6G Challenges

Introduction

Generally, each cellular technology has an average lifecycle of 10 years from original conception to full commercialization. As 5G is becoming a commercial reality, all mobile operators and OEMs around the world are competing for market primacy. 5G was originally envisioned to be a technological shift compared to 4G in terms of aggregate spectral and energy efficiency, as well as, latency and reliability. After nearly eight years of intensive academic research and industrial testing, the following lessons emerge: a) 5G can indeed support emerging data-hungry applications (e.g. ultra-fast broadband, high-definition video streaming), mainly through advances in the massive MIMO; b) 5G is still falling short of supporting the so-called Internet-of-Everything (IoE), where myriads of devices in a geographic cube require either low-latency, ultra-reliable connectivity or wireless Gpbs Internet access by availing of the mm-wave/ THz spectrum. 5G has several inherent limitations and difficulties to completely fulfill its target goals. The development of different datacentric, automated processes are proving to exceed the capabilities defined by key performance indicators of 5G. Challenges are faced in simultaneously accommodating several applications, such as haptics, telemedicine, and connected autonomous vehicles, that use long packets with ultra-high reliability and high data rates vis-a-vis short packet applications used for ultra-reliable low-latency communication (URLLC).

In addition, next generation virtual and augmented reality-based applications, such as holographic teleportation require sub-millisecond level latency and very high data-rates seem difficult to be fulfilled by 5G networks.

Consequently, a number of experts are asking the grand question: “What comes after 5G?” as also “Are we approaching the limits of wireless communications?”. The answers to these questions are not straightforward, especially taking into account the ever increasing number of connected devices for the years to come.

In this context, rudimentary idea of 6G has boldly gained ground, for eg 6G will (a) push the communication to higher frequency bands (mm-wave and THz), (b) create smart radio environments through reconfigurable surfaces and (c) remove conventional cell structures (aka cell-free massive MIMO).

Unfortunately, transforming these still theoretical or simulated concepts into commercially viable solutions is a very challenging exercise.[i]

Emerging Machine Learning Schemes

6G will use artificial intelligence (AI) as an integral part that has the capability to optimize a variety of wireless network problems like convex optimization schemes, matching theory, game theory, heuristic, and brute force algorithms. However, these solution approaches might suffer from the issue of high complexity which in turn degrades the capacity of a system. Machine learning is capable of optimizing various complex mathematical problems including the problems that cannot be modeled using mathematical equations.

A self-organizing (i.e., self-operating) network offers optimization, management, configuration, and planning in an efficient, fast manner. Self-organizing network was systematically outlined in 3GPP Release 8. However, the traditional self-organizing networking scheme is still to be realized. Self-sustaining 6G systems have to adapt to the highly dynamic environment sustainably.

Machine learning (ML) is considered one of the key drivers of 6G. Distributed machine learning can lead to high-performance computing by enabling parallel computation of machine learning models at distributed locations. There can be two possible ways, such as the data-parallel approach and model-parallel approach, to distribute the machine learning tasks. The data parallel approach might not be feasible for many machine learning models that cannot be split up into parts. Distributed machine learning models can have many possible ways of realisation that include centralized, tree-based decentralized, parameter server-based decentralized, and fully distributed. Challenge will lie in managing huge amount of data generated by some of the use case scenarios. For instance, autonomous driving cars can generate as much as 4 TB of data every day. In this scenario, real-time interaction is required in an incremental manner with most of the storage and compute happening at an edge node. Centralized machine learning based on onetime training can also be used. However, the model trained via centralized machine learning might not produce good results in real-time scenarios.

