Modelling heterogeneous future wireless cellular networks: An analytical study for interaction of 5G femtocells and macro-cells

doi:10.1016/j.future.2020.07.049

Future Generation Computer Systems

Volume 114, January 2021, Pages 82-95

https://doi.org/10.1016/j.future.2020.07.049 Get rights and content

Highlights

•
A new solution approach is introduced for steady state solution of queuing networks.
•
New approach can incorporate large queuing capacities of two interacting systems.
•
The interaction between 5G based femtocells and macrocells are modelled.
•
Results are provided comparatively with existing solution approaches and simulation.
•
Accuracy and the efficacy of the new approach are demonstrated.

Abstract

The unprecedented growth in the number of mobile nodes, connected devices, and data traffic lead to the dense deployment of fifth generation (5G) networks. In such networks, small cells or femtocells are becoming quite popular mainly because of their features such as consistent coverage regardless of the location, transmit power adjustment, energy efficiency, and their ability of providing higher data rates compared to traditional cellular networks. Although the number of femtocells in a heterogeneous cellular network environment can be increased with an aim of satisfying high data traffic in next-generation 5G networks, higher numbers of small cell deployments may result in unnecessary, and frequent handovers. Therefore, the need for improvement and optimization of femtocell handover paradigms as well as their coexistence with other technologies should be investigated in the era of 5G cellular networks. This study proposes an analytical model which can be used to analyse integrated heterogeneous wireless cellular networks which consist femtocells. Integration of fourth and fifth generation wireless cellular systems is considered for this purpose. The state diagram of the proposed model can be represented by a two dimensional Markov Chain. A new, novel decomposition approach is employed for solving the system for state probabilities and to obtain performance measures such as mean queue length, throughput and response time. The main advantage of the new solution method is its capability of considering interaction of two systems with large queuing capacities unlike the existing matrix based solution approaches such as spectral expansion and matrix geometric methods. The proposed analytical solution approach has been validated using simulation and a system of simultaneous equations.

Graphical abstract

Introduction

Mobile wireless communications were initiated with first-generation voice-only systems, a few decades ago. Since then, the world has witnessed a steady and gradual evolution of mobile wireless communication towards the second, third, and fourth generation wireless networks. All IP based fourth generation Long-Term Evolution (LTE) networks have become a part of everyday life due to the ever increasing popularity of smart devices, new emerging multimedia applications, and exponential rise in wireless data demand [1]. Fifth generation networks are also expected to further improve the quality of service (QoS). Along with that, 5G networks are expected to integrate human-to-machine and machine-to-machine communications. Sensors, smart devices, accessories, and other devices are expected to become wireless communication entities with the advent of 5G networks [1]. This ever-increasing data demand and usage have the potential to cause severe traffic congestion in 5G networks. Heterogeneous networks where small cells are deployed within larger macro cells [2] aim to provide infrastructures in order to cope with the high volumes of traffic caused by this demand. Small cells or femtocells are able to provide better data rates for potentially smaller numbers of active users while consuming less energy when compared to the Wi-Fi access points. A typical infrastructure with femtocell deployments consists of five components; a digital subscriber line (DSL) or cable router, femtocell device, cellular tower (macrocell), mobile operator network, and ISP Internet line [3]. Although deployment of femtocells in heterogeneous networks is quite common with an attempt to satisfy the high data traffic demand, the interaction between different technologies can be critical. For example if 5G based femtocells are deployed within 4G macrocells, the heterogeneous nature of interaction in terms of system characteristics such as service capacity, queue capacities, and various handover probabilities can introduce challenges in estimation of QoS and in turn optimization of system configuration. Furthermore, frequent or unnecessary handovers have the potential to introduce problems in terms of QoS, as well as mobility management in 5G networks. Therefore, analysis and evaluation of interaction between cellular systems is necessary to study and possibly enhance the performance of such networks. This study aims to develop a model and a novel solution approach for heterogeneous cellular architectures where 5G small cell technologies are integrated effectively in macrocells.

