Mobility Load Balancing Research Articles

Reinforcement learning based mobility load balancing in cellular networks: a two-layered approach

Reinforcement learning based mobility load balancing in cellular networks: a two-layered approach

Fuzzy Logic for the Improving of Handover Decision and the Adaptive Adjustment of Control Parameters in 5G Wireless Networks

Handover process is one of the most important aspects of mobility management in 5G wireless networks. It becomes a hot topic for researchers because it constitutes a guarantee of communication continuity during the user's movement, in addition to being the basic step on which the mobility load balancing process depends to distribute the load between the cells. The focus on this process is whether by providing solutions to improve the handover decision-making, or by modifying the values of the handover control parameters in a way that it guarantees the reduction of handover problems, because the inaccurate or unnecessary modification of these parameters values will cause a degradation in the quality of service. This paper presents a study targeting two mechanisms to improve handover decision-making and selection of handover control parameters adaptively based on different schemes. The first one, based on a learning model called LIM2 and the second one is based on fuzzy logic and is called RHOT-FLC. The results show that the RHOT-FLC mechanism, which relies on fuzzy logic and takes into account the user's velocity provides better performance in term of average throughput, packet drop rate, average HOPP probability, average HO latency, HO failure.

A Framework to Address Mobility Management Challenges in Emerging Networks

The key enablers for emerging cellular networks such as densification, concurrent operation at multiple bands and harnessing mmWave spectrum give birth to a peculiar set of new network management challenges. One such key challenge is the user mobility management. In this article, we identify the key issues that render current mobility management paradigm inadequate for delivering the expected Quality of Experience (QoE) and resource efficiency in emerging and future cellular networks. Together, these challenges call for paradigm shift in the way mobility is managed in cellular networks. We present an Advanced Mobility Management and Utilization Framework (A-MMUF) that can enable this paradigm shift by transforming mobility management from being a reactive to a proactive process. The core idea of A-MMUF is to build upon Mobility Prediction Models (MPMs) to predict various attributes of user mobility and traffic patterns such as next candidate cell for handover (HO), time of HO, future cell loads. These predictions are then leveraged to not only improve HO process for better QoE and less signaling overhead, but also to enable proactive automation to further maximize network performance in terms of energy and spectrum efficiency. However, understandably, the gains offered by the A-MMUF hinge on the accuracy of the MPMs, which may vary, not only with the choice of underlying machine learning technique, but also with the volume, variety, and fidelity of the training data. We analyze the potential gains of A-MMUF through three case studies: namely proactive HO, proactive Mobility Load Balancing (P-MLB), and proactive Energy Savings (P-ES). In addition to demonstrating significant gains in all three case studies, the results provide useful insights into the agility vs accuracy tradeoff that can be leveraged for choosing optimal machine learning models for practical deployment of A-MMUF.

Decentralized Deep Reinforcement Learning Meets Mobility Load Balancing

Mobility load balancing (MLB) aims to solve the problem of uneven resource utilization in cellular networks. Since network dynamics are usually complicated and non-stationary, conventional model-based MLB methods fail to cover all scenarios of cellular networks. On the other hand, deep reinforcement learning (DRL) can provide a flexible framework to learn to distribute cell load evenly without explicit modeling of the underlying network dynamics. In this paper, we introduce a novel decentralized DRL-based MLB method where each cell has a DRL agent to learn its handover parameters and antenna tilt angle. As the number of cells increases, the decentralized framework is more computationally efficient than its centralized counterpart by dividing the action space. Furthermore, our designed decentralized DRL architecture only requires readily known information defined in existing cellular standards, and it can achieve a more balanced cell load distribution than the centralized DRL one by using individual reward functions. To provide realistic performance evaluation, a network simulator is introduced strictly following the Third Generation Partnership Project (3GPP) specifications. Furthermore, field data is used to construct the underlying cellular environment. Extensive evaluations have been conducted to demonstrate the fact that the introduced decentralized DRL-based MLB method can achieve a more balanced cell load distribution and a better performance of edge users than the state-of-the-art MLB methods.