Communication Technologies

A 6G system may use novel communication technologies like terahertz communication, quantum communication, 3D wireless communication, visible light communication, nanoscale communication, and holographic communication. Terahertz communication offers several advantages, but several challenges must be resolved to enable its use in 6G. These challenges involve the design of efficient transceivers with advanced adaptive array technologies to increase its range. Another important aspect of 6G is the use of 3D communication which involves the integration of ground and airborne networks. Unarmed aerial vehicles and low-orbit satellites can be used as base stations for 3D communication. Novel schemes are necessary for resource allocation and mobility handling for 3D communication networks. Nanoscale communication is a new communication technology that uses an extremely short wavelength for communication and is suitable for a distance of 1 m or cm. Key challenges of nanoscale communication are nanoscale transceiver design and channel modeling. Visible light communication can be used to enable several 6G applications using a visible light spectrum that ranges from 430 THz to 790 THz. The main advantage of visible light communication is the use of illumination sources for lighting and communication, also light communication offers a substantial large bandwidth and interference-free communication from radio frequency waves. However, visible light communication with low-range, novel transceivers must be designed to enable different visible light communication-based applications. Several challenges must be resolved to enable 6G with visible light communication, that include connectivity of light-emitting diode to the Internet, inter-cell interference, mobility and coverage etc. To enable 6G with a high capacity, one can deploy light-emitting diodes for visible light communication in a dense fashion. However, they will suffer from intercell interference due to various use cases and implementations.

Quantum communication has the inherent feature of high security, which makes it preferred for 6G. The simultaneous achievement of long-distance and high rates is contradictory in quantum communication. Therefore, repeaters must be used to enable secure long-distance, high-data-rate quantum communication. However, current repeaters cannot be used for quantum communication, and new repeaters must be designed.

Networking Technologies

Novel networking technologies for 6G are nano-networking, bio-networking, optical networking, and 3D networking. The operation of the n-IoT is based on molecular communication. Different materials, such as graphene and meta materials can be used to build nanometer-range devices. B-IoT using biological cells are used for communication using IoT. B-IoT and n-IoT are seemingly integral parts of future 6G smart services but have several implementation challenges. The design of physical layer technologies for molecular communication is a challenging task. Apart from physical layer techniques, novel routing schemes will be required to be used because of the substantially different nature of B-IoT and n-IoT compared with traditional IoT. Further-more, novel models must be devised for a 3D communication network due to its substantially different nature compared with a 2D network. Routing protocol design with low computational complexity and short-range communication are required to take care of limited energy, short-range communication, and low computing capabilities of nano-nodes and bio-nodes.

Computing Technologies

6G system involves a wide variety of sources of different smart applications that generate an enormous amount of data. High-performance and quantum computing must be used to enable intelligent data analytics. Quantum computing is expected to revolutionize the field of computing by enabling higher speeds that users have never experienced until now. Other than quantum computing, intelligent edge computing is required for 6G to provide intelligent on-demand computing and on-demand storage capabilities with extremely low latency to end nodes.

For instance, consider the training of an intelligent caching agent in XR applications. Federated learning can be used to train the edge agent efficiently by reducing wireless resource usage through sending only model updates (that have much less size compared with the whole training data) to the edge/ cloud server. Similarly, federated learning can be used to enable intelligence in an adaptive transceiver.

Higher Frequency Bands

Wireless systems for 6G will potentially rely on: i) Millimetre-wave technologies (30 to 300 GHz); ii) THz technologies (300GHz to 3 THz); and (iii) Free space optics (FSO). For example, utilizing spectrum available from 100 GHz to 1 THz, will have the potential to deliver data rates in excess of 10 Gbps.

A major challenge that will be faced at designing HW, where bond wires cause considerable signal degradation. The effects of bond wires are difficult to characterize for large signal applications, such as power amplifiers and also for phase critical applications, such as beamformers for phased arrays, where these can introduce undesirable side lobe levels. Traditional metallic split-block packages provide excellent performance but are bulky and heavy. Cutting-edge techniques in micro-machining and LTCC technology yield compact and low-cost solutions. Additive manufacturing techniques, such as metal coated 3D printing of plastic devices can realize low-cost, light weight and compact devices.

