Advanced Multicarrier Technologies for Future Radio Communication: 5G and Beyond

By Hanna Bogucka, Adrian Kliks and Pawel Kryszkiewicz

A practical review of state-of-the-art non-contiguous multicarrier technologies that are revolutionizing how data is transmitted, received, and processed

This book addresses the advantages and the limitations of modern multicarrier technologies and how to meet the challenges they pose using non-contiguous multicarrier technologies and novel algorithms that enhance spectral efficiency, interference robustness, and reception performance. It explores techniques using non-contiguous subcarriers which allow for flexible spectrum aggregation while achieving high spectral efficiency and flexible transmission and reception at lower OSI layers. These include non-contiguous orthogonal frequency division multiplexing (NC-OFDM), its enhanced version, non-contiguous filter-bank-based multicarrier (NC-FBMC), and generalized multicarrier. 

Following an overview of current multicarrier technologies for radio communication, the authors examine particular properties of these technologies that allow for more efficient usage within key directions of 5G. They examine the principles of NC-OFDM and discuss efficient transmitter and receiver design. They present the principles of FBMC modulation and discuss key challenges for FBMC communications while comparing performance results with traditional OFDM. They move on from there to a fascinating discussion of GMC modulation within which they clearly demonstrate how that technology encompasses all of the advantages of previously discussed techniques, as well as all imaginable multi- and single-carrier waveforms. 

• Addresses the problems and limitations of current multicarrier technologies (OFDM)
• Describes innovative techniques using non-contiguous multicarrier waveforms as well as filter-band based and generalized multicarrier waveforms
• Provides a thorough review of the practical limitations and solutions for evolving and breakthrough 5G communication technologies
• Explores the future outlook for non-contiguous multicarrier technologies as regards their greater industrial realization, hardware practicality, and other challenges

Advanced Multicarrier Technologies for Future Radio Communication: 5G and Beyondis an indispensable working resource fortelecommunication engineers, researchers and academics, as well as graduate and post-graduate students of telecommunications. At the same time, it provides a fascinating look at the shape of things to come for telecommunication industry executives, telecom operators, regulators, policy makers, and economists. 

• Language: English
• Publisher: Wiley
• Release date: Jul 12, 2017
• ISBN: 9781119168928

Book preview

Preface

Increasing demand by mobile radio customers (persons and devices) for higher data rates, multimedia services, and more bandwidth, as well as anticipated traffic related to the Internet of Things, creates unprecedented challenges for future mobile communication systems. There seems to be the general consensus on the future Fifth Generation (5G) wireless communication directions and expected key performance indicators to meet these challenges, that is, to aim at achieving significantly higher system capacity, connectivity, energy and spectral efficiencies, while lowering the end-to-end latency for some mission-critical applications. Concerning the network capacity and spectrum usage enhancement, they result from network densification and spectrum aggregation [1]. Spectrum aggregation refers to making use of possibly discontinuous frequency bands and, thus, larger amounts of electromagnetic spectrum. It is known to be possible through a technique called Carrier Aggregation (CA), which has been proposed for the Long Term Evolution Advanced (LTE-A) standard in order to achieve the throughput of 1 Gbps in the downlink for the Fourth Generation (4G) systems in a 20 MHz channel [2]. Although CA applied in LTE-A is a step toward spectrum aggregation, its flexibility in aggregating any kind of spectrum fragments is limited, and the proposed protocols do not allow for dynamic spectrum access and aggregation.

New multicarrier transmission techniques using noncontiguous subcarriers are known to be capable of flexible spectrum aggregation [3, 4] and allow for flexibility of various kinds, specially at adaptive physical and medium access control layers. By applying cognitive spectrum sharing using these techniques in both licensed and unlicensed frequency bands of the future heterogeneous networks, more spectrum can be effectively used, and interference among cells and nodes can be avoided. Dynamic aggregation of potentially noncontiguous fragments of bands in a wide frequency range poses a number of challenges for the baseband processing, antenna and Radio Frequency (RF) transceiver design, particularly in the dynamically changing radio environment. In our book, we present these promising technologies and answer how to meet the mentioned 5G challenges with noncontiguous multicarrier technologies and novel algorithms enhancing spectral efficiency, interference robustness, and reception performance. It is apparent that the deployment of future flexible radios and spectrally agile waveforms has received and is still receiving the necessary scientific recognition.