Intelligent coordinated self-optimizing handover scheme for 4G/5G heterogeneous networks

Mobile connections and applications are growing unprecedentedly with the advent and increasing popularity of wireless services worldwide, which greatly increases the demands on data traffic. This led to considering millimeter-wave bands and ultra-dense deployment as part of the key enabler solutions in fifth-generation (5G) networks. However, these solutions radically increase the number of handovers (HOs), thus increasing the rate of unnecessary HO and dropping call probabilities. In this regard, optimizing HO control parameters appropriately is the main factor that can efficiently address HO issues during user mobility. This paper proposes a fuzzy-coordinated self-optimizing HO scheme to achieve a seamless HO while users move in multi-radio access networks. The proposed scheme resolves the conflict between mobility robustness and load balancing functions by utilizing a fuzzy system considering three input parameters: signal-to-interference-plus-noise ratio, cell load and UE speed. Simulation results show that the proposed scheme manages to control the mobility optimization in terms of ping-pong HO, radio link failure and HO latency over different mobile speed scenarios. Moreover, the proposed scheme reduces the outage probability compared to other schemes from literature.

Classification of Request-Based Mobility Load Balancing in Fog Computing

Every day, more and more data is being produced by the Internet of Things (IoT) applications. IoT data differ in amount, diversity, veracity, and velocity. Because of latency, various types of data handling in cloud computing are not suitable for many time-sensitive applications. When users move from one site to another, mobility also adds to the latency. By placing computing close to IoT devices with mobility support, fog computing addresses these problems. An efficient Load Balancing Algorithm (LBA) improves user experience and Quality of Service (QoS). Classification of Request (CoR) based Resource Adaptive LBA is suggested in this research. This technique clusters fog nodes using an efficient K-means clustering algorithm and then uses a Decision Tree approach to categorize the request. The decision-making process for time-sensitive and delay-tolerable requests is facilitated by the classification of requests. LBA does the operation based on these classifications. The MobFogSim simulation program is utilized to assess how well the algorithm with mobility features performs. The outcome demonstrates that the LBA algorithm’s performance enhances the total system performance, which was attained by (90.8%). Using LBA, several metrics may be examined, including Response Time (RT), delay (d), Energy Consumption (EC), and latency. Through the on-demand provisioning of necessary resources to IoT users, our suggested LBA assures effective resource usage.

Proactive Load Balancing Through Constrained Policy Optimization for Ultra-Dense Networks

Designing an intelligent self-organizing network (SON) architecture is challenging for future wireless networks. To meet the needs of SON, the reactive self-organizing model of the traditional network needs to be transformed into an active and interactive one. Due to the user mobility and small coverage of cells in ultra-dense networks (UDNs), the network load usually becomes unbalanced, leading to deteriorated network performance, such as low throughput, radio link failure, and poor user experience. Therefore, the technique of mobility load balancing (MLB) is critical to ensuring a seamless user experience among cells. This letter proposes an active and interactive MLB strategy for UDNs, which transforms the original reactive MLB into a forward-aware and active one. In particular, user mobility is first predicted based on the Bayesian additive regression tree (BART). Then, with the mobility predictions, the joint mobility robust optimization and MLB problem subject to users’ rate constraint is solved via safe reinforcement learning. The pertaining simulation results show that the proposed method can improve the network performance and realize intelligent mobile management for future UDNs.

Enhanced performance of long-term evolution small-cell networks using improved mobility algorithms

With an ever-increasing number of user equipment (UE) and bandwidth demands of new applications, the deployment of dense heterogeneous cellular networks has been adopted in many network scenarios. The small cells are unable to unload the traffic due to the random UEs mobility and cells deployment. This degrades the network performance such as handover success, throughput, and load distribution. To address such a problem, we propose an enhanced mobility load-balancing algorithm for small cells. The conventional mobility load balancing (MLB) algorithms study only the fixed or adaptive thresholds of the network separately to consider the load balance process, while other MLB algorithms consider the neighboring cells of the network those experience unnecessary MLB actions. The proposed load balancing algorithms study overloaded cells and neighbors using the proposed efficiency parameter, B. To identify the overloaded cells, we propose B which compares between a pre-defined threshold and the network threshold to categorize medium and overloaded cells, based on B, the algorithm is triggered. Then, to control the shifted load to a target neighbor cell, we propose the Rescue factor, f. the f ensures the load of the target cell after handover is equal or less than Ɓ. The simulation results showed a lower standard deviation and higher throughput and physical resource block utilization than surveyed algorithms.

