In recent years, the Internet of Things (IoT) has been studied in the context of the Metaverse to provide users with immersive cyber-virtual experiences in mixed-reality environments. This survey introduces six typical IoT applications in the Metaverse, including collaborative healthcare, education, smart city, entertainment, real estate, and socialization. In the IoT-inspired Metaverse, we also comprehensively survey four pillar technologies that enable augmented reality (AR) and virtual reality (VR), namely, responsible artificial intelligence (AI), high-speed data communications, cost-effective mobile edge computing (MEC), and digital twins. According to the physical-world demands, we outline the current industrial efforts and seven key requirements for building the IoT-inspired Metaverse: immersion, variety, economy, civility, interactivity, authenticity, and independence. In addition, this survey describes the open issues in the IoT-inspired Metaverse, which need to be addressed to eventually achieve the convergence of physical and cyber worlds.

AS ONE of the new communication scenarios in 5th-generation (5G) mobile communication systems, ultra-reliable and low-latency communications (URLLC) have stringent requirements on latency (around 1 ms) and reliability (up to 99.99999%). Nevertheless, existing 5G URLLC alone cannot fulfill all the Key Performance Indicators (KPIs) in emerging mission-critical applications like industrial automation, intelligent transportation, telemedicine, Tactile Internet, and Virtual/Augmented Reality (VR/AR). The 6th generation (6G) communication systems need to meet additional requirements on some of the following KPIs in combination with URLLC: high spectrum efficiency (SE)/throughput/energy efficiency (EE)/network availability/security as well as low Age of Information (AoI)/jitter/round-trip delay. These new requirements pose unprecedented challenges in terms of design methodologies and enabling technologies in 6G. To fill the gap between 5G URLLC and the diverse KPI requirements of the neXt generation URLLC (xURLLC), novel methodologies and innovative technologies are much needed.

Mobile networks have to cope with an ever increasing demand for video streaming, an application that imposes high quality requirements for user satisfaction. In this paper, we present an analytical model to calculate two important quality measures for streaming traffic, namely the video startup delay distribution and the buffer starvation probability. The queuing-theoretic model differs from related work by incorporating data flow dynamics of the considered cell, as well as the dynamics of fluctuating interference from neighboring cells, with the goal of accurately representing a multi-cellular environment. In this regard, we propose a finite-volume method to approximate the solution of the involved system of partial differential equations, where other approaches from comparable work were faced with numerical problems. We also evaluate two simplified versions of the model, where only the interference dynamics or all coupling terms are omitted, respectively. The model allows us to study the impact of different parameters on buffered video streaming performance. Additionally, we propose a user-centric metric to measure quality of experience (QoE). To the best of our knowledge, our approach is novel and has not been covered by comparable work. The presented results can help to design future cellular networks with enhanced video streaming experience.

We propose a mechanism for distributed resource management and interference mitigation in wireless networks using multi-agent deep reinforcement learning (RL). We equip each transmitter in the network with a deep RL agent that receives delayed observations from its associated users, while also exchanging observations with its neighboring agents, and decides on which user to serve and what transmit power to use at each scheduling interval. Our proposed framework enables agents to make decisions simultaneously and in a distributed manner, unaware of the concurrent decisions of other agents. Moreover, our design of the agents' observation and action spaces is scalable, in the sense that an agent trained on a scenario with a specific number of transmitters and users can be applied to scenarios with different numbers of transmitters and/or users. Simulation results demonstrate the superiority of our proposed approach compared to decentralized baselines in terms of the tradeoff between average and 5thpercentile user rates, while achieving performance close to, and even in certain cases outperforming, that of a centralized information-theoretic baseline. We also show that our trained agents are robust and maintain their performance gains when experiencing mismatches between train and test deployments.

The term Tactile Internet broadly refers to a communication network that is capable of delivering control, touch, and sensing/actuation information in real-time. The Tactile Internet is currently a topic of interest for various standardization bodies. The emerging IEEE P1918.1 standards working group is focusing on defining a framework for the Tactile Internet. The main objective of this work is to present the IEEE P1918.1 reference architecture framework for the Tactile Internet. The paper provides an indepth survey of various architectural aspects including the key entities, the interfaces, the functional capabilities, and the protocol stack. A case study has been presented as a manifestation of the architecture. Performance evaluation demonstrates the impact of functional capabilities and the underlying enablers on user-level utility pertaining to a generic Tactile Internet application.