Multicarrier modulation and multiplexing are a form of Frequency-Division Multiplexing (FDM), where data are transmitted across several narrowband streams using different carrier frequencies. The most known example is the Orthogonal Frequency-Division Multiplexing (OFDM). In the recent years, however, increasing research effort has been focused on some other forms of multicarrier modulation and multiplexing, which enhance the properties of OFDM or employ nonorthogonal subcarriers, use discontinuous frequency bands, and apply subcarrier shaping. In our book, we focus on new multicarrier transmission techniques using noncontiguous subcarriers such as PNon-contiguous Orthogonal Frequency-DivisionMultiplexing (NC-OFDM), its enhanced version, Generalized Multicarrier (GMC) multiplexing, or its special case, namely the Non-contiguous Filter-Bank Multi-Carrier (NC-FBMC) technique. These are techniques capable of flexible spectrum aggregation, flexible transmission and reception methods achieving high spectral efficiency or energy efficiency toward meeting the 5G radio system challenges. We believe that in the coming years, the work on novel multicarrier technologies will be at a height of culmination for application in future radio communication systems (5G and beyond).

In Chapter 1, we discuss the challenges and bottlenecks of the future and 5G radio communication technology based on the spectral agility of waveforms and on the flexibility and efficiency of spectrum usage. The need for practical solutions and implementation based on novel multicarrier technologies are emphasized.

Chapter 2 entitled Multicarrier technologies in radio communication systems presents the state of the art in multicarrier technologies for radio communication. We present the principles of multicarrier schemes, OFDM, as well as other known multicarrier techniques. In that chapter, we also address the key advantages and issues in designing multicarrier systems, such as nonlinear distortions, Peak-to-Average Power Ratio (PAPR) reduction techniques, transmission parameter adaptation, reception techniques, and synchronization.

In Chapter 3 on Noncontiguous OFDM for future radio communications, we introduce the principles of NC-OFDM as a well-suited technique for future 5G radio communications, able to aggregate discontinuous spectrum bands. Efficient NC-OFDM transmitter and receiver designs are discussed. Moreover, key techniques for enhanced NC-OFDM communications are addressed: reduction of the Out-of-Band (OOB) power to aggregate the fragmented spectrum and to limit and control interference generated to the adjacent frequency bands, spectrum aggregation dynamics, PAPR reduction, signal reception, and the particularly difficult problem of synchronization in the face of the reduced number of used subcarriers and possible interference from frequency-adjacent systems.

In Chapter 4 on Generalized multicarrier techniques for 5G radio, we introduce the idea of Generalized Multicarrier (GMC) modulation. It is shown that it encompasses all existing multicarrier techniques, as well as all theoretically imaginable multi- and single-carrier waveforms. Some interesting features of this flexible and generalized waveform description are discussed, showing its potential for the application in future 5G (and beyond) radio communications and flexible programmable transceivers. Moreover, key issues for GMC communications are addressed, such as higher PAPR and increased complexity of the GMC transceivers, including adaptive transmission and reception algorithms.

Chapter 5 entitled Filter-bank-based multicarrier technologies presents the principles of Filter-BankMulti-Carrier (FBMC) modulation, which has been recently heavily researched worldwide and is being proposed for some of the 5G radio interfaces. In this technique, the OOB power is filtered on the per-subcarrier basis. Efficient Offset Quadrature Amplitude Modulation (OQAM)-based FBMC transmitter and receiver design with reduced computational complexity is discussed. The prototype-filter design and related receiver techniques are addressed. Other challenges of this technique are also covered, as well as other filter-bank-based techniques recently proposed: filtered OFDM, Cosine-Modulated Multitone signaling, Filtered Multi-Tone (FMT), Universal Filtered Multicarrier (UFMC), or Generalized Frequency Division Multiplexing (GFDM).