The main motivation of this study is to present a realistic analytical model and the corresponding solution approach, which can be used for the femtocell/macrocell interaction. This interaction has recently become quite critical with the deployment of 5G based femtocells. Furthermore, femtocells are expected to be a strong candidate for the Internet of Things (IoT) paradigm realization [4]. The analytical modelling attempts for these kinds of interactions can be effectively used for performance analysis, specification of system characteristics under various conditions and traffic loads, and optimization of the system configurations such as the queuing capacities. The efficient mobility management in the proposed heterogeneous network architecture is also critical; therefore, the interactions between network layers due to the handover traffic are also analysed.

The integrated system is modelled as a two-stage open queuing network with multiple servers. A new, approximate product form analytical solution approach is proposed in this study for such networks, which can be used to evaluate and optimize the interaction between heterogeneous cellular networks. The proposed solution approach is able to handle larger state spaces compared to the existing solution approaches in the literature. This provides us an important advantage for realistic modelling of these interactions since the queuing capacities of systems considered do not suffer from the limitations introduced by the solution approaches. Since the proposed solution approach is approximate, the numerical results are validated using simulations. Please note that this study focuses on queueing models of packet forwarding and feedback performance of 4G/5G macrocell and femtocell interaction. Similar to the studies such as [5], [6], [7], and [8], where M/M/1, M/M/c, and M/M/1/K/ $\infty$ queuing abstractions are successfully employed for modelling real time systems, we mainly employed the simulations for the verification of the proposed model as well as the new solution approach presented. We have not focused on the implementation of a testbed for benchmarking studies. Experimental setups are known to be costly in terms of the financial requirements, and more importantly, in terms of the time required to run experiments. Furthermore, the results obtained through benchmarking are also known for the difficulties associated in terms of extrapolation [9]. Analytical modelling and solution approaches are essential contributions for studying the interactions similar to the ones presented in this work. The simulation results show that the approximate analytical solution approach performs very well in terms of accuracy and efficacy while it is able to consider large state spaces. The contributions of this study can be summarized as follows:

•
The interaction between 5G based femtocells and macrocells are modelled using a queuing theory based abstraction.
•
A new and novel product form solution approach is introduced for steady state solution which is able to incorporate large queuing capacities of two systems in interaction unlike the existing analytical solution approaches for similar systems.
•
Since the capacities of the two systems are not limited, numerical results are presented for the analysis of different system characteristics while the queuing capacities of systems in interaction are chosen more realistically.
•
The proposed new approximate solution approach is validated for accuracy as well as efficacy using simulation.

Section snippets

Related work

Analytical models are known to have an advantage over simulation tools and testbeds, as they are based on mathematical theory and have lower computational complexity. Such an advantage can become highly valuable when the traffic load is relatively high, or the number of user equipment is large. The computation time required for the simulation scenario of relatively complex systems can be significantly higher than that of analytical models [9]. Furthermore, through analytical modelling

System model

In this paper, an analytical model is proposed to analyse the interaction between existing LTE technologies like 4G and next-generation future technologies (5G). The primary goal of our design is to provide an abstraction for the system, where the effects of system characteristics such as the expected traffic load, the number of available channels, queuing capacities, and the handover probabilities on important QoS measures such as the mean response time, throughput and mean queue length can be

New product form solution

In order to express the effective mean arrival rate $λ_{e}$ , let us consider a simple M/M/1 queuing system with feedback. Assuming that the rate of feedback is $θ μ$ , we can calculate the effective mean arrival rate as $λ_{e} = f (λ, θ)$ , where the throughput is given as $γ = λ$ . Following this, the system turns into a simple M/M/1 queuing system with an arrival rate of $λ_{e}$ . The effective arrival rate $λ_{e}$ is derived as follows:

Iteration	Feedback, $δ$	$λ_{e}$
0		$λ$
1	$λ θ$	$λ + λ θ = λ (1 + θ)$
2	$λ (1 + θ) θ$	$λ + λ θ + λ θ^{2} =$
		$λ (1 + θ + θ^{2})$
3	$λ (1 + θ + θ^{2}) θ$	$λ + λ θ + λ θ^{2} + λ θ^{3} = λ (1 +$
		$θ + θ^{2} + θ$

