Adnan Abu-Dayya
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Adnan Abu-Dayya has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Ali Imran
AI4Networks Research Center, School of Electrical and Computer Engineering, The University of Oklahoma, Norman, OK, USA
14 shared publications - Haneya Naeem Qureshi
AI4Networks Research Center, School of Electrical and Computer Engineering, The University of Oklahoma, Tulsa, OK, USA
5 shared publications - Marvin Manalastas
AI4Networks Research Center, School of Electrical and Computer Engineering, The University of Oklahoma, Tulsa, OK, USA
5 shared publications - Syed Muhammad Asad Zaidi
AI4Networks Research Center, School of Electrical and Computer Engineering, The University of Oklahoma, Tulsa, OK, USA
5 shared publications - Ali Rizwan
Qatar Mobility Innovations Centre, Qatar University, Doha, Qatar
4 shared publications - Elias Yaacoub
Qatar University
4 shared publications
Latest Publications
- 4 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 10 (11) , 5023-5041
LiDAR in Connected and Autonomous Vehicles: Perception, Threat Model, and Defense
Connected and Autonomous Vehicles (CAVs) are referred to as self-driving vehicles that will become an essential component of future intelligent transportation systems. These CAVs will be equipped with various sensors for perceiving their surroundings and onboard computing capabilities to process sensor data in real-time. Light Detection and Ranging (LiDAR) is one of the essential sensors used for detecting objects and accurate distance estimation. However, LiDAR sensors are susceptible to several types of attacks. Adversaries can exploit LiDAR sensors either physically, by sending signals directly to the sensor, or digitally, by manipulating LiDAR data after gaining access to the in-vehicle network. Over the past few years, there has been significant research on the vulnerabilities, attack models, and security of LiDAR sensors. However, to our knowledge, no comprehensive survey exists that addresses these aspects of autonomous vehicle security. This paper aims to bridge this gap by presenting an overview of LiDAR-based perception, data processing, threat models, and defense mechanisms for LiDAR sensors in CAVs. We believe this paper will serve as a valuable reference for researchers, providing a clear understanding of cyber-physical attacks and defense strategies related to LiDAR sensors in autonomous vehicles and related fields.
- 4 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 10 (11) , 5023-5041
LiDAR in Connected and Autonomous Vehicles: Perception, Threat Model, and Defense
Connected and Autonomous Vehicles (CAVs) are referred to as self-driving vehicles that will become an essential component of future intelligent transportation systems. These CAVs will be equipped with various sensors for perceiving their surroundings and onboard computing capabilities to process sensor data in real-time. Light Detection and Ranging (LiDAR) is one of the essential sensors used for detecting objects and accurate distance estimation. However, LiDAR sensors are susceptible to several types of attacks. Adversaries can exploit LiDAR sensors either physically, by sending signals directly to the sensor, or digitally, by manipulating LiDAR data after gaining access to the in-vehicle network. Over the past few years, there has been significant research on the vulnerabilities, attack models, and security of LiDAR sensors. However, to our knowledge, no comprehensive survey exists that addresses these aspects of autonomous vehicle security. This paper aims to bridge this gap by presenting an overview of LiDAR-based perception, data processing, threat models, and defense mechanisms for LiDAR sensors in CAVs. We believe this paper will serve as a valuable reference for researchers, providing a clear understanding of cyber-physical attacks and defense strategies related to LiDAR sensors in autonomous vehicles and related fields.