The compact physical size and power efficiency requirements will become more challenging at mm-wave and THz frequencies. Hybrid beam forming will be best suited to implement large number of antenna elements along with high efficiency amplifiers. Performance parameters, such as the noise figure, output power and power efficiency degrade significantly for high operating frequencies. Demodulation of higher order modulated signals also becomes more challenging as phase noise increases at higher frequencies. Advanced array SP techniques are needed to complement the transceiver design to address these challenges. Novel techniques, such as spatially oversampled antennas and new phased array architectures would be required to be leveraged to provide a solution to size, weight and power consumption of large mm- wave/ THz antenna arrays.

Calibration, verification, and measurement traceability at THz frequency bands remains a major challenge. For time domain systems, a major challenge is the establishment of standardized measurement and calibration, whereas for VNA systems, solutions are being sought for high precision waveguides and interconnects. Electrooptic sampling (EOS) shows much promise as a complementary approach to THz measurements, although it is yet to extend the bandwidth to 1.5 THz and to improve resolution.

Channel Estimation

Extensive research in 5G has been concerned with the need to reduce training overheads in pilot-based channel estimation. In 6G networks, viable solutions become extremely challenging due to the massive scale-up and connectivity demands, in particular: (i) supporting high data rates (Gbps) in high-mobility scenarios—will require dealing with much shorter channel coherence times; (ii) ultra-low latency requirements will see transmission intervals substantially shortened, and (iii) the number of parameters to estimate will be massively large as a consequence of the scaling (not only of antennas/ APs, but also in the numbers of users/devices).

The high-dimensional channels will therefore need to be estimated in severe under-sampling constraints, which might render pilot-based (coherent) estimation approaches unfeasible, particularly under high-mobility or low-latency requirements. Blind (non-coherent) estimation approaches, stand as promising alternatives and might play a key role. However, these approaches typically require knowledge of the (high-dimensional) received signal covariance matrix, which will again need to be acquired from a limited number of samples. To that end, research in the fields of random matrix theory (RMT) and high-dimensional statistics will be relevant; in particular, to develop accurate estimators of large covariance matrices (and their eigen-spectrum) under limited sampling. Other, more recent approaches based on machine learning (ML) might also play a significant role.

Adaptive Filtering

In beamforming, transmitted signals are dynamically adapted (via digital precoding) to the propagation conditions, effectively mitigating interference and noise. Adaptive beamforming can be seen as a linear filter with a particular design objective, e.g., to maximize the signal-to-noise ratio (SNR). Optimal solutions for the beamformer (and associated receiver filters) require the covariance matrix of the aggregated interference and noise; unknown in practice, this needs to be estimated from observed samples. Current solutions rely on classical estimators, such as the sample covariance matrix (SCM), which will return a poor estimate in high-dimensional 6G scenarios, due to:

• Finite sampling (scarcity of samples): While the numbers of antennas and user devices will scale up massively, strict low-latency and high- mobility requirements will impose a rather limited number of training (observed) samples.
• Temporal correlation: The ultra-dense and highly decentralized deployments, e.g., with thousands of distributed access points in CF networks, will be subject to non-perfect synchronization (e.g., between interference and desired signals) which might in turn produce non-stationarity effects.
• Outlying samples: With millions of interconnected devices (from electrical/ smart appliances to connected vehicles), it is also reasonable to expect multiple sources of impulsive noise and, in security-sensitive applications/ environments, eventual sources of intentional interference (jamming).

Traditional estimators (e.g., SCM) rely on the sufficient availability of samples (the number of samples should be far greater than the number of signals). Under the conditions above, however, the mismatch between true and estimated covariance leads to highly inaccurate filters with severe performance losses, in terms of connectivity, reliability, and data rates. A major challenge is then to develop filtering solutions which are robust to the effects of finite-sampling, temporal correlation, and corrupted (outlying) samples.

Intelligent Wireless Energy Harvesting

Wireless energy harvesting can be one of possible ways to enable sustainable operation of 6G. Wireless energy harvesting covers numerous harvesting scenarios: dedicated radio frequency harvesting sources, interference-aware harvesting, and ambient sunlight harvesting. However, substantial variations exist in harvested energy for these wireless energy harvesting sources. Therefore, an intelligent power control must be developed for energy-harvesting devices. Traditional power control schemes for energy-harvesting devices assume the known system state (incoming harvesting energy and wireless channel), but this information is not available practically. Machine learning can be used to predict the future system state and address these challenges.