Chapter 6 on Multicarrier technologies for flexible spectrum usage discusses Dynamic Spectrum Access (DSA) and sharing options for the future multicarrier technologies meeting the desired features of 5G communications. Some interesting DSA methods based on game theory, spectrum pricing, and the so-called coopetition are discussed. The issue of the required information signaling is confronted against the required spectral efficiency. Coexistence of the new cognitive radio technologies with the incumbent (licensed) systems is considered. In particular, spectrum aggregation using NC-OFDM and NC-FBMC in the real-world scenarios in the presence of Global System for Mobile Communications (GSM) and Universal Mobile Telecommunications System (UMTS) system base stations and terminals is discussed and evaluated.

Finally, the book is summarized in Chapter 7, presenting Conclusions and Future Outlook. This chapter summarizes the key observations obtained from the totality of the presented work. The chapter also includes the discussion of the future outlook for presented technologies in terms of their greater industrial realization, hardware practicality, and other challenges.

Poznań

April 10, 2017

List of Abbreviations

Chapter 1

Introduction

Intensive development of mobile radio communication systems can be observed since the deployment of cellular telephone systems, which revolutionized communication in modern society. The pioneer working cellular system was the analog First Generation (1G) Nordic Mobile Telephony (NMT) system deployed in 1981, first in Scandinavia, and then, in some European and Asian countries [5]. In the following years, there have been other analog, and then digital, Second Generation (2G) cellular systems introduced with increasing continental coverage, that is, European Global System for Mobile Communications (GSM) and American IS-95, followed by the global Third Generation (3G) systems, that is, Universal Mobile Telecommunications System (UMTS) and IMT-2000 [6]. For these systems, a number of high-data-rate transmission schemes and improvements have been implemented, leading to the Fourth Generation (4G) standards, significantly increasing the 3G-systems' capacity, coverage, and mobility. Apart from the mobile radio communication standards, the local Wireless Local Area Networks (WLANs) are constantly improving to support very high data rates, for example, the IEEE 802.11.n standard using the Multiple Input, Multiple Output (MIMO) technology allows transmission from 54 to 600 Mbit/s (at a maximum net data rate) [7].

Future wireless networks are challenged with keeping up with the constantly increasing demand by mobile devices for higher data rates, multimedia services support, and ever more bandwidth. In Figure 1.1, contemporary wireless communication standards are presented, reflecting the dependence of their achievable data rates and mobility.

Illustration of Contemporary wireless standards.

Figure 1.1 Contemporary wireless standards.

(Based on [8].)

Note that the upper-right corner of the picture in Figure 1.1 remains empty. This is because high mobility usually means high velocity and high dynamics of the radio communication channel. This entails limitations on the possible data rates. At the same time, it is visible that subsequent generations of mobile communication systems have been aiming at higher mobility and higher data rates. The trend toward Fifth Generation (5G) (and beyond 5G) mobile communication seems to have even more demanding assumptions.

1.1 5G Radio Communications

According to the Cisco predictions, annual global IP traffic will surpass the zettabyte (1000 exabytes) threshold in 2016, and the two zettabyte threshold in 2019. (…) By 2019, global IP traffic will pass a new milestone figure of 2.0 zettabytes per year. (…) Traffic from wireless and mobile devices will exceed traffic from wired devices by 2019. (…) Globally, mobile data traffic will increase 10-fold between 2014 and 2019 [9]. According to the recent Ericsson Mobility Report [10], the number of mobile subscriptions at the beginning of 2016 totaled 7.3 billion, while it is predicted that in 2019, this number will be 9.3 billion. Moreover, communication of billions of machines and devices that are expected to comprise the Internet of Things poses even greater challenges, never encountered before. That is why 5G wireless communication is the focus of research and industry interest, aiming at achieving 1000 times the system capacity, 10 times the energy efficiency, data rate, and spectral efficiency, 25 times the average mobile cell throughput, and significantly lower latency compared with today's 4G [11]. The paradigms for future 5G systems provided in [12] are ultrahigh capacity (1000 times higher per square-kilometer), ultralow latency (lower than 1 ms), massive connectivity (100 times higher), ultrahigh rate (up to 10 Gbps), ultralow energy consumption. Although these performance targets do not need to be met simultaneously, they provide the basis for the Gbps user experience for 5G networks [12]. In the practical real-world scenarios, 5G networks should support data rates exceeding 10 Gbps in the indoor and dense outdoor environments, and several 100 Mbps in urban and suburban environments, while 10 Mbps should be accessible almost everywhere, including rural areas in both developed and developing countries [13].

