Conditional Handover (CHO) by the 3rd Generation Partnership Project (3GPP) enables efficient user mobility between Base Stations (BSs) by preselecting and preparing Target BSs (T-BSs). However, CHO relies on signal strength for T-BS selection, leading to resource blocking on multiple T-BSs due to signal fluctuations. Existing state-of-the-art methods use deep learning to narrow the list of T-BSs but still lack an effective method for resource reservation timing. This paper presents in-time CHO (iCHO) which exploits historical mobility data to estimate user dwell time at the current BS to reduce resource reservation duration. The proposed iCHO employs a Multivariate Multi-output Single-step Prediction (MMSP) model that leverages a multi-task learning approach to simultaneously predict the minimal list of required T-BSs together with the user dwell time. The model demonstrates remarkable performance across two mobility datasets of different scales, achieving T-BS prediction accuracies of 98% and 95%. It also ensures a 100% handover success rate with a minimum of three and four predicted T-BSs for both datasets, respectively, significantly limiting the list of T-BSs. Moreover, the MMSP model achieves a Mean Absolute Error (MAE) of 19 s and 45 s when predicting the user’s dwell time at the current BS. By utilizing these predictions, iCHO reserves resources at the minimum number of T-BSs immediately before handover. Thus, iCHO can save up to 99% of resources from blockage as compared to the CHO, enabling operators to increase revenue by serving up to eighteen more users with the saved resources.

An automated analysis of respiratory sounds using Deep Learning (DL) plays a pivotal role in the early detection of lung diseases. However, current DL methods often examine the spatial and temporal characteristics of respiratory sounds in isolation, which inherently limit their potential. This study proposes a novel DL framework that captures spatial features through convolution operations and exploits the spatiotemporal correlations of these features using temporal convolution networks. The proposed framework incorporates Multi-Level Temporal Convolutional Networks (ML-TCN) to considerably enhance the model accuracy in detecting anomaly breathing cycles and respiratory recordings from lung sound audio. Moreover, a transfer learning technique is also employed to extract semantic features efficiently from limited and imbalanced data in this domain. Thorough experiments on the well-known ICBHI 2017 challenge dataset show that the proposed framework outperforms state-of-the-art methods in both binary and multi-class classification tasks for respiratory anomaly and disease detection. In particular, improvements of up to 2.29% and 2.27% in terms of the Score metric, average sensitivity and specificity, are demonstrated in binary and multi-class anomaly breathing cycle detection tasks, respectively. In respiratory recording classification tasks, the classification accuracy is improved by 2.69% for healthy-unhealthy binary classification and 1.47% for healthy, chronic, and non-chronic diagnosis. These results highlight the marked advantage of the ML-TCN over existing techniques, showcasing its potential to drive future innovations in respiratory healthcare technology.

The mobile network traffic patterns in urban areas significantly diverge depending on commercial and residential establishments. These regional traffic patterns provide crucial clues for predicting traffic patterns precisely. Previous studies have employed a combination of time-series and convolutional Deep Learning (DL) models to effectively capture the correlation of the regional features and traffic patterns. Despite promising results, these approaches are limited in identifying pattern similarities among sparsely located regions and can be further improved. To this end, this study proposes a GEospatial clustering and residual COnvolutional temporal long Short-term memory (GECOS) framework consisting of clustering and DL components. The proposed Urbanflow Peak Clustering (UPC) component exploits the peak traffic times of daily mobile data to obtain the groups of cells with similar traffic patterns apart from their geographical diversity. The UPC improves the scalability of existing algorithms and enables DL components to improve their accuracy by recognizing unique regional patterns and localizing the training targets. The proposed Residual Convolutional TCN-LSTM (RCTL) serves as the DL component of GECOS that improves TCN-LSTM structure through layer-wise feature transfer and enhances long-term dependency learnability. The RCTL ensures more accurate capturing of extensive spatiotemporal features through structural enhancements. The experiments conducted on real-world mobile traffic data showcase 43% improvement by GECOS compared to state-of-the-art models, enabling precise traffic engineering policies by operators.