- 27 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (12) , 15693-15705
AI-Assisted Joint Search Approach for mmwave Cell Discovery Using Sparse MDT Data
High signal directivity and sensitivity to blockages make the mmWave base station (BS) discovery a challenging problem in emerging mobile networks. Existing solutions include methods that rely on the exhaustive periodic beam sweeping and thus have high latency and low mmWave cell discovery rate. Joint methods where macro-BS determines the mmWave cell for a given user equipment based solely on spatial proximity are prone to Non Line of Sight (NLoS) conditions. Recent Artificial Intelligence (AI)-based solutions address the above problems but rely on impractical assumption of having complete minimization of drive test (MDT) reports traces at the assisting macro BSs in all areas of interest. This paper is the first to present an AI-based framework that can utilize very sparse MDT style data to enable NLoS-aware low latency mmWave cell discovery, hereafter referred to as AI -enabled S parse Data based M mWave cell disc O very and EN-DC activation framework (AISMO). In AISMO framework, we first gather historic MDT traces of mmWave users containing signal strength and Radio Link Failure (RLF) indicators. We then augment this highly sparse MDT data using a variety of interpolation, domain knowledge, and AI-based techniques to create augmented mmWave coverage maps (mW-Amaps) while incorporating the NLoS conditions through the RLF traces that are inherently embedded in the data. The mW-Amaps are then used by the macro BS to determine the optimal mmWave cell for a given user location. Results show that our proposed domain knowledge-based MDT data augmentation approach, which we call Weighted Nearest Neighbor Count (WNNC), outperforms other data sparsity alleviation techniques for mW-Amap creation with accuracy of 96%. Second best in terms of accuracy is a deep learning-based solution, that has almost a 30 $x$ faster training time than the WNNC. To evaluate AISMO's performance in a realistic 5 G deployment scenario, we present a case study where AISMO is used to enable E-UTRAN New-Radio Dual-Connectivity (EN-DC) between macro BS and mmWave small cells. Results show that compared to the state-of-the-art nearest mmWave BS approach that leads to 26% EN-DC activation failure due to NLoS conditions, AISMO practically offers zero EN-DC failure rate by avoiding unnecessary mmWave cell searches when the UE is in NLoS condition. Similarly, when compared with the state-of-the-art AI-based mmWave cell discovery through sparse data only, AISMO enables 3x more EN-DC activations leading to proportionally higher use of mmWave band.
- 27 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (12) , 15693-15705
AI-Assisted Joint Search Approach for mmwave Cell Discovery Using Sparse MDT Data
High signal directivity and sensitivity to blockages make the mmWave base station (BS) discovery a challenging problem in emerging mobile networks. Existing solutions include methods that rely on the exhaustive periodic beam sweeping and thus have high latency and low mmWave cell discovery rate. Joint methods where macro-BS determines the mmWave cell for a given user equipment based solely on spatial proximity are prone to Non Line of Sight (NLoS) conditions. Recent Artificial Intelligence (AI)-based solutions address the above problems but rely on impractical assumption of having complete minimization of drive test (MDT) reports traces at the assisting macro BSs in all areas of interest. This paper is the first to present an AI-based framework that can utilize very sparse MDT style data to enable NLoS-aware low latency mmWave cell discovery, hereafter referred to as AI -enabled S parse Data based M mWave cell disc O very and EN-DC activation framework (AISMO). In AISMO framework, we first gather historic MDT traces of mmWave users containing signal strength and Radio Link Failure (RLF) indicators. We then augment this highly sparse MDT data using a variety of interpolation, domain knowledge, and AI-based techniques to create augmented mmWave coverage maps (mW-Amaps) while incorporating the NLoS conditions through the RLF traces that are inherently embedded in the data. The mW-Amaps are then used by the macro BS to determine the optimal mmWave cell for a given user location. Results show that our proposed domain knowledge-based MDT data augmentation approach, which we call Weighted Nearest Neighbor Count (WNNC), outperforms other data sparsity alleviation techniques for mW-Amap creation with accuracy of 96%. Second best in terms of accuracy is a deep learning-based solution, that has almost a 30 $x$ faster training time than the WNNC. To evaluate AISMO's performance in a realistic 5 G deployment scenario, we present a case study where AISMO is used to enable E-UTRAN New-Radio Dual-Connectivity (EN-DC) between macro BS and mmWave small cells. Results show that compared to the state-of-the-art nearest mmWave BS approach that leads to 26% EN-DC activation failure due to NLoS conditions, AISMO practically offers zero EN-DC failure rate by avoiding unnecessary mmWave cell searches when the UE is in NLoS condition. Similarly, when compared with the state-of-the-art AI-based mmWave cell discovery through sparse data only, AISMO enables 3x more EN-DC activations leading to proportionally higher use of mmWave band.