System parameters

All the parameters used in this paper are taken from relevant literature. The radius of the femtocell is set to be 20 m [24] and the radius of macrocell is 1000 m [24]. The channel capacity of macrocell 4G LTE is 32 [24] and channel capacity of 5G femtocell is 1 taken from [29], [31], [36], [37] and [28]. Table 3 shows the summary of the parameters used together with the related literature. M/M/1/L system is preferred for 5G femtocell since in resource constrained systems and environments, it

Conclusion

In this paper integrated heterogeneous wireless networks are considered for performance evaluation. Existing 4G LTE and future generation 5G networks are modelled as a two stage tandem queuing networks. A new analytical solution method is proposed using the integrated network model. The proposed method can be used to analyse the quality of service measures such as throughput, response times, idle probabilities, and Mean Queue Lengths. In order to have seamless communication vertical handover

CRediT authorship contribution statement

Mahnoor Yaqoob: Conceptualization, Methodology, Software, Writing - original draft, Writing - review & editing. Orhan Gemikonakli: Conceptualization, Methodology, Writing - review & editing. Enver Ever: Conceptualization, Methodology, Software, Validation, Writing - original draft, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Mahnoor Yaqoob received the BS degree in Software Engineering from Fatima Jinnah Women University Pakistan in 2016 and MS degree in Computer Engineering from Middle East Technical University Northern Cyprus Campus in 2020. Her current research interests include wireless cellular communications, computer networks, analytical modelling, queueing theory, machine learning and artificial intelligence.

References (40)

Al-TurjmanF.
5g-enabled devices and smart-spaces in social-iot: an overview
Future Gener. Comput. Syst.
(2019)
AntonelliF. et al.
Analytical modeling of performance indices under epistemic uncertainty applied to cloud computing systems
Future Gener. Comput. Syst.
(2020)
ShenD. et al.
Stochastic modeling of dynamic right-sizing for energy-efficiency in cloud data centers
Future Gener. Comput. Syst.
(2015)
YangK.H.
Chapter 7 - stepping through finite element analysis
RenY. et al.
Modeling content transfer performance in information-centric networking
Future Gener. Comput. Syst.
(2017)
ÖzçevikM.E. et al.
End to end delay modeling of heterogeneous traffic flows in software defined 5g networks
Ad Hoc Netw.
(2017)
FischerW. et al.
The markov-modulated poisson process (mmpp) cookbook
Perform. Eval.
(1993)
AgiwalM. et al.
Next generation 5g wireless networks: A comprehensive survey
IEEE Commun. Surv. Tutor.
(2016)
Al-TurjmanF. et al.
Small cells in the forthcoming 5g/iot: Traffic modelling and deployment overview
IEEE Commun. Surv. Tutor.
(2018)
ChandrasekharV. et al.
Femtocell networks: a survey
IEEE Commun. Mag.
(2008)

MondalA. et al.

Buffer size evaluation of openflow systems in software-defined networks

IEEE Syst. J.

(2018)

Prados-GarzonJ. et al.

Modeling and dimensioning of a virtualized mme for 5g mobile networks

IEEE Trans. Veh. Technol.

(2016)

ZhaoJ. et al.

Modelling and optimization of packet forwarding performance in software-defined wan

Future Gener. Comput. Syst.

(2020)

BanksJ.

Discrete Event System Simulation

(2005)

OualiK. et al.

Modeling d2d handover management in 5g cellular networks

KirsalY. et al.

Modelling and analysis of vertical handover in highly mobile environments

J. Supercomput.

(2015)

KirsalY. et al.

A generic analytical modelling approach for performance evaluation of the handover schemes in heterogeneous environments

Wirel. Pers. Commun.

(2014)

LiuM. et al.

Cooperative fog-cloud computing enhanced by full-duplex communications

IEEE Commun. Lett.

(2018)

LiX. et al.

Optimal pricing and service selection in the mobile cloud architectures

IEEE Access

(2019)

Fernández-SegoviaJ.A. et al.

A teletraffic model for the physical downlink control channel in lte

Telecommun. Syst.