Conflict Resolution Strategy in Handover Management for 4G and 5G Networks

Fifth-generation (5G) cellular networks offer high transmission rates in dense urban environments. However, a massive deployment of small cells will be required to provide wide-area coverage, which leads to an increase in the number of handovers (HOs). Mobility management is an important issue that requires considerable attention in heterogeneous networks, where 5G ultra-dense small cells coexist with current fourth-generation (4G) networks. Although mobility robustness optimization (MRO) and load balancing optimization (LBO) functions have been introduced in the 3GPP standard to address HO problems, non-robust and nonoptimal algorithms for selecting appropriate HO control parameters (HCPs) still exist, and an optimal solution is subjected to compromise between LBO and MRO functions. Thus, HO decision algorithms become inefficient. This paper proposes a conflict resolution technique to address the contradiction between MRO and LBO functions. The proposed technique exploits received signal reference power (RSRP), cell load and user speed to adapt HO margin (HM) and time to trigger (TTT). Estimated HM and TTT depend on a weighting function and HO type which is represented by user status during mobility. The proposed technique is validated with other existing algorithms from the literature. Simulation results demonstrate that the proposed technique outperforms existing algorithms overall performance metrics. The proposed technique reduces the overall average HO ping-pong probability, HO failure rate and interruption time by more than 90%, 46% and 58%, respectively, compared with the other schemes overall speed scenarios and simulation time.

Tackling control channel saturation for load balancing in multi-channel mobile ad hoc networks

Multiple wireless channels in Mobile Ad hoc Networks (MANET) promises to improve Quality of Service (QoS) and their well coordinated usage can result in significant operational improvements. Traditional methods implementing multi-channels in MANET uses control channel based approach for synchronization and coordination among multiple channels. Control channel is a separate and dedicated channel reserved for performing channel negotiations and other control operations. In multi-channel MANET, conventionally all mobile devices have to resort to the control channel for channel reservation during a well defined beacon interval. This approach requires tight synchronization and works well when magnitude of traffic and node density is moderate. With the rise in traffic and node density, the control channel becomes saturated and network operation is compromised. The paper proposes Parallel Rendezvous Multi-Channel MAC (PRMMAC) protocol based on the concept of multi-channel communications and results reflects that it resolves the basic saturation of the control channel and ensures a balanced load among multiple channels catering to data communication needs. Key findings of the proposed system has been observed in terms of throughput, frame delivery ratio and delay under varied networks and channel configurations and significant performance improvement in terms of throughout delay and Frame Delivery Ratio (FDR) has been observed in comparison to other protocols of its category.

Reliability and Mobility Load Balancing in Next Generation Self-organized Networks: Using Stochastic Learning Automata

Self-organizing networking (SON) is an automation technology designed to make the planning, configuration, management, optimization and healing of mobile radio access networks simpler and faster. Most current self-organization networking functions apply rule-based recommended systems to control network resources which seem too complicated and time-consuming to design in practical conditions. This research proposes a cognitive cellular network empowered by an efficient self-organization networking approach which enables SON functions to separately learn and find the best configuration setting. An effective learning approach is proposed for the functions of the cognitive cellular network, which exhibits how the framework is mapped to SON functions. One of the main functions applied in this framework is mobility load balancing. In this paper, a novel Stochastic Learning Automata has been suggested as the load balancing function in which approximately the same quality level is provided for each subscriber. This framework can also be effectively extended to cloud-based systems, where adaptive approaches are needed due to unpredictability of total accessible resources, considering cooperative nature of cloud environments. The results demonstrate that the function of mobility robustness optimization not only learns to optimize HO performance, but also it learns how to distribute excess load throughout the network. The experimental results demonstrate that the proposed scheme minimizes the number of unsatisfied subscribers (Nus) by moving some of the edge users served by overloaded cells towards one or more adjacent target cells. This solution can also guarantee a more balanced network using cell load sharing approach in addition to increase cell throughput outperform the current schemes.

Mobility robustness optimization and load balancing in self-organized cellular networks: Towards cognitive network management

Mobility robustness optimization and load balancing in self-organized cellular networks: Towards cognitive network management

Low Complexity Radio Mobile Speed Estimation Using Count of Received Envelop Peaks and Troughs

Future high densification wireless networks come with high handoff rates, which require knowledge of mobile speed. Mobile speed estimation is crucial for optimizing handover to reduce call drops and network signaling flow, optimize traffic scheduling, improve quality of service, achieve resource optimization, mobility load balancing, channel quality feedback enhancement, and energy efficiency. In this paper, we present a low complexity mobile speed estimation model using count of peaks and troughs of the received signal envelop. We simulated the model in Matlab® and our result shows that the model has a maximum error of 0.25 m/s. The model has two advantages. First, it does not require measurement of the received signal power; it only counts envelop peaks and troughs. Second, the model is independent of DC offset inherent in the radio receivers. However, the model has one limitation—it does not give the crossing component of a mobile’s velocity.