Mobile traffic data in urban regions shows differentiated patterns during different hours of the day. The exploitation of these patterns enables highly accurate mobile traffic prediction for proactive network management. However, recent Deep Learning (DL) driven studies have only exploited spatiotemporal features and have ignored the geographical correlations, causing high complexity and erroneous mobile traffic predictions. This paper addresses these limitations by proposing an enhanced mobile traffic prediction scheme that combines the clustering strategy of daily mobile traffic peak time and novel multi Temporal Convolutional Network with a Long Short Term Memory (multi TCN-LSTM) model. The mobile network cells that exhibit peak traffic during the same hour of the day are clustered together. Our experiments on large-scale real-world mobile traffic data show up to 28% performance improvement compared to state-of-the-art studies, which confirms the efficacy and viability of the proposed approach.

Registration areas (RAs) play a pivotal role in the localization of UEs in B5G/6G mobile networks for instant service delivery, as they define a region where the network is certain of UE presence in active and idle modes. A UE must update its registration with the network when it changes its RA, hence, it is desirable to increase RA sizes to minimize registration updates but this elevates the paging overhead and vice versa. Conventionally, RAs are manually defined at the network initiation and they largely remain static afterward. This preliminary study proposes a tailored dynamic registration approach, where a dynamic RA is tailored for a UE according to its movement pattern and rate. This is achieved through a Lightweight Recurrent deep learning Model (LRM) that approximates the region of UE presence for the next defined period. The proposed input sequence aggregation and output sequence compression mechanisms in LRM significantly reduce the computational footprint. The preliminary evaluation with open-source dataset confirms that tailored dynamic registration achieves tradeoff between paging and registration and reduces their signaling overheads by an average 54% and 65%, respectively, compared to conventional static RAs. Further, an average 51% reduction in learning time by LRM showcases its robustness and practical viability.

Unmanned Ariel Vehicle (UAV) services with 5G connectivity is an emerging field with numerous applications. Operator-controlled UAV flights and manual static flight configurations are major limitations for the wide adoption of scalability of UAV services. Several services depend on excellent UAV connectivity with a cellular network and maintaining it is challenging in predetermined flight paths. This paper addresses these limitations by proposing a Deep Reinforcement Learning (DRL) framework for UAV path planning with assured connectivity (DUPAC). During UAV flight, DUPAC determines the best route from a defined source to the destination in terms of distance and signal quality. The viability and performance of DUPAC are evaluated under simulated real-world urban scenarios using the Unity framework. The results confirm that DUPAC achieves an autonomous UAV flight path similar to base method with only 2% increment while maintaining an average 9% better connection quality throughout the flight.

Data aggregation is an important approach in IoT sensor networks since it reduces data transfer while also preserving energy and bandwidth. This research investigates the challenge of time-efficient data aggregation in wireless sensor networks, which is critical in military, civilian, and industrial applications. Effective data aggregation algorithm design and optimization are required for quick and interference-free data collection. Machine learning has received attention for outperforming classical heuristic techniques. The research presents the first Graph Neural Network (GNN) model for data aggregation in IoT sensor networks, which incorporates Graph Attention Networks (GATs) and fully connected layers. The GNN-based model learns network topology and node attributes, creating node embeddings and correcting sensor node transmitting time slots. With a centralized training procedure and adapted execution for network size change, the proposed approach achieves satisfactory performance compared to the heuristic algorithm.

Fast data aggregation is crucial for facilitating critical Internet of Things services as it enables the collection of sensory data within strict volume and time constraints. Over the past decades, the data aggregation scheduling problem for minimum latency has garnered significant research attention. Existing approaches to this problem typically schedule all data transmissions based on an aggregation tree, which is constructed without secondary interference. However, such interference can introduce delays when scheduling a transmission from a node to its parent in the tree. To this end, this study proposes an approach called Active Neighbor EXploitation (ANEX) that enables sensor nodes to switch their parents by identifying active neighbors for potential connectivity, irrespective of the receivers established in the tree. Additionally, the scheme prioritizes scheduling nodes with the fewest unscheduled active neighbors, thereby allowing for more concurrent transmissions. ANEX is evaluated through theoretical analysis and extensive simulations under various scenarios. The results demonstrate that ANEX achieves up to 86% faster aggregation compared to the state-of-the-art approach while maintaining an equivalent time complexity.