(2017)

Cited by (11)

An effective model for network selection and resource allocation in 5G heterogeneous network using hybrid heuristic-assisted multi-objective function
2024, Expert Systems with Applications
Multiple service requirements in various communication environments are fulfilled by the fifth-generation (5G) mobile communication network. Heterogeneous Networks (HetNet) have been introduced as a newly developed network structure. Conventional cellular networks have prominent limitations in terms of frequency utilization, coverage, proliferation of transmission services, and power consumption. The previously used approaches provide precise results but are time-consuming, making them tricky to apply for huge networks, and makes them complicated to solve multifaceted problems. In order to decrease complementary interference and accomplish spectrum sharing, an effective resource allocation technique is necessary. Likewise, interference produced by the potential Secondary Users (SUs) must be kept at a minimum value within the authorized network. Considering the perspective of the supplementary networks, it is vital to delegate networks to SUs in order to lower the gross contamination and manage the overall network’s quantity. An efficient method is developed on a heuristic basis in order to solve the issues regarding resource allocation in 5G networks. With the aid of every user device’s network properties, network selection, and resource allocation are considered the two major concerns that need to be attained in 5G heterogeneous network systems. To meet this requirement, the multi-objective constraints are considered with the help of the Integrated African Vultures with Grasshopper Optimization Algorithm (IAVGOA). The constraints that are taken into consideration are communication cost, interference between all the SU to the primary user, Quality of Service (QoS), and congestion control. The performance of the executed model is examined using various metrics. When compared with the existing models, the proposed method achieves impressive results in appropriately selecting the network and allocating the resources over 5G networks.
Ubiquitous learning models for 5G communication network utility maximization through utility-based service function chain deployment
2024, Computers in Human Behavior
The main problem of deploying service function chains in virtualized 5G networks is dealt with in wireless 5G communications and effective ubiquitous learning models. The aim is to ensure differentiated network performance for a wide range of services while maximizing the collaborative revenue of infrastructure operators and wireless virtual operators. To achieve this, a utility-based service function chain deployment strategy is introduced, tailored to the specific characteristics of Ubiquitous Learning based 5G-RAN-Architecture. Consideration is given to the virtual operator's maximum tolerable end-to-end latency, minimum service rate requirements, and the infrastructure operator's constraints on computing and link resources. It also looks at how different deployment scenarios for service function chains affect network performance and creates a utility model using a business framework. The ultimate objective is to optimize the collective revenue of infrastructure operators and virtual operators. The approach leverages genetic algorithms and Matlab's Linprog function for iterative problem-solving. The graph clearly shows that the SFC deployment algorithm in this study uses less infrastructure resources for Front Haul links. This implies that the algorithm successfully reduces the load on Front Haul lines, which in turn lowers the cost of SFC deployment and makes it easier to deploy more SFCs across the infrastructure. This work contributes to the evolution of 5G wireless communications and its seamless integration with ubiquitous learning models.
Handover parameter for self-optimisation in 6G mobile networks: A survey
2023, Alexandria Engineering Journal
One of the most crucial issues in mobile networks is ensuring reliable and stable connectivity during mobility. In recent years, numerous research has examined Fourth Generation (4G) and Fifth Generation (5G) mobile networks to address issues related to handover (HO) self-optimisation. Different approaches have been developed to identify the best Handover Control Parameters (HCPs) settings, including Time-To-Trigger (TTT) and Handover Margin (HOM), since managing HCP values is a key factor in determining the efficiency and accuracy of handover decisions. The purpose of this work is to address the challenges associated with HO management in Sixth Generation (6G) mobile networks, where a massive number of diverse devices, high mobility, and varying network conditions (e.g., Ultra Dense Network (UDN), Heterogeneous Network (HetNet) and Millimetre Waves (mmWaves)) are established. This, in turn, presents complex HO issues which require solutions that can adapt to different deployment settings. To provide a brief background of mobility management, the paper presents the basic HO concept, HO history, and HO procedure. Additionally, a comparison of HO in 4G to 6G is discussed to gain a better understanding of technology development and its effect on HO solutions. Furthermore, the basics of Mobility Robustness Optimisation (MRO) is explained, which involve HCP values. Previous MRO approaches have made significant advancements; however, they often lack adaptability and robustness to dynamic network conditions. By leveraging Artificial Intelligence (AI) algorithms with MRO techniques, this approach offers an improved solution to optimize handover decisions in a self-optimizing manner, thereby enhancing the overall network performance. To achieve this goal, it is necessary to analyze previous MRO and AI-integrated solutions and make comparisons between them. Additionally, the advantages and disadvantages of these solutions are considered. The contribution of this work lies in identifying the directions for HO self-optimization in 6G deployment. It demonstrates that deploying an AI-based solution would benefit future MRO deployments. This survey will aid in the analysis of mobility management challenges, particularly for the future mobile MRO self-optimization implementation in future technologies.
Performance evaluation of a mobile hotspot supporting quasi-random traffic
2023, IET Networks
An energy efficient dynamic small cell on/off switching with enhanced k-means clustering algorithm for 5G HetNets
2023, International Journal of Communication Networks and Distributed Systems
Femtocell-Based Interference Reduction Methodology for Self-organized Networks: A Green Network Prospective
2023, Lecture Notes in Electrical Engineering