In future 5G systems, it is envisioned that the physical resources of a single network will be dynamically shared between the virtual end-to-end networks called “slices” and the network is “sliced”. The dynamic sharing of resources can bring about pooling gains, but different slices can easily influence each other. Focusing on slicing the radio access network, a slice management entity is required to steer the radio resource management (RRM) so that all of the slices are satisfied and negative inter-slice influences are minimized. The steering of RRM can be done by adjusting slice-specific control parameters in scheduler and admission controller mechanisms. We use a model-free reinforcement learning (RL) framework and train an agent as a slice manager. Simulation results show that such agents are capable of relatively quickly learning how to steer the RRM. Furthermore, a hybrid method of Jacobian-matrix approximation with RL approach has been devised and shown to be a practical and efficient solution.

Increased usage of wireless technologies in unlicensed frequency bands inevitably increases the co-channel interference. Hence, for applications such as ultra-reliable-low-latency-communications (URLLC) in factory automation, the interference should be avoided. An intelligent coexistence management entity, which dynamically distributes the time and frequency resources, has been shown to be greatly beneficial in boosting efficiency and avoiding crippling interruptions of the wireless medium. This entity also supports multi-connectivity schemes, which are crucial for industry-level reliability requirements. The proposed governing technique of the coexistence management is a deep reinforcement learning (DRL) method, which is a model-free framework and channel allocation decisions are learned merely by interactions with the environment. The simulation results have shown that the employed method can greatly increase the reliability of the wireless network, when compared with legacy methods.

Ultra-Reliable Low-Latency Communications (URLLC) are considered as one of the key services of the upcoming fifth generation (5G) of wireless communications systems. Enabling URLLC is especially challenging due to the strict requirements in terms of latency and reliability. Multi-connectivity is a powerful approach to increase reliability. However, most of the current research is restricted to single-user scenarios, neglecting the challenges of multi-cellular, multi-user systems, i.e., interference and the competition for limited resources. In this article, we develop analytic comparisons of different connectivity approaches, showing that multi-connectivity may not always be optimal in the considered scenario. Moreover, we propose and evaluate novel resource allocation approaches based on stable matching theory to enable wireless URLLC. We extend the pure many-to-one stable matching procedure by utilizing the optimal connectivity approach for each user, optimizing the maximum number of matched resources, and providing a resource reservation mechanism for users suffering from bad channel conditions. System-level simulations demonstrate that the proposed algorithm outperforms baseline resource allocation approaches in outage probability by up to three orders of magnitude. Even in a highly loaded system, an outage probability in the range of ${10^{-5}}$ is achieved.

With the fifth generation (5G) of mobile networks the challenging field of ultra-reliable low latency communications (URLLC) is envisioned. In this area of research, reliability requirements are commonly formulated only in terms of packet loss rates. However, sparse packet loss can usually be tolerated by an application. Furthermore, the terms reliability and availability are usually loosely defined and not sufficiently separated in the wireless communications field. This paper applies fundamental metrics from dependability theory to wireless communications, providing more powerful key performance indicators (KPIs) for URLLC. In particular, we derive analytic expressions which reflect the up- and downtime distributions as well as the mission availability in a wireless multi-connectivity scenario with selection combining and Rayleigh fading. Further, we propose an appropriate approximation for the computationally complex expression of the mission availability.