Cell-Free Massive MIMO

Cell-free massive MIMO (CF MaMi) has been proposed to overcome the boundary effect of cellular networks. In CF MaMi, many access points (APs) distributed in a geographic coverage area coherently serve many users in the same time frequency resources. There are no cells, and hence, no boundary effects. Using many APs, CF MaMi offers many degrees of freedom, high multiplexing gain, and high array gain. As a result, it can provide huge energy efficiency and spectral efficiency. These gains can be obtained with simple Signal Processing (SP) due to the favorable propagation and channel hardening properties of MaMi technology. CF MaMi offers macro-diversity gains. It promises no dead zone in a not so planned network. In CF MaMi each AP has a few antennas, hence it is expected to be built by low-cost, low-power components and simple APs. Challenge in CF MaMi is when all APs participate in serving all users through the backhaul connections with one or several central processing units (CPUs). This is not scalable in the sense that it is not implementable when the network size (number of APs and/ or number of users) grows large.

Designing a scalable structure is one of the main challenges of CF MaMi. Another challenge is regarding power control that is central in CF MaMi since it needs to control the near-far effects e.g. the max-min fairness or the total energy efficiency. This requires huge front/ back-hauling overhead. Provided that, it is very difficult for the CPU to have perfect knowledge of large-scale fading coefficients associated with a potentially unprecedented number of APs and users. Some heuristic power control schemes with local AP processing are possible candidates.

One of the ultimate aims of CF MaMi research is designing a distributed SP scheme. Otherwise, the system is not scalable. Compared to other linear processing schemes such as zero-forcing (ZF) and minimum mean-square error (MMSE), the performance of conjugate beamforming is far below. To cover the gap between conjugate beamforming and ZF/ MMSE, very large numbers of service antennas are required. Currently, there are no distributed SP schemes available for the uplink in CF MaMi. Even with the simple matched filtering, the (processed) signals from each AP needs to be sent to the CPU for signal detection which is a challenge.[ii]

Designing a scalable structure is one of the main challenges of CF MaMi. Owing to the path loss, only 10-20% of the total number of APs really participate in serving a given user. Network-centric approach can be adopted to address this challenge by dividing the APs are divided into disjoint clusters, with each cluster coherently serving users in their joint coverage area. Network- centric-based systems still have boundaries, and hence, are the final solution. By contrast, in user-centric approach, each user is served by its selected subset of APs. In this approach, there are no boundaries. There are several simple methods to implement user-centric approach, such as each user chooses some of its closest APs or chooses a subset of APs which contribute to most of the total received power of the desired signal. Yet, the cluster-set changes quickly requiring more control signaling. Hence, designing a practical user-centric approach is a challenging exercise and requires urgently intensive research.

Reconfigurable Smart Reflecting Surfaces

High frequency communication suffers from significant path loss, coping with which will require use of reconfigurable smart reflecting surfaces. A reflective surface is a planar aperture synthesized using an array of sub-wavelength elements (or unit cells). Such a surface can be used to modulate the incoming waves into a desired wavefront upon reflection. Due to their sub-wavelength unit cell sampling, reflective surfaces can be considered a distinct form of meta-surfaces, synthesized using an array of metamaterial unit cell elements. There has been a substantial amount of research conducted in EM wave control using meta-surface apertures with applications ranging from imaging to EM invisibility.

Different from the phased array technology, meta-surfaces do not rely on phase shifting circuits to modulate the phase of the incoming waves. Instead, meta-surfaces rely on a holographic principle to achieve the desired phase modulation. The incoming wave illuminating the aperture surface acts as a reference-wave, which is converted to a desired wavefront upon reflection from the reflective meta-surface aperture. The major advantage of meta-surfaces is that they can synthesize any arbitrary waveform using this simple, yet strong, holographic principle without the need for expensive and power hungry phase shifters.