View all citing articles on Scopus

Orhan Gemikonakli is an Honorary Professor in telecommunications at University of Middlesex, UK. He received his Ph.D. degree from the King’s college London, UK, and he has been visiting professor at the University of Camerino, Italy. His current research interests include computer networks, wireless communication systems, Internet of Things, wireless sensor networks, parallel computing paradigms, integrated circuits, and performance/reliability modelling. He serves in various Programme Committees, and he is a reviewer for various journals. He is a member of ACM and IEEE.

Enver Ever is an Associate Professor at Middle East Technical University Northern Cyprus Campus. He received B.Sc. degree from the Department of Computer Engineering, Eastern Mediterranean University, Cyprus, in 2002, and the M.Sc. degree in computer networks as well as the Ph.D. degree in performance evaluation of computer networks and communication systems from Middlesex University in 2004 and 2008, respectively. He was with Bradford University as a Post-Doctoral Research Associate for a year. He also served as a Lecturer/Senior Lecturer in the Computer and Communications Engineering Department, Middlesex University from 2008 to 2013. His current research interests include computer networks, wireless communication systems, cloud computing, wireless sensor networks, wireless multimedia sensor networks, integrated circuits, and performance/reliability modelling. He serves on various program committees and received the Exemplary Reviewer Award for his contributions as a reviewer.

View full text

Modelling heterogeneous future wireless cellular networks: An analytical study for interaction of 5G femtocells and macro-cells

Highlights

Abstract

Graphical abstract

Introduction

Section snippets

Related work

System model

New product form solution

System parameters

Conclusion

CRediT authorship contribution statement

Declaration of Competing Interest

Future Gener. Comput. Syst.

Future Gener. Comput. Syst.

Future Gener. Comput. Syst.

Future Gener. Comput. Syst.

Ad Hoc Netw.

Perform. Eval.

Next generation 5g wireless networks: A comprehensive survey

IEEE Commun. Surv. Tutor.

Small cells in the forthcoming 5g/iot: Traffic modelling and deployment overview

IEEE Commun. Surv. Tutor.

Femtocell networks: a survey

IEEE Commun. Mag.

Buffer size evaluation of openflow systems in software-defined networks

IEEE Syst. J.

Modeling and dimensioning of a virtualized mme for 5g mobile networks

IEEE Trans. Veh. Technol.

Modelling and optimization of packet forwarding performance in software-defined wan

Future Gener. Comput. Syst.

Discrete Event System Simulation

Modeling d2d handover management in 5g cellular networks

Modelling and analysis of vertical handover in highly mobile environments

J. Supercomput.

A generic analytical modelling approach for performance evaluation of the handover schemes in heterogeneous environments

Wirel. Pers. Commun.

Cooperative fog-cloud computing enhanced by full-duplex communications

IEEE Commun. Lett.

Optimal pricing and service selection in the mobile cloud architectures

IEEE Access

A teletraffic model for the physical downlink control channel in lte

Telecommun. Syst.