In the context of 5G systems, the emergence of various use cases with diverse requirements has attracted great attention to network slicing. In a single physical network, several instances of logical end-to-end networks, i.e. slices, will be instantiated to fulfill these requirements. To this end, slices should share the resources of the physical network, which consist of Core Network (CN) and Radio Access Network (RAN) resources. Herein, we focus on the Radio Resource Management (RRM) in the context of network slicing. To maximize the pooling gains, dynamic resource sharing is preferred over static sharing. However, dynamic resource sharing can lead to undesirable inter-slice influences, in particular, the contention on radio resources. In this article, we show that, although the radio resources are dynamically shared among the users of different slices, proper slice management can realize slice protection. This is achieved by adjusting the fraction of radio resources allocated to the different slices by the Packet Scheduler (PS) and by limiting the number of users admitted to the network via Admission Control (AC). We propose an iterative algorithm to optimize the parameters of PS and AC in order to ensure that the service level agreements are satisfied. Extensive system-level simulations have shown that a central entity that tunes these control parameters can greatly increase the network's performance.

Fifth generation (5G) of mobile networks are expected to serve multiple heterogeneous use cases. These use cases are extremely diverse in terms of service requirements and bundling them in a single monolithic network is a challenge. Network slicing is identified as one of the main enablers of 5G systems, where multiple logical End-to-End (E2E) networks share the resources of a single physical network. Radio Resource Management (RRM) in a sliced network should be able to simultaneously fulfill the required services of slices, dynamically share the network and assure the independence of slices so that slices cannot affect each other negatively. In this paper, we study the existing mechanisms that provide similar features in legacy networks and demonstrate the contributions and shortcomings of such existing RRM mechanisms in a sliced network. Thereafter, we argue the need for a new entity that will complement the existing RRM mechanisms to be slice-aware. With the aid of system-level simulations, we compare different slicing schemes and illustrate the drawbacks of legacy networks in fulfilling the objectives of a fully sliced network. Moreover, we illustrate the capabilities of the slice-aware RRM in steering the network's Key Performance Indicators (KPIs).

Mobile communication continues to play an important role in the modern economy, including consumer, health, education, logistics, and other major industries. At the same time, the current Internet has created a key infrastructure component for our modern world, having an impact on almost every aspect of our daily lives. The Internet democratized access to information and has enabled emerging economies to participate in the modern global economy. We are now approaching the next big wave of the Internet innovation: the Tactile Internet. The widely used term “Tactile Internet” was coined and defined by the IEEE P1918.1 as: “A network or network of networks for remotely accessing, perceiving, manipulating or controlling real or virtual objects or processes in perceived real time by humans or machines.”

Wireless communication systems have been evolving since the first generation. With the fifth generation of wireless systems, not only the evolutionary aspect of increased data rates is tackled but also the revolutionary aspect. Here, emerging use cases such as massive machine-type communication and ultrareliable low-latency communication will play a crucial role. Within this context, applications with stringent latency and reliability requirements are emerging. Wireless reliability is understood as successfully transmitting the desired amount of data within a given time. Diversity techniques, such as multiconnectivity, are potential solutions to achieve stringent reliability requirements. However, in a multiuser scenario, in which resources are shared, this might not always be possible. In this paper, we discuss the feasibility of various multiconnectivity approaches and propose a matching theory-based algorithm together with a novel scheduler aiming to guarantee the reliability requirements of as many users as possible in a multicellular, multiuser system. System-level simulations demonstrate that the proposed approach achieves 100% reliability for the fifth-percentile users in a highly loaded system. The maximum gain of fifth-percentile user throughput as compared to a static multiconnectivity approach is 150%.

The fifth-generation (5G) of mobile networks are supposed to address radically new types of applications that have not been possible with previous generations of wireless technologies. Ultra-Reliable Low-Latency Communication (URLLC) is one 5G use case category imposing strict requirements on the availability, reliability, and latency together with zero mobility interruption. Aiming for a common understanding, we put forward a comprehensive description of performance metrics based on reliability theory focusing on availability and reliability metrics as a function of time. We establish a link to wireless communications by studying Rayleigh fading channels in a multi-connectivity system. The reliability metric definitions and analytic models presented in this paper contribute to enable a systematic and accurate performance analysis for URLLC. Numerical evaluations demonstrate the importance of considering not only availability alone but combinations of availability and time-based reliability statistics jointly.