It is also anticipated in [13] that 5G networks will not be based on one specific radio-access technology. Rather, 5G communication system will consist of a portfolio of access and connectivity solutions addressing the demands and requirements of mobile communication beyond 2020. The specification of 5G will include the development of a new flexible air interface, which will be directed to extreme mobile broadband deployments, and target high-bandwidth and high-traffic-usage scenarios, as well as new scenarios that involve mission-critical andreal-time communications with extreme requirements in terms of latency and reliability [13]. Apart from the extended mobile broadband and mission-critical communication, other use cases considered for 5G radio are massive machine-type communication, broadcast/multicast services, and vehicular communication.

Similar vision on 5G capabilities is presented by the European experts of 5G Infrastructure Association in [14]. According to this vision, by increasing key performance indicators (data rates, data volume, reliability, mobility, energy efficiency, density of served devices, inverse of the end-to-end latency, and service development time) indicated in the radar diagram in Figure 1.2, the 5G communication technologies will be an economy booster, paving new ways to organize the business sector of service providers, as well as fostering new business models supported by advanced information and communication technologies. They will provide user-experience continuity, the Internet of Things (machine type of communication), and mission-critical (low-latency) services.

Illustration of 5G disruptive capabilities.

Figure 1.2 5G disruptive capabilities.

(Based on [14].)

To support these 5G radio communication system requirements, in particular for the mobile broadband use case, new spectrum is required. The World Radiocommunication Conference (WRC) in 2015 took a key decision that it will provide enhanced capacity for these kind of systems, that is, to allocate the 694–790 MHz frequency band in ITU Region-1 (Europe, Africa, the Middle East, and Central Asia) for the mobile broadband radio services. Full protection has been given to the incumbent systems operating in this frequency band (e.g., digital television broadcasting). Moreover, WRC-15 decided to include studies in the agenda for the next WRC in 2019 for the identification of bands above 6 GHz. Higher frequency ranges in the millimeter-wave frequency bands are also studied widely to use there unlicensed bands for 5G transmission, as well as cognitive radio technologies that allow for spectrum sharing and, thus, higher efficiency of frequency-band utilization.

Technologies that look promising for 5G include massive MIMO antenna systems, energy-efficient communications, cognitive radio networks, and small cells (pico- and femtocells) including extremely small, mobile femtocells. In part, these technologies will tackle the problem of the relatively poor wireless service inside buildings. Wireless devices are used indoors about 80% of the time, yet today's cellular architecture relies mostly on outdoor base stations. Such indoor wireless communications will increase the energy efficiency of wireless systems, because by separating indoor traffic from outdoor traffic, the base station would face less pressure in allocating radio spectrum and could transmit with lower power. For mobile users in cars, trains, and buses, mobile femtocells can also improve service quality [11]. There is the general consensus on the forecast that the increased traffic will be handled by heterogeneous networks, which will use different types of network nodes equipped to handle various transmission power levels and data processing capabilities and support different radio-access technologies [15]. These technologies are supported by different types of backhaul links. Thus, their interoperability is understood more broadly than the interworking of wireless local area networks and cellular networks.

Low-power micro nodes (base stations and mobile terminals) and high-power macro nodes can be maintained under the management of the same operator, sharing the same frequency band. Thus, joint radio resource and interference management needs to be provided to ensure the coverage of low-power nodes. Moreover, the nodes can use discontinuous bands and aggregate fragmented spectrum. For this purpose, new transmission techniques are being considered for future heterogeneous communications [3, 4]. Enhanced Orthogonal Frequency-Division Multiplexing (OFDM) and Filter-Bank Multi-Carrier (FBMC) are multicarrier technologies that can be mentioned as examples of application in the future 5G radio interfaces. By applying intelligent spectrum sharing methods, strong interference among cells can be avoided, as well as interference originating from coexisting systems. The objective of heterogeneous networks targets the improvement of overall capacity as well as cost-effective coverage extension and green radio solution by deploying additionalnetwork nodes within the local area range, such as low-power nodes in micro-, picocells, home-evolved Node-Bs (HeNBs), femto nodes, and relay nodes. Moreover, future 5G networks will support device-to-device (D2D) communications omitting intermediary base station, allowing devices at close distance to each other to communicate without going through the main network infrastructure. This kind of communication is viewed as very effective for traffic offloading and for improving spectrum reuse in densely populated areas.