Cluster-Based Load Balancing Algorithm for Ultra-Dense Heterogeneous Networks

In a highly dense heterogeneous cellular network, the loads across cells are uneven due to random deployment of cells and the mobility of user equipments (UEs). Such unbalanced loads result in performance degradation such as throughput and handover success. In order to solve the uneven load problem for better network performance, we propose a cluster-based mobility load-balancing algorithm for heterogeneous cellular networks. Traditional mobility load balancing (MLB) schemes that consider only the adjacent neighbors cannot provide enough improvement in network performance. On the other hand, the previous MLB schemes consider neighbors in the entire network suffer from unnecessary MLB actions. However, in the load balancing process, the proposed algorithm considers overloaded cells and their neighbors within then-tiers. First, the algorithm models the network as a directed multi-graph and constructs clusters taking the overloaded cells and their n-tier neighbors. Therefore, by adjusting cell individual offset parameters of the cells in the clusters the algorithm achieves load balancing locally. Since load balancing is performed inside the clusters, the network can be optimized more efficiently by avoiding unnecessary MLB actions. Simulations show that the proposed algorithm distributes the load across the network more evenly than other MLB algorithms, and in a low UE velocity scenario, it increases the overall network throughput by 6.42% compared to a non-optimized network without an MLB algorithm.

A Self-Tuning Algorithm for Optimal QoE-Driven Traffic Steering in LTE

Due to the wide diversity of services in mobile networks, cellular operators have changed their focus from Quality of Service (QoS) to Quality of Experience (QoE). To manage this change, Self-Organizing Networks (SON) techniques have been developed to automate network management, with traffic steering as a key use case. Traditionally, traffic steering aims to balance traffic volume or load among adjacent cells. Although more advanced schemes have been devised to balance QoE among cells, these do not guarantee that the overall system QoE is improved. In this work, a novel self-tuning algorithm for parameters in a classical mobility load balancing scheme is proposed to steer traffic among adjacent cells in a Long-Term Evolution (LTE) network driven by QoE criteria. Unlike previous approaches, based on heuristic rules, the proposed algorithm takes a gradient ascent approach to ensure that parameter changes always improve the overall system QoE. For this purpose, the impact of parameter changes on system QoE is estimated with an analytical network performance model that can be adjusted with statistics taken from the real network. The proposed algorithm is tested in a system-level simulator implementing a realistic LTE scenario. Results show that the method outperforms classical load and QoE mobility load balancing schemes.

BO_CEM: A novel CEM platform

SummaryIn this paper, we present our novel platform called Bo_CEM. It is composed of three levels: QoE Data Sources, QoE Monitoring System, and QoE Management System. The QoE Data Sources is an umbrella retrieving information from all sources. The monitoring systems are based on fuzzy logic in order to design a novel unique value called QoE score which is our main contribution as it uses customer profile indicators in addition to network KPIs. Finally, the QoE Management System is a high level of decision using the QoE Score as decision metric. The utility of the model is validated via three cases: The first one consists on the enhancement of the troubleshooting prioritization process for NOC Teams. The use of our new approach leads to increasing the operator's profit while customers still satisfied by the expected QoS. The second used case presents a new approach for SON Mobility Load Balancing. It allows the enhancement of the traffic load over network and increases the average QoE_score per e‐NodeB. In the last case, our novel metric called QoE_Score is used to evaluate the QoE perceived by Wi‐Fi customers in order to alleviate the LTE‐A network and route traffic to Wi‐Fi network.