In the context of 5G mobile networks, several new use cases with various requirements with respect to throughput, latency, coverage, etc., should be addressed. To avoid deployment of separate networks for each of the use cases, the concept of network slicing has been introduced, where several logical networks share a single physical network. However, the accommodation of networks with diverse requirements in a single physical network is a new challenge. In this work, we study the effects of a mapping layer, which supervises the network over a service area and manages the allocation of radio resources to slices to guarantee their target service requirements. To do so, we propose an adaptation algorithm based on minimizing deviations from slice requirements. The results show that by utilizing the mapping layer, the resources can be shared efficiently and fairly and the deviations of Key Performance Indicators (KPIs) from the Service Level Agreement (SLA) targets are reduced compared to distributed control methods that are typically used in legacy and current cellular systems.

Besides classical performance indicators such as throughput and capacity, other metrics like reliability, availability, and latency are becoming increasingly important for designing and analyzing wireless networks. This work provides novel insights into how high availability can be achieved. For doing so, a preexisting model for detailed analysis of the signal-to-interference- plus-noise ratio (SINR) is adapted and applied to various intra-frequency multi-connectivity architectures. In addition, the impact of different carrier frequencies, i.e., 2.5 GHz and 15 GHz, is investigated as well. The developed models accurately capture the system performance and reveal distinct advantages and disadvantages of the investigated architectures. It is demonstrated that, especially, dynamic clustering mechanisms in combination with single frequency network transmissions are suited to achieve extremely high availability while using resources efficiently.

Millimeter-wave frequency bands enable the deployment of small-sized antenna arrays in the fifth generation of mobile networks, which provide high beamforming gain for both transmitter and receiver. In order to study user mobility, a computationally efficient simulator is needed that runs with a time step resolution that is higher than transmission-time-interval of cellular networks. In such simulators, typically the average downlink signal-to-interference-and-noise ratio (SINR) is used for radio link failure detection and throughput calculation. In this paper, models of desired and interfering signals are formulated first, by considering the impact of antenna beamforming at transmitter and receiver. Then, a closed-form expression of average downlink SINR is derived by taking into account the scheduling probabilities of the users. In addition, the closed form expression is approximated by Monte Carlo method. Simulations are performed to analyze the complexity and accuracy of Monte Carlo method along with the impact of user location and beamforming gain on average downlink SINR. Results confirm that the average downlink SINR is approximated accurately, and the computational complexity of the proposed method is tolerable. In addition, the results reveal that user location and beamforming have significant impact on the average downlink SINR.

The Mobile Internet connects billions of smart phones and laptops. With this global connectivity, the stage is set for the emergence of: (a)Tactile Internetto deliver haptic experiences to remote users, and (b) flexible and integratedBackhaul/Fronthaulnetworks to support demands of such applications.

Enhancing the connectivity reliability is one of the most challenging requirements for the design of future wireless communications systems. The scope of this paper is to leverage the existing tool set of reliability theory for enabling reliable communication in wireless systems. Definitions, concepts, and methods of reliability theory are applied and extended to wireless communications networks, which are modeled as a repairable system. The steady-state and transient system behaviour are considered. Two new key performance indicators (KPIs) for the reliability analysis of wireless communications systems are introduced, namely mean time to first failure (MTTFF) and interval reliability (IR), and a closed form expression is derived for the MTTFF. By evaluating an exemplary scenario, the trade-off between availability, reliability and throughput is discussed.

The support of mission-critical use cases is an ambitious goal of the upcoming fifth generation of mobile networks. In addition to challenging requirements regarding reliability and latency, particular use cases need to be supported in high mobility scenarios as well. To avoid extensive system simulations and enable studies of very small outage probabilities, we integrate mobility effects into an existing signal-to-interference- plus-noise ratio (SINR) model by combining the model with results from a mobility simulation. An evaluation of the model corroborates that high mobility can severely deteriorate the reliability performance in traditional single-connectivity architectures. As potential countermeasures, inter- and intra-frequency multi-connectivity are identified. Results show that high reliability of 99.999% or greater becomes possible if a sufficient number of connections is utilized.