Service mobility in Multi-access Edge Computing (MEC) paradigm is necessary to provide ultra-Reliable Low Latency Communications for the erratically roaming MEC users. It involves relocation of containerized application services to a strategically selected optimal edge host. During relocation, service containers are unavailable (downtime), resulting in the interruption of ongoing user sessions and increased operational expenses for the network operator. Prolonged service downtime degrades perceived quality of experience for users, and this study handles this problem by proposing a downtime-aware Policy Learning based Capped Downtime (PLCD) service mobility strategy. It exploits Deep Actor-Critic prowess for effectively deciding when and where to relocate a containerized application service while taking user mobility and MEC server resource fluctuations into account. Efficacy of the proposed PLCD strategy is confirmed through simulation experiments, and results indicate over 90% average reduction in service downtime comparing to a baseline scheme.

Software-defined networking provides a programmable and flexible way to manage the network by separating and centralizing the control plane. The data plane entities like software-defined switches and routers use flow entries in flow tables for forwarding the packets. However, the limited switch memory restricts the number of flow entries in the flow tables. This leads to flow table overflow and flow entry reinstallation problems, which severely degrade the network performance. This requires a comprehensive policy for timely eviction of inactive flow entries to avoid overflows and optimally maintain flow tables usage. To this end, many studies have been proposed, but none of them have suggested detailed eviction strategy for UDP flows. This paper proposes a UDP flow eviction strategy which periodically updates the statistical information of UDP flows through reinforcement learning and utilizes it to evict inactive UDP flows. This eviction strategy is combined with the existing TCP flow eviction method to form an eviction system that takes into account the protocol-specific characteristics of the flow. Through three traffic-based experiments, we found that the proposed system reduces the number of overflow occurrences by 27% and flow entries reinstallation by 28%, compared to the random and FIFO policies, resulting in 15% reduction in control signaling overhead.

Current reactive mobility management in cellular networks becomes a bottleneck for ultra-low latency 5G services and severely degrades the QoS. To satisfy the ultra-low latency requirement of 5G services, proactive mobility management is essential where next PoA of the user is predicted with minimal error. Recent studies have used different deep learning algorithms for this purpose, but their results are unacceptable in real networks due to low accuracy. This paper exploits the distributional learning capability of Generative Adversarial Network (GAN) to propose MoGAN for the prediction of user’s next PoA. The generator in MoGAN uses Gated Recurrent Unit to learn the distribution of time-series data and generates the next PoA. Meanwhile, the discriminator evaluates the generated output against the real data to determine its correctness. The model is trained in adversary mode by using the output from the discriminator. The dataset utilized in training and evaluation is collected from one of the university campuses, and the results show 96.33% of prediction accuracy, which is 5% higher than the previous study. Furthermore, MoGAN is more robust under limited data conditions, as it achieves above 90% accuracy with only 50% of the dataset.

An essential requirement in operating a carriergrade network (CGN) is ensuring the high availability and reliability. Software-defined networking (SDN) is expected to address such requirement while improving the network management. One challenging issue faced in the process of enhancing the reliability of SDN-enabled CGN is how to achieve rapid recovery with minimal effort. There are two well-known approaches to determine the failover scope: end-to-end (global) detouring and local detouring. Particularly, the local detouring approach provides an efficient means to achieve faster recovery, as it locally detours the disrupted flows around the failed network components using a preconfigured alternative path. However, it requires thousands of flow entries per switch to be configured. To address the technical challenges, we propose a fault-tolerant forwarding table design (FFTD), which groups the flows using group entries and aggregates the flows using a tagging mechanism for scalable and rapid recovery from the dual-failures of switches or links without overburdening the controller and the flow table's memory. Our extensive emulation results reveal that the proposed FFTD satisfies the CGN's 50 ms recovery requirement. Additionally, it reduces the alternate path flow storage requirement by up to 99%.