- 28 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 9 (1) , 130-145
D-RAN: A DRL-Based Demand-Driven Elastic User-Centric RAN Optimization for 6G & Beyond
With highly heterogeneous application requirements, 6G and beyond cellular networks are expected to be demand-driven, elastic, user-centric, and capable of supporting multiple services. A redesign of the one-size-fits-all cellular architecture is needed to support heterogeneous application needs. While several recent works have proposed user-centric cloud radio access network (UCRAN) architectures, these works do not consider the heterogeneity of application requirements or the mobility of users. Even though significant gains in performance have been reported, the inherent rigidity of these methods limits their ability to meet the quality of service (QoS) expected from future cellular networks. This paper addresses this need by proposing an intelligent, demand-driven, elastic UCRAN architecture capable of providing services to a diverse set of use cases including augmented/virtual reality, high-speed rails, industrial robots, E-health, and more applications. The proposed framework leverages deep reinforcement learning to adjust the size of a user-centered virtual cell based on each application’s heterogeneous requirements. Furthermore, the proposed architecture is adaptable to varying user demands and mobility while performing multi-objective optimization of key network performance indicators (KPIs). Finally, numerical results are presented to validate the convergence, adaptability, and performance of the proposed approach against meta-heuristics and brute-force methods.
- 28 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 9 (1) , 130-145
D-RAN: A DRL-Based Demand-Driven Elastic User-Centric RAN Optimization for 6G & Beyond
With highly heterogeneous application requirements, 6G and beyond cellular networks are expected to be demand-driven, elastic, user-centric, and capable of supporting multiple services. A redesign of the one-size-fits-all cellular architecture is needed to support heterogeneous application needs. While several recent works have proposed user-centric cloud radio access network (UCRAN) architectures, these works do not consider the heterogeneity of application requirements or the mobility of users. Even though significant gains in performance have been reported, the inherent rigidity of these methods limits their ability to meet the quality of service (QoS) expected from future cellular networks. This paper addresses this need by proposing an intelligent, demand-driven, elastic UCRAN architecture capable of providing services to a diverse set of use cases including augmented/virtual reality, high-speed rails, industrial robots, E-health, and more applications. The proposed framework leverages deep reinforcement learning to adjust the size of a user-centered virtual cell based on each application’s heterogeneous requirements. Furthermore, the proposed architecture is adaptable to varying user demands and mobility while performing multi-objective optimization of key network performance indicators (KPIs). Finally, numerical results are presented to validate the convergence, adaptability, and performance of the proposed approach against meta-heuristics and brute-force methods.
- 4 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (2) , 2381-2394
A Data Driven Framework for QoE-Aware Intelligent EN-DC Activation
In emerging 5G networks, User Equipment (UE) camps traditionally on 4G network. Later on, if the UE requests a 5G service, it can simultaneously camp on 4G and 5G through a 3GPP-standardized approach known as E-UTRAN New-Radio Dual-Connectivity (EN-DC). In EN-DC, poor radio conditions in either 4G or 5G network can be detrimental to user Quality-of-Experience (QoE). Although operators want to maximize EN-DC activation to fully utilize the 5G network, sub-optimal parameter configuration to turn on EN-DC can compromise retainability and integrity Key Performance Indicators (KPIs) due to excessive radio link failures (RLFs) or voice call muting after EN-DC activation. While the need to maximize the EN-DC activation is obvious for maximizing the 5G network's utility, RLF and mute avoidance are equally important to maintain the QoE requirements. This paper presents the first solution to optimally configure the EN-DC activation parameters to achieve the desired trade-off between maximizing 5G sites utility and user QoE. We collect two sets of data from real network to develop machine learning models to predict RLF and muting, respectively. We also investigate and compare the potential of various under-sampling, oversampling, and synthetic data generation techniques including Tomek Links and Generative Adversarial Networks (GAN) for their potential to address the data imbalance problem inherent in the real network training data. Leveraging these models, we formulate and solve two QoE-aware optimization problems that can maximize EN-DC activation while minimizing either RLF or voice muting. System level simulation-based results show that compared to state-of-the-art solution that does not take into account RLF or voice muting risk in EN-DC activation, the proposed solution can intelligently determine EN-DC activation criteria that minimize the risk of RLF and voice muting while giving the operator's desired level of priority to maximize 5G network utilization.