Capacity and ultra-reliable communication are some of the requirements for 5th generation (5G) networks. One of the candidate technologies to satisfy capacity requirement is standalone Ultra Dense Network (UDN). However, UDNs are characterized by fast change of received signal strength that creates mobility challenges in terms of increased handovers and connection failures. In this paper, a low layer multiconnectivity scheme is presented for standalone UDN aiming at ultra-reliable communication that is free of interruptions from handover procedures and connection failures. Furthermore, the problem in managing of the set of serving cells, that are involved in multiconnectivity for each user, is formulated. By using numerical method, feasible scheme for management of the set of serving cells is derived. Performance of the proposed multiconnectivity scheme is evaluated and compared against single connectivity. It is shown that the proposed multiconnectivity scheme outperforms single connectivity considerably in terms of connection failures and cell-edge throughput.

Multi-connectivity is a promising concept for addressing challenging requirements in next generation wireless networks. We put forward a modeling framework for analyzing signal-to-interference-and-noise ratio (SINR) distributions in inter-frequency multi-connectivity scenarios. The most important features are a best server association based on random shadowing, multiple path loss models, and intra-/inter-frequency shadowing cross-correlation. Furthermore, we consider diverse antenna types, such as sectorized antennas and antenna arrays with beamforming, to accurately model the distinct properties of conventional as well as upcoming millimeter wave carrier frequencies. In the analysis, we focus on the lower tail of the SINR distributions in order to explore the availability performance. The modeling results, which are corroborated by simulations, show that certain combinations of carrier frequencies can significantly improve the availability as well as the throughput performance compared to single-frequency usage.

5thGeneration (5G) mobile networks are required to support transmission of capacity demanding services such as real-time remote computing without any interruption. One of the candidate solution for high capacity is Ultra Dense Networks (UDNs). However, UDNs are characterized by fast change of the received signal by a user. The fast change of the signal and high speed of users create too many handovers and connection failures such as handover failures and Radio Link Failures (RLFs). Consequently, conventional handovers and connection failures are the major sources of service interruption. To achieve ultra-reliable communication by tackling the service interruptions, this paper proposes a novel multi-connectivity scheme that uses fast selection of serving cell from a set of prepared cells. A similar feature in Long Term Evolution - Advanced (LTE-A) that is defined under Coordinated Multi-Point (CoMP) transmission is Dynamic Point Selection (DPS). However, in DPS a cell is selected dynamically for transmission of only data signals. Transmission of mobility related control signals is performed through one primary cell which is changed through a conventional handover. Unlike DPS, this paper proposes that the selected cell, from the set of prepared cells, is used for transmission of both data and control signals. Simulation results show that the connection failures due to RLFs are considerably resolved by the proposed scheme.

High reliability and low latency are key requirements for a number of upcoming use cases and applications in mobile communication networks. In order to address the requirements, we present statistical models which capture the most important radio propagation phenomena, such as path loss, shadowing, and small scale fading. The models enable the evaluation of trade-offs between reliability, achievable rate, latency, and bandwidth. Additionally, extensions for joint channel coding across multiple frequency layers are presented. Numerical evaluations reveal that the antenna configuration has a significant impact on the reliability performance. Further, considerable performance gains of joint coding across multiple frequency layers are observed.

Powered by next generation mobile networking capabilities, the Tactile Internet will be able to transport touch and actuation in real-time. Enabled by suitable robotics and haptics equipment at the edges, and an unprecedented communications network, the Tactile Internet will provide a true paradigm in creating skill-set delivery networks. The fifth generation (5G) mobile communications systems will underpin this emerging Internet at the wireless edge. This paper presents the most important technology concepts which lay at the intersection of the larger Tactile Internet and the emerging 5G systems. Specifically, the paper presents some of the most important Tactile Internet applications, outlines the key technical requirements, and covers end-to-end architectural aspects of the Tactile Internet.

Nowadays telecommunication systems consume a huge amount of energy. While energy efficiency methods have already been studied for linear antennas, this paper focuses on 2D antenna arrays. An antenna array with many elements gives the network operator a high degree of flexibility in network optimization, which can be used not only for improving the quality of service, but also for increasing the energy efficiency. The energy efficiency can be improved by adapting the array shape, i.e., by switching off certain elements of the array. This work presents three novel solutions for the energy efficiency optimization, which are based on Q-learning, biological neural networks, and random drop, respectively. Simulation results demonstrate that significant improvements in terms of energy efficiency can be achieved, while the 5thpercentile of user throughput and coverage performances decrease only marginally.

