter used in this measurement is the RSRQ value as one of the basic parameters of KPI (Key Performance Indicator). The results of drivetest data obtained the farthest coverage area of 789 m, namely the FDD (Frequency Division Duplexing) technology with frequency of 1800 MHz and TDD with frequency of 2300 MHz for 633 m.

ANALISIS DROP CALL PADA JARINGAN WIDEBAND CODE-DIVISION MULTIPLE ACCESS (WCDMA) DI CLUSTER RENON

Drop call is an example of a problem that is often experienced by users of voice telecommunications services. From this condition, this riset raises the topic of drop call analysis on WCDMA networks in the Renon cluster (nodeB MYAMIN_CR, AKABA_PL, and BUNDARANRENON_PL). Data retrieval needed (Call Set Up Success Rate (CSSR), Call Drop Ratio (CDR), RSCP and Ec/No) is done by drive test using G-NetTrack Pro software. Based on the analysis of the results of the drive test, there were 10 drop call problems. The problems indentified cause by missing neighbors, handover failure, bad spot, coverage problems and overshoot. Bad spots, coverage problems, and overshoot are the initial causes of missing neighbors and handovers failure. Missing neighbors and handover failure begins with the poor value of RSCP and Ec/No, so that the cellphone will try to do a handover. To analyze the drive test results, drive test data is carried out using Atoll software. From the comparison found extreme conditions resulting from obstacle which cause a significant difference in value between the results of the drive test and simulation. The high attenuation rate drop call location due to the obstacle also results in RSCP and Ec/No values being obtained when measurements in the field are low. The low parameter values of RSCP and Ec / No then cause a drop call.

Machine Learning for Performance Prediction in Mobile Cellular Networks

In this paper, we discuss the application of machine learning techniques for performance prediction problems in wireless networks. These problems often involve using existing measurement data to predict network performance where direct measurements are not available. We explore the performance of existing machine learning algorithms for these problems and propose a simple taxonomy of main problem categories. As an example, we use an extensive real-world drive test data set to show that classical machine learning methods such as Gaussian process regression, exponential smoothing of time series, and random forests can yield excellent prediction results. Applying these methods to the management of wireless mobile networks has the potential to significantly reduce operational costs while simultaneously improving user experience. We also discuss key challenges for future work, especially with the focus on practical deployment of machine learning techniques for performance prediction in mobile wireless networks.

Statistical Inference on Spectrum Data for Design and Enforcement of Harm Claim Thresholds

Harm claim thresholds (HCTs) are a promising approach for regulators to specify interference limits in a technology-neutral fashion, and a useful parameter spectrum access systems can use to manage the aggregate interference caused by transmitters they control. However, existing literature provides very little guidance how HCTs should be set and enforced. In this paper, we propose a detailed regulatory framework for gathering and processing of measurement data for enforcing and setting HCTs. We introduce the central concepts of stratification and weighting of measurement data, and show their importance in ensuring representativeness of measurements and enabling robust estimation of statistical confidence on results. For deriving HCT thresholds from measurements, we propose additional representativeness criteria that a regulator should apply to avoid underestimation of field strength levels related to existing wireless services. We demonstrate application of our proposed framework using an extensive drive test data set, and show that the chosen HCT percentile is critical in determining how much data needs to be gathered for enforcement. We also show how spatial prediction techniques can be used to deal with data sets that have been collected non-uniformly over the region of interest, emphasizing the need for modern bias-corrected techniques.

Near Ground Path Loss Prediction for UMTS 2100 MHz Frequency Band Over Propagating Over a Smooth-Earth Terrain

In this paper, path loss prediction for near ground propagation of third generation (3G)-based Universal Mobile Telecommunications System (UMTS) network signal in the 2100 MHz frequency band over a smooth-earth terrain is presented. Particularly, the attenuation due to diffraction is estimated based on ITU-R Recommendation P.526-13 for diffraction over smooth earth. Furthermore, the total pathloss is determined using the Blomquist empirical model which combined free-space loss, plane-earth loss and the diffraction loss over smooth earth. In the study, two drive tests are conducted for the UMTS 2100 MHz frequency band in suburban area of Uyo. The Blomquist empirical model was tuned with respect to the first drive test pathloss dataset. The results show that with the training data (first drive test data), the untuned Blomquist empirical model has RMSE=10.21344 dB, Prediction Accuracy = 89.92%, minimum Prediction error = 9.02 dB and maximum Prediction error = -34.05 dB. Also, with the training data, the optimized Blomquist empirical model has RMSE=1.625388dB, Prediction Accuracy = 98.48%, minimum Prediction error = 5.34 dB and maximum Prediction error = -5.40 dB. Furthermore, with the cross validation data (second drive test data), the optimized Blomquist empirical model has RMSE=1.831368 dB, Prediction Accuracy = 98.24%, minimum Prediction error = 5.25 dB and maximum Prediction error = -6.15 dB. The results show that for the given terrain under study, the tuned or optimized Blomquist empirical model can effectively predict the pathloss for the UMTS network signal in the 2100 MHz frequency band.

Introducing automated procedures in 3G network planning and optimization

Third generation (3G) networks have been launched for quite some time and cellular operators have already started to face the challenges when operating such kind of networks, compared to their 2G predecessors. In this paper, we give an overview of the network planning and optimization processes in 3G networks. Emphasis is given on the interdependence between coverage and capacity, which increases the complexity of operating 3G networks in a cost effective and quality assuring manner. We stretch out the importance of introducing automated procedures in these challenging tasks and describe main inputs and outputs of these procedures, such as initial network configuration and constraints, drive test data, and key performance indicators statistics and goals. We use two case studies from a operational 3G system to indicate the improvements on the network when these automated procedures are employed.