- 4 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (2) , 2381-2394
A Data Driven Framework for QoE-Aware Intelligent EN-DC Activation
In emerging 5G networks, User Equipment (UE) camps traditionally on 4G network. Later on, if the UE requests a 5G service, it can simultaneously camp on 4G and 5G through a 3GPP-standardized approach known as E-UTRAN New-Radio Dual-Connectivity (EN-DC). In EN-DC, poor radio conditions in either 4G or 5G network can be detrimental to user Quality-of-Experience (QoE). Although operators want to maximize EN-DC activation to fully utilize the 5G network, sub-optimal parameter configuration to turn on EN-DC can compromise retainability and integrity Key Performance Indicators (KPIs) due to excessive radio link failures (RLFs) or voice call muting after EN-DC activation. While the need to maximize the EN-DC activation is obvious for maximizing the 5G network's utility, RLF and mute avoidance are equally important to maintain the QoE requirements. This paper presents the first solution to optimally configure the EN-DC activation parameters to achieve the desired trade-off between maximizing 5G sites utility and user QoE. We collect two sets of data from real network to develop machine learning models to predict RLF and muting, respectively. We also investigate and compare the potential of various under-sampling, oversampling, and synthetic data generation techniques including Tomek Links and Generative Adversarial Networks (GAN) for their potential to address the data imbalance problem inherent in the real network training data. Leveraging these models, we formulate and solve two QoE-aware optimization problems that can maximize EN-DC activation while minimizing either RLF or voice muting. System level simulation-based results show that compared to state-of-the-art solution that does not take into account RLF or voice muting risk in EN-DC activation, the proposed solution can intelligently determine EN-DC activation criteria that minimize the risk of RLF and voice muting while giving the operator's desired level of priority to maximize 5G network utilization.
- Open Access
- 22 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Unauthorized Drone Detection: Experiments and Prototypes
The increase in the number of unmanned aerial vehicles a.k.a. drones pose several threats to public privacy, critical infrastructure and cyber security. Hence, detecting unauthorized drones is a significant problem which received attention in the last few years. In this paper, we present our experimental work on three drone detection methods (i.e., acoustic detection, radio frequency (RF) detection, and visual detection) to evaluate their efficacy in both indoor and outdoor environments. Owing to the limitations of these schemes, we present a novel encryption-based drone detection scheme that uses a two-stage verification of the drone’s received signal strength indicator (RSSI) and the encryption key generated from the drone’s position coordinates to reliably detect an unauthorized drone in the presence of authorized drones.
- 22 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Open AccessUnauthorized Drone Detection: Experiments and Prototypes
The increase in the number of unmanned aerial vehicles a.k.a. drones pose several threats to public privacy, critical infrastructure and cyber security. Hence, detecting unauthorized drones is a significant problem which received attention in the last few years. In this paper, we present our experimental work on three drone detection methods (i.e., acoustic detection, radio frequency (RF) detection, and visual detection) to evaluate their efficacy in both indoor and outdoor environments. Owing to the limitations of these schemes, we present a novel encryption-based drone detection scheme that uses a two-stage verification of the drone’s received signal strength indicator (RSSI) and the encryption key generated from the drone’s position coordinates to reliably detect an unauthorized drone in the presence of authorized drones.
- Open Access
- 23 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 67140-67151
A Hybrid Deep Learning-Based (HYDRA) Framework for Multifault Diagnosis Using Sparse MDT Reports
Diminishing viability of manual fault diagnosis in the increasingly complex emerging cellular network has motivated research towards artificial intelligence (AI)-based fault diagnosis using the minimization of drive test (MDT) reports. However, existing AI solutions in the literature remain limited to either diagnosis of faults in a single base station only or the diagnosis of a single fault in a multiple BS scenario. Moreover, lack of robustness to MDT reports spatial sparsity renders these solutions unsuitable for practical deployment. To address this problem, in this paper we present a novel framework named Hy brid D eep Learning-based R oot Cause A nalysis (HYDRA) that uses a hybrid of convolutional neural networks, extreme gradient boosting, and the MDT data enrichment techniques to diagnose multiple faults in a multiple base station network. Performance evaluation under realistic and extreme settings shows that HYDRA yields an accuracy of 93% and compared to the state-of-the-art fault diagnosis solutions, HYDRA is far more robust to MDT report sparsity.