Improved throughput and reliable communication that is free of mobility failures are some of the requirements for 5thGeneration (5G) mobile networks. One of the cost-effective solutions to meet capacity requirement is standalone ultra dense network that use the same spectrum and a cloud radio access. However, stand alone ultra dense networks are prone to mobility challenges. Previous work proposed and evaluated multi-connectivity that co-ordinates transmission on both the user plane and control plane; this enables improvement not only in throughput but also in mobility robustness. One of the major component of multi-connectivity is managing the “active set” which is the set of co-ordinating cells. Proper procedures for managing user-specific active set was defined for 3G Soft Handover. This paper revisits prior art schemes for active set management and proposes new adaptive one for 5G Networks. Performance evaluation is provided with the help of elaborated models for simulation. Among other things, it is shown that the adaptive active set management scheme reduces the signaling overhead associated to active set updates by around 19% compared to static active set management.

In order to cope with the wireless traffic demand explosion within the next decade, operators are underlying their macrocellular networks with low power base stations in a more dense manner. Such networks are typically referred to as heterogeneous or ultra-dense small cell networks, and their deployment entails a number of challenges in terms of backhauling, capacity provision, and dynamics in spatio-temporally fluctuating traffic load. Self-organizing network (SON) solutions have been defined to overcome these challenges. Since self-organization occurs in a plethora of biological systems, we identify the design principles of immune system self-regulation and draw analogies with respect to ultra-dense small cell networks. In particular, we develop a mathematical model of an artificial immune system (AIS) that autonomously activates or deactivates small cells in response to the local traffic demand. The main goal of the proposed AIS-based SON approach is the enhancement of energy efficiency and improvement of cell-edge throughput. As a proof of principle, system level simulations are carried out in which the bio-inspired algorithm is evaluated for various parameter settings, such as the speed of small cell activation and the delay of deactivation. Analysis using spatio-temporally varying traffic exhibiting uncertainty through geo-location demonstrates the robustness of the AIS-based SON approach proposed.

The long-term ambition of the Tactile Internet is to enable a democratization of skill, and how it is being delivered globally. An integral part of this is to be able to transmit touch in perceived real-time, which is enabled by suitable robotics and haptics equipment at the edges, along with an unprecedented communications network. The fifth generation (5G) mobile communications systems will underpin this emerging Internet at the wireless edge. This paper presents the most important technology concepts, which lay at the intersection of the larger Tactile Internet and the emerging 5G systems. The paper outlines the key technical requirements and architectural approaches for the Tactile Internet, pertaining to wireless access protocols, radio resource management aspects, next generation core networking capabilities, edge-cloud, and edge-AI capabilities. The paper also highlights the economic impact of the Tactile Internet as well as a major shift in business models for the traditional telecommunications ecosystem.

Analytical models enable accurate and quick assessment of performance metrics in network planning and system design. In wireless networks, the signal-to-interference-and-noise ratio (SINR) is of key importance since other metrics, such as throughput and capacity, strongly depend on the SINR. In this work, we characterize the SINR by a composition of log-normal random variables describing shadow fading and propose a comprehensive framework for modeling SINR distributions at specific user locations. In contrast to existing works, we include shadowing cross-correlation, noise power, and the best server policy in a single framework. Especially, the best server policy, which captures the influence of shadowing on the selection of the serving base station, is frequently neglected in analytical models. Moreover, we put forward SINR models for non-coherent joint transmission in dynamic multi-point networks. Finally, numerical evaluations show the applicability of the models but also reveal the limits of them.