An important limitation in the design process of a reflective surface is the achievable phase range of the unit cells synthesizing the aperture. Ideally, each unit cell across the reflective surface should provide a full phase control across a phase range of 0-2π. However, achieving this full-phase range may require a complex unit cell topology, with multiple layers complicating the hardware architecture. Another important constraint in the design process of a reflective surface is the quantization of the phase range of the unit cells. Even if a full phase range of 0-2π is achieved, the number of quantization levels used to discretize this phase range has a direct impact on the fidelity of the synthesized wavefront upon reflection from the reflective surface.

The challenges of the static meta-surfaces can be offset by Intelligent Reflective Surface (IRS) has the capability to dynamically tune the reflection response of the aperture in an all-electronic manner. This is particularly important as communication environments have dynamic characteristics, namely variations in the number of connected users and non-static location distribution over time. Thus, the capability to intelligently change the characteristics of the reflected wavefront to meet their dynamic metrics plays a crucial role in future wireless communication systems. The dynamic modulation of the IRS can be achieved using low-power semiconductor elements, e.g., PIN diodes. This advantage of IRS makes them a promising candidate to replace the power-hungry phase shifting apertures.

Conclusion

Being critical is a self-directed and self-correcting capacity to think otherwise about an issue in order to see the limitations of ones’ or others’ past or present theoretical or practical reasoning about this issue. In essence, it is about reflexivity. We need to be critical because, in our age of digital transformation, society faces considerable challenges and risks related to its core values. R&D spending on computing and telecommunications has fuelled technological changes in society and related problems at unprecedented speed and scale. It remains difficult not to believe that the systems which we use daily are implicated in the changes and challenges of our current world. This is a world that runs at internet speed, where individuals are becoming heavily dependent on technologies, if not addicted to some of them, to the point that the most powerful democracies can be manipulated by foreign powers. Technology can have unintended consequences, however, technology is a solution, not the cause of problems. The dilemma(s) we face are about the values that we choose. Today’s obvious lesson is that tech giants develop agile apps and digital platforms in ways that sense and respond to our “needs” and create business value. Further, they learn more about us the more we use their platforms in order to accumulate the highest possible value. Norms and values are important, and whereas risks appear at all levels, the greatest difficulties lie at the collective and, in particular, at the societal level where individuals cannot do much by themselves(https://www.tandfonline.com/doi/full/10.1080/0960085X.2018.1471789).

With the eyes working at 576 megapixels (https://www.discovery.com/science/how-old-is-the-sun-), the networking and mobile world has lots to catch up to meet the basic needs of eye itself. So 6G is yet another step in that direction.

6G systems will unlock the full potential of smart cities via enabling Internet of everything based smart services. AI will be an integral part of the 6G wireless system to solve complex network optimization problems. Terahertz communication will be considered as one of the key communication bands for 6G systems. Novel models and architectures will emerge for the 6G Physical Layer. Furthermore, new models must emerge for quantum communication that is currently in infancy.[iii]

This article analyzes the major challenges that will impact the 6G architecture and design creation, including optical communication in open spaces, energy harvesting, wireless charging, machine learning as well as Cell Free MIMO upgrades.

[i] The road to 6G: Ten physical layer challenges for communications engineers, Matthaiou, M., Yurduseven, O., Ngo, H.-Q., Morales-Jimenez, D., Cotton, S., & Fusco, V. (2021). The road to 6G: Ten physical layer challenges for communications engineers. IEEE Communications Magazine, 59(1). https://doi.org/10.1109/MCOM.001.2000208

[ii] Ultra-Massive Multiple Input Multiple Output Technologies for 6G Wireless Networks, Ravilla Dilli,1,* Ravi Chandra M L2,# and Deepthi Jordhana3,#

[iii] 6G: Opening New Horizons for Integration of Comfort, Security and Intelligence Guan Gui, Senior Member, IEEE, Miao Liu, Member, IEEE, Fengxiao Tang, Member, IEEE, Nei Kato, Fellow, IEEE, and Fumiyuki Adachi, Life Fellow, IEEE