- 23 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 67140-67151
Open AccessA Hybrid Deep Learning-Based (HYDRA) Framework for Multifault Diagnosis Using Sparse MDT Reports
Diminishing viability of manual fault diagnosis in the increasingly complex emerging cellular network has motivated research towards artificial intelligence (AI)-based fault diagnosis using the minimization of drive test (MDT) reports. However, existing AI solutions in the literature remain limited to either diagnosis of faults in a single base station only or the diagnosis of a single fault in a multiple BS scenario. Moreover, lack of robustness to MDT reports spatial sparsity renders these solutions unsuitable for practical deployment. To address this problem, in this paper we present a novel framework named Hy brid D eep Learning-based R oot Cause A nalysis (HYDRA) that uses a hybrid of convolutional neural networks, extreme gradient boosting, and the MDT data enrichment techniques to diagnose multiple faults in a multiple base station network. Performance evaluation under realistic and extreme settings shows that HYDRA yields an accuracy of 93% and compared to the state-of-the-art fault diagnosis solutions, HYDRA is far more robust to MDT report sparsity.
- Top Cited PapersOpen Access
- 29 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (12) , 11439-11455
On the Detection of Unauthorized Drones—Techniques and Future Perspectives: A Review
The market size of civilian drones is tremendously increasing and is expected to reach 1.66 million by the end of 2023. The increase in number of civilian drones poses several privacy and security threats. To safeguard critical assets and infrastructure and to protect privacy of people from the illegitimate uses of commercial drones, a drone detection system is inevitable. In particular, there is a need for a drone detection system that is efficient, accurate, robust, cost-effective and scalable. Recognizing the importance of the problem, several drone detection approaches have been proposed over time. However, none of these provides sufficient performance due to the inherited limitations of the underlying detection technology. More specifically, there are trade-offs among various performance metrics e.g., accuracy, detection range, and robustness against environmental conditions etc. This motivates an in-depth study and critical analysis of the existing approaches, highlighting their potential benefits and limitations. In this paper, we provide a rigorous overview of the existing drone detection techniques and a critical review of the state-of-the-art. Based on the review, we provide key insights on the future drone detection systems. We believe these insights will provide researchers and practicing engineers a holistic view to understand the broader context of the drone detection problem.
- 29 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (12) , 11439-11455
Top Cited PapersOpen AccessOn the Detection of Unauthorized Drones—Techniques and Future Perspectives: A Review
The market size of civilian drones is tremendously increasing and is expected to reach 1.66 million by the end of 2023. The increase in number of civilian drones poses several privacy and security threats. To safeguard critical assets and infrastructure and to protect privacy of people from the illegitimate uses of commercial drones, a drone detection system is inevitable. In particular, there is a need for a drone detection system that is efficient, accurate, robust, cost-effective and scalable. Recognizing the importance of the problem, several drone detection approaches have been proposed over time. However, none of these provides sufficient performance due to the inherited limitations of the underlying detection technology. More specifically, there are trade-offs among various performance metrics e.g., accuracy, detection range, and robustness against environmental conditions etc. This motivates an in-depth study and critical analysis of the existing approaches, highlighting their potential benefits and limitations. In this paper, we provide a rigorous overview of the existing drone detection techniques and a critical review of the state-of-the-art. Based on the review, we provide key insights on the future drone detection systems. We believe these insights will provide researchers and practicing engineers a holistic view to understand the broader context of the drone detection problem.
- 8 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 71 (6) , 6158-6172
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.
- 8 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 71 (6) , 6158-6172
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.
- 18 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 8 (2) , 570-583
A Data-Driven Self-Optimization Solution for Inter-Frequency Mobility Parameters in Emerging Networks
Densification and multi-band operation means inter-frequency handovers can become a bottleneck for mobile user experience in emerging cellular networks. The challenge is aggravated by the fact that there does not exist a method to optimize key inter-frequency handover parameters namely A5 time-to-trigger, A5-threshold1 and A5-threshold2. This paper presents a first study to analyze and optimize the three A5 parameters for jointly maximizing three key performance indicators that reflect mobile user experience: handover success rate (HOSR), reference signal received power (RSRP), and signal-to-interference-plus-noise-ratio (SINR). As analytical modeling cannot capture the system-level complexity, we exploit a data-driven approach. To minimize the training data generation time, we exploit shapley additive explanations (SHAP) sensitivity analysis. The insights from SHAP analysis allow the selective collection of the training data thereby enabling the easier implementation of the proposed solution in a real network. We show that joint RSRP, SINR and HOSR optimization problem is non-convex and solve it using genetic algorithm (GA). We then propose an intelligent mutation scheme for GA, which makes the solution 5x times faster than the legacy GA and 21x faster than the brute force search. This paper thus presents first solution to implement computationally efficient closed-loop self-optimization of inter-frequency mobility parameters.