Enhanced Inter-Cell Interference Co-ordination (eICIC) enables expanding the range of small cells (and thereby better Macro cell offloading) by protecting users in the expanded area through blanking subframes in the Macro cells. Range expansion is achieved by modifying the handover parameters. However, those handover parameters are already modified by the existing feature Mobility Robustness Optimization (MRO). The interaction between MRO and eICIC is particularly challenging since real deployments face a mixture of Rel. 10/11 users (which are able to perfectly exploit eICIC) and legacy Rel. 8/9 users. Furthermore, the details of the eICIC impact on mobility (failures) is not very well explored, in particular due to the lack of simulation tools capable of investigating both eICIC and MRO at the same time. In Release 11, there was a proposal for a new feature which enables proper interaction between MRO and eICIC in the presence of both Rel. 8/9 and Rel. 10/11 users. The feature allows setting up separate MRO statistics for Rel. 8/9 users and for Rel. 10/11 users. The paper presents comparison of this new feature with simpler workarounds. The results were created with a system level simulator which has elaborated mobility and eICIC model. The main conclusions are that mobility problems also exist for Rel. 10/11 users, and simple workaround solutions cannot solve the interaction problem.

Revolutionary use cases for 5G, e.g., autonomous traffic or industrial automation, confront wireless network engineering with unprecedented challenges in terms of throughput, latency, and resilience. Especially, high resilience requires solutions that offer outage probabilities around 10-6 or less, which is close to carrier-grade qualities but far below what is currently possible in 3G and 4G networks. In this context, multi-connectivity is understood as a promising architecture for achieving such high resilience in 5G. In this paper, we analyze an elementary multi-connectivity solution, which utilizes macro-as well as microdiversity, and evaluate trade-offs between power consumption, link usage, and outage probability. To elaborate, we consider exponential path loss, log-normal shadowing, shadowing cross-correlation, and Nakagami-m small scale fading, and derive analytical models for the outage probability. An evaluation of the multi-connectivity system in a hexagonal cellular deployment reveals that optimal operating points with respect to the number of links and resources exist. Moreover, typical 5G aspects, e.g., frequent line of sight in dense networks and multiple antenna branches, are shown to have a beneficial impact (fewer links needed, more power saved) on ideal operating points and overall utility of multi-connectivity.

The use of small cell deployments in heterogeneous network (HetNet) environments is expected to be a key feature of 4G networks and beyond, and essential for providing higher user throughput and cell-edge coverage. However, due to different coverage sizes of macro and pico base stations (BSs), such a paradigm shift introduces additional requirements and challenges in dense networks. Among these challenges is the handover performance of user equipment (UEs), which will be impacted especially when high velocity UEs traverse picocells. In this paper, we propose a coordination-based and context-aware mobility management (MM) procedure for small cell networks using tools from reinforcement learning. Here, macro and pico BSs jointly learn their long-term traffic loads and optimal cell range expansion, and schedule their UEs based on their velocities and historical rates (exchanged among tiers). The proposed approach is shown to not only outperform the classical MM in terms of UE throughput, but also to enable better fairness. In average, a gain of up to 80% is achieved for UE throughput, while the handover failure probability is reduced up to a factor of three by the proposed learning based MM approaches.

Device-to-device (D2D) communication enables mobile devices to directly communicate with each other without the help of infrastructure and to reuse radio resources within cellular networks. When D2D user equipments (UEs) use the same radio resources for discovery and communication as the cellular network, intercell interference (ICI) may degrade the D2D discovery performance. Therefore, ICI between the cellular network and D2D UEs should be coordinated to avoid performance degradation. In this paper, we evaluate the benefits of time-domain intercell interference coordination (ICIC) approaches in D2D discovery by using almost blank subframes (ABS). Also, we present the benefits of using multi-hop discovery.

Handover is one of the most critical functions in a cellular network and it supports the mobility of user equipments (UEs) across multiple cells. It is usually based on signal measurements from candidate base stations and filtering the signal measurements becomes necessary due to channel imperfections. In LTE, handover measurements are filtered using layer 3 (L3) filters in the UE, whereby the UE checks for the handover entry condition at discrete sampling instants of the L3 filter. The selection of sampling period becomes crucial in improving the handover performance of the network. Therefore, it is necessary to understand the trade-offs related to the sampling period of an L3 filter. The goal of this paper is to derive the handover failure (HF) expression as a function of L3 sampling periods in LTE. A geometric handover model is used for derivations and all analytical expressions are verified through simulations.