Load Balancing for Ultradense Networks: A Deep Reinforcement Learning-Based Approach

In this paper, we propose a deep reinforcement learning (DRL) based mobility load balancing (MLB) algorithm along with a two-layer architecture to solve the large-scale load balancing problem for ultra-dense networks (UDNs). Our contribution is three-fold. First, this work proposes a two-layer architecture to solve the large-scale load balancing problem in a self-organized manner. The proposed architecture can alleviate the global traffic variations by dynamically grouping small cells into self-organized clusters according to their historical loads, and further adapt to local traffic variations through intra-cluster load balancing afterwards. Second, for the intra-cluster load balancing, this paper proposes an off-policy DRL-based MLB algorithm to autonomously learn the optimal MLB policy under an asynchronous parallel learning framework, without any prior knowledge assumed over the underlying UDN environments. Moreover, the algorithm enables joint exploration with multiple behavior policies, such that the traditional MLB methods can be used to guide the learning process thereby improving the learning efficiency and stability. Third, this work proposes an offline-evaluation based safeguard mechanism to ensure that the online system can always operate with the optimal and well-trained MLB policy, which not only stabilizes the online performance but also enables the exploration beyond current policies to make full use of machine learning in a safe way. Empirical results verify that the proposed framework outperforms the existing MLB methods in general UDN environments featured with irregular network topologies, coupled interferences, and random user movements, in terms of the load balancing performance.

A QoE-Driven Traffic Steering Algorithm for LTE Networks

Due to the diversity of mobile services and raising user expectations, mobile network management has changed its focus from Quality of Service (QoS) to Quality of Experience (QoE). As a consequence, classical network optimization procedures must be updated accordingly. One of these procedures is traffic sharing, whose aim is to redistribute traffic among adjacent cells so as to provide an adequate QoE to subscribers. In this work, a novel QoE-driven traffic sharing algorithm based on mobility load balancing is proposed for LTE networks offering services of very different nature. Unlike previous approaches, where the aim was to balance some QoS indicator, the aim here is to equalize the QoE provided by all cells in the network. For this purpose, the handover margins between adjacent cells are tuned on a per-adjacency or per-service basis based on QoE measurements collected in the network management system. Method assessment is based on a dynamic system-level simulator implementing a realistic LTE scenario. Results show that the proposed QoE-driven traffic sharing algorithm alleviates QoE problems by equalizing user QoE throughout the scenario.

Assessment of Deep Learning Methodology for Self-Organizing 5G Networks

In this paper, we present an auto-encoder-based machine learning framework for self organizing networks (SON). Traditional machine learning approaches, for example, K Nearest Neighbor, lack the ability to be precisely predictive. Therefore, they can not be extended for sequential data in the true sense because they require a batch of data to be trained on. In this work, we explore artificial neural network-based approaches like the autoencoders (AE) and propose a framework. The proposed framework provides an advantage over traditional machine learning approaches in terms of accuracy and the capability to be extended with other methods. The paper provides an assessment of the application of autoencoders (AE) for cell outage detection. First, we briefly introduce deep learning (DL) and also shed light on why it is a promising technique to make self organizing networks intelligent, cognitive, and intuitive so that they behave as fully self-configured, self-optimized, and self-healed cellular networks. The concept of SON is then explained with applications of intrusion detection and mobility load balancing. Our empirical study presents a framework for cell outage detection based on an autoencoder using simulated data obtained from a SON simulator. Finally, we provide a comparative analysis of the proposed framework with the existing frameworks.

Learning-Aided Multiple Time-Scale SON Function Coordination in Ultra-Dense Small-Cell Networks

To satisfy the high requirements on operation efficiency in the 5G network, self-organizing network (SON) is envisioned to reduce the network operating complexity and costs by providing SON functions, which can optimize the network autonomously. However, different SON functions have different time scales and inconsistent objectives, which leads to conflicting operations and network performance degradation, raising the needs for SON coordination solutions. In this paper, we devise a multiple time-scale coordination management scheme (MTCS) for densely deployed SONs, considering the specific time scales of different SON functions. Specifically, we propose a novel analytical model named ${M}$ time-scale Markov decision process, where SON decisions made in each time-scale consider the impacts of SON decisions in other ${M}-{1}$ time scales on the network. Furthermore, in order to manage the network more autonomously and efficiently, a Q-learning algorithm for SON functions in the proposed MTCS scheme is proposed to achieve a stable control policy by learning from history experience. To improve energy efficiency, we then evaluate the proposed MTCS scheme with two functions of mobility load balancing and energy saving management with designed network utility. The simulation results show that the proposed SON coordination scheme significantly improves the network utility with different quality of experience requirements while guaranteeing stable operations in wireless networks.