- 18 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 8 (2) , 570-583
A Data-Driven Self-Optimization Solution for Inter-Frequency Mobility Parameters in Emerging Networks
Densification and multi-band operation means inter-frequency handovers can become a bottleneck for mobile user experience in emerging cellular networks. The challenge is aggravated by the fact that there does not exist a method to optimize key inter-frequency handover parameters namely A5 time-to-trigger, A5-threshold1 and A5-threshold2. This paper presents a first study to analyze and optimize the three A5 parameters for jointly maximizing three key performance indicators that reflect mobile user experience: handover success rate (HOSR), reference signal received power (RSRP), and signal-to-interference-plus-noise-ratio (SINR). As analytical modeling cannot capture the system-level complexity, we exploit a data-driven approach. To minimize the training data generation time, we exploit shapley additive explanations (SHAP) sensitivity analysis. The insights from SHAP analysis allow the selective collection of the training data thereby enabling the easier implementation of the proposed solution in a real network. We show that joint RSRP, SINR and HOSR optimization problem is non-convex and solve it using genetic algorithm (GA). We then propose an intelligent mutation scheme for GA, which makes the solution 5x times faster than the legacy GA and 21x faster than the brute force search. This paper thus presents first solution to implement computationally efficient closed-loop self-optimization of inter-frequency mobility parameters.
- Open Access
Machine Learning Aided Holistic Handover Optimization for Emerging Networks
In the wake of network densification and multi-band operation in emerging cellular networks, mobility and handover management is becoming a major bottleneck. The problem is further aggravated by the fact that holistic mobility management solutions for different types of handovers, namely inter-frequency and intra-frequency handovers, remain scarce. This paper presents a first mobility management solution that concurrently optimizes inter-frequency related A5 parameters and intra-frequency related A3 parameters. We analyze and optimize five parameters namely A5-time to trigger (TTT), A5-threshold1, A5-threshold2, A3-TTT, and A3-offset to jointly maximize three critical key performance indicators (KPIs): edge user reference signal received power (RSRP), handover success rate (HOSR) and load between frequency bands. In the absence of tractable analytical models due to system level complexity, we leverage machine learning to quantify the KPIs as a function of the mobility parameters. An XGBoost based model has the best performance for edge RSRP and HOSR while random forest outperforms others for load prediction. An analysis of the mobility parameters provides several insights: 1) there exists a strong coupling between A3 and A5 parameters; 2) an optimal set of parameters exists for each KPI; and 3) the optimal parameters vary for different KPIs. We also perform a SHAP based sensitivity to help resolve the parametric conflict between the KPIs. Finally, we formulate a maximization problem, show it is non-convex, and solve it utilizing simulated annealing (SA). Results indicate that ML-based SA-aided solution is more than 14x faster than the brute force approach with a slight loss in optimality.
Open AccessMachine Learning Aided Holistic Handover Optimization for Emerging Networks
In the wake of network densification and multi-band operation in emerging cellular networks, mobility and handover management is becoming a major bottleneck. The problem is further aggravated by the fact that holistic mobility management solutions for different types of handovers, namely inter-frequency and intra-frequency handovers, remain scarce. This paper presents a first mobility management solution that concurrently optimizes inter-frequency related A5 parameters and intra-frequency related A3 parameters. We analyze and optimize five parameters namely A5-time to trigger (TTT), A5-threshold1, A5-threshold2, A3-TTT, and A3-offset to jointly maximize three critical key performance indicators (KPIs): edge user reference signal received power (RSRP), handover success rate (HOSR) and load between frequency bands. In the absence of tractable analytical models due to system level complexity, we leverage machine learning to quantify the KPIs as a function of the mobility parameters. An XGBoost based model has the best performance for edge RSRP and HOSR while random forest outperforms others for load prediction. An analysis of the mobility parameters provides several insights: 1) there exists a strong coupling between A3 and A5 parameters; 2) an optimal set of parameters exists for each KPI; and 3) the optimal parameters vary for different KPIs. We also perform a SHAP based sensitivity to help resolve the parametric conflict between the KPIs. Finally, we formulate a maximization problem, show it is non-convex, and solve it utilizing simulated annealing (SA). Results indicate that ML-based SA-aided solution is more than 14x faster than the brute force approach with a slight loss in optimality.