As future small cell base stations (SCBSs) are set to be multi-mode capable (i.e., transmitting on both licensed and unlicensed bands), a cost-effective integration of both technologies coping with peak data demands is crucial. Using tools from reinforcement learning, a distributed cross-system traffic steering framework is proposed whereby SCBSs leverage WiFi, to autonomously optimize their long-term performance over the licensed spectrum band, as a function of the traffic load and users' heterogeneous Quality of Service (QoS) requirements. The proposed traffic steering solution is validated in a Long-Term Evolution (LTE) simulator augmented with WiFi hotspots. Remarkably, it is shown that the proposed cross-system learning-based approach outperforms several benchmark algorithms and traffic steering policies, with gains reaching up to 200% when using a traffic-aware scheduler as compared to the classical proportional fair (PF) scheduler.

In this paper, we study the strategic coexistence of macro and pico cells and investigate a utility based resource allocation technique for heterogeneous networks. Aiming at finding an objective driven allocation of resources, we focus on time domain inter-cell interference coordination (ICIC) techniques. Macro and pico cells optimally select their almost blank subframe (ABS) muting rate and cell range bias values, so that inter-cell offloading and enhanced ICIC (e-ICIC) are jointly performed. A theoretical closed form solution is provided for our utility based resource allocation algorithm as well as a heuristic solution. To substantiate our theoretical findings, system level simulations are carried out in which the proposed heuristic solution is compared to a baseline ICIC technique. Interestingly, our proposed solution is shown to yield significant gains of up to 36% compared to the baseline approach.

In this paper, we investigate enhanced Inter-Cell Interference Coordination (e-ICIC) techniques for Heterogeneous Networks (HetNets), consisting of a mix of macro and picocells. We model this strategic coexistence as a multi-agent system in which decentralized interference management and cell association strategies inspired from Reinforcement Learning (RL) are devised. Specifically, we focus on time and frequency domain ICIC techniques in which picocells optimally learn their cell range bias and downlink transmit power allocation. In turn, the macrocell optimizes its transmission by serving its own users while adhering to the picocell interference constraint. To substantiate our theoretical findings, system level simulations are carried out in which our proposed solution is compared with a number of existing ICIC approaches, such as resource partitioning, fixed cell range expansion (CRE) and fixed Almost Blank Subframe (ABS). Interestingly, our proposed solution is shown to yield substantial gains of up to 125% compared to static ICIC approaches.

Autonomous interference management solutions for inter-cell interference coordination (ICIC) are of utmost importance. In this paper, the coexistence between macro and small cells is studied whereby different ICIC techniques pertaining to different deployment and information assumptions are evaluated. Inspired from Evolutionary Game Theory (EGT), decentralized strategies are devised, in which small cell Base Stations (BSs) exchange information through a central controller, and adapt their strategies based on instantaneous and average payoffs of the small cell population. In contrast, when distributed operation is aimed at, using tools from Reinforcement Learning (RL) small cells learn by interacting with their environment through trials and-errors, and autonomously optimize their strategies based on a mere feedback. In particular, we compare the performance of decentralized Q- learning, Fuzzy Q-learning, improved Q-learning and expertness-based Q-learning procedures. Finally, the overall performance of the network in terms of average peruser data throughput and convergence are carried out in an LTE-A system level simulator.

We investigate the resource allocation problem for the downlink of a spatial multiplexing (SM) based multiple input multiple output orthogonal frequency division multiple access (MIMO-OFDMA) based network. We address this problem in the context of the utility-based resource allocation presented in earlier papers in order to enhance the efficiency and guarantee fairness among users by exploiting multiuser diversity, frequency diversity, as well as time diversity. We design utility functions with respect to the average rate in order to simultaneously provide two services, real-time and best-effort. We propose a low-complexity joint subchannel and power control algorithm that converges very fast. Simulation results indicate that using the proposed method achieves near-optimum solutions and the available resources are distributed more fairly among users.