The topic of 5G heterogeneous networks has gained much momentum in the industry and research community very recently. The 3rd Generation Partnership Project (3GPP) Long Term Evolution Advanced (LTE-A) has started a new study item to investigate heterogeneous network deployments as an efficient way to improve system capacity as well as effectively enhance network coverage [16]. It has attracted the attention of IEEE 802.16j standardization. There have been a number of special issues in leading scientific journals and magazines focused on 5G communications and heterogeneous network key issues and prospective performance (e.g., Refs [17–20]), as well as dedicated events and workshops at major scientific conferences. Undoubtedly, a need is still recognized by both industry and academia to better understand and elaborate more on the technical details and performance gains that can be made possible by future heterogeneous networks. To support the heterogeneity of radio-access technologies and networks, radio resource sharing, and infrastructure sharing in the 5G communication context, virtualization of networks and their control is intensively researched.

1.2 Challenges for Future Radio Communications

Future heterogeneous networks come with the challenges, and there are important technical issues that still need to be addressed for successful deployment and operation of these networks. In theory, the overall capacity scales with the number of small cells deployed in a unit area. By reducing each cell's radius and by introducing more cells in a given area, more capacity can be offered, and spectrum reuse can be increased. However, as cells get closer, the hyperdensification of networks is challenged in many ways. Let us consider the following equation based on the capacity of an additive white Gaussian noise (AWGN) channel. The throughput c01-math-001 of a user in a cellular system is upper-bounded by [21]:

1.1

equation

where c01-math-003 denotes capacity, c01-math-004 denotes the base station signal bandwidth, c01-math-005 (load factor) denotes the number of users sharing the given base station, c01-math-006 (spatial multiplexing factor) denotes the number of spatial streams between a base station and a user device, and c01-math-007 denotes the desired signal power, while c01-math-008 and c01-math-009 denote the interference and noise power, respectively, observed at the receiver. The signal bandwidth c01-math-010 can be increased by using additional spectrum, if it is available. The load factor c01-math-011 ( c01-math-012 ) can be decreased through cell splitting, which involves deploying a larger number of base stations and ensuring that user traffic is distributed as evenly as possible among all the base stations. Spatial multiplexing factor c01-math-013 can be increased using a larger number of antennas (with suitable correlation characteristics) at the base station and user devices.

Cell splitting has the favorable side effect of reducing the path loss between a user device and base stations, which increases both desired and interfering signal levels c01-math-014 and c01-math-015 , effectively lowering the impact of thermal noise c01-math-016 . As a result, interference mitigation is paramount for link efficiency improvement in modern cellular systems. This requires a combination of adaptive resource coordination among transmitters and advanced signal processing at the receivers. The aforementioned parameters for wireless capacity enhancement may be viewed under a common umbrella of network densification. Network densification is a combination of spatial densification (which increases the ratio c01-math-017 ) and spectrum aggregation (which increases c01-math-018 ) [21]. Spectral aggregation refers to using larger amounts of electromagnetic spectrum, spanning from 500 MHz to the millimeter-wave bands (30–300 GHz). Aggregating potentially noncontiguous fragments of bandwidth across such disparate frequency bands poses numerous challenges for antenna and Radio Frequency (RF) transceiver design, which need to be overcome in order to support spectral aggregation.

Overall, noncontiguous multicarrier techniques, such as Non-contiguous Orthogonal Frequency-Division Multiplexing (NC-OFDM) Non-contiguous Filter-Bank Multi-Carrier (NC-FBMC), have been recognized as suitable candidates for 5G transmission due to their potential for achieving spectrally efficient communications by aggregating and exploiting fragmented unoccupied spectrum while simultaneously achieving high data rates [3]. Both these techniques possess the ability to efficiently use fragmented spectrum opportunities as well as perform spectrum shaping in order to suppress interference that may affect nearby wireless transmissions. To counteract the potential for significant interference resulting from NC-OFDM Out-of-Band (OOB) power emission, several techniques have been proposed in the literature, which are designed to significantly suppress these sidelobes in order to make coexistence between systems utilizing adjacent spectrum feasible [4]. On the other hand, NC-FBMC, which applies subcarriers filtering, handles the OOB power emission at the required level. This comes at the expense of increased computational complexity. One can envision that the OOB level can be adjusted by adaptively modifying the filter characteristics, that is, by adaptive waveform design. However, so far, there has been no published work on this kind of pulse-shapeadaptation. Given the possible constraints of limited computational and energy resources available via a user equipment and other elements of the radio-environment context, a practical approach to this problem that achieves a balance between the OOB interference mitigation efficiency and its associated costs is needed. Moreover, there are many issues of such spectrum aggregating, noncontiguous multicarrier signal reception. One of them is synchronization in the presence of self-interference (among subcarriers) and external in-band (especially narrowband) interference, the other being reception/detection quality.

As shown earlier, in a hyperdense deployment, not only desired signal strength but also interference from other cells increases. Increasing other-cell interference needs to be mitigated, and mobility management mechanism is required as the mobile users see cell edges more frequently. Furthermore, as some privately owned small cells implement restricted access schemes, they can generate/receive strong uncoordinated interference to/from external cells sharing the same radio resources [22]. The deployment of small cells is mostly unplanned, so a network self-organizing mechanism needs to be developed. The self-organizing features of small cells can be generally classified into three processes: (i) self-configuration, where cells are automatically configured by downloaded software; (ii) self-healing, where cells can automatically perform failure recovery; (iii) self-optimization, where cells constantly monitor the network status and optimize their settings to improve coverage and reduce interference [23]. The promising performance gain by deploying more cells can only be achieved by successfully addressing these problems. It can be noted that there have been some achievements in recent years, in the development of enhanced intercell interference coordination in 3GPP LTE-Advanced systems [24]. Still, intelligent incentive schemes are needed to motivate privately owned small cells to open up for wider access.

Besides the issue of meeting the overwhelming traffic demands, network operators around the world now realize the importance of managing their cellular networks in an energy efficient manner and reducing the amount of c01-math-019 emission levels [25]. As a result, the terminology of green cellular network has become very popular recently, showing that the energy efficiency as one of the key performance indicators for cellular network design [26, 27]. Although the deployment of small-cell networks is seen to be a promising way of catering to increasing traffic demands, the dense and random deployment of small cells and their uncoordinated operation raise important questions about the implication of energy efficiency in such multitier networks. Besides introducing small cells into existing macrocell networks, another effective technique is to introduce sleep mode in macrocell base stations and offload the traffic to smaller or more energy-efficient cells [28, 29]. Moreover, in some cases of dense small-cell deployment, sleep mode can also be possible and advantageous for open-access picocells to reduce energy consumption in the area [30]. Proper traffic balancing between cells of diverse coverage and networks will also result in higher Quality ofExperience (QoE) for the end users by lowering the probability of blocked calls [31].

In the 5G Era, networks, systems, and nodes will need to be context-aware, utilizing context information in a real-time manner based on networks, devices, applications, and the user and his/her environment. This context awareness will allow improvements in the efficiency of existing services and help provide more user-centric and personalized services. For example, networks will need to be more aware of application requirements, QoE metrics, and specific ways to adapt the application flows to meet the QoE needs of the user. The context-based adaptations of various transmission and network parameters will have to take into account the following context information: device-level context, application context, user context, environment context, and network context [32, 33]. This context information itself consists of different parts/components, each of which affects the individual steps of the decision-making process in a different way, as shown in [34]. There are two important performance aspects related to the use of context information: signaling overhead and information reliability. They directly relate to key performance metrics such as network capacity and energy efficiency.

1.3 Initiatives for the Future Radio Interface Definition

The common understanding of future 5G communication and considered frequency bands of hundred of GHz implies that using one radio interface to address this wide range of frequency bands is not a good approach. This is because propagation characteristics, implementation aspects, and compatibility issues are different for different frequency ranges. Therefore, the overall