The shift to 6G communications: vision and requirements – Human-centric Computing and Information Sciences

Next-generation wireless networks will consist of massive number of connected devices and with the base stations (BSs)/access points (APs) leading to mMTC. Multiple BSs/APs may serve one or more devices simultaneously to form a coordinated multi-point (CoMP) transmission [39]. The huge amount of data produced by massive devices will require very high-performance processing units and robust backhauling links. The central processing units may utilize ML and AI algorithms and the backhauling links may utilize optical fiber and or photonic communications. Remote user, in 6G communication systems, can use several relays or transmitters for a remote user to transmit, and the user’s SINR may be improved by using the technique of diversity as in virtual MIMO systems.

By intelligent networking, all the end devices would be aware of the location and features of BSs/APs in their vicinity, and all of the BSs/APs would be aware of the locations, features, and QoS requirements of devices in their vicinity. Robust interference management/optimization techniques can be applied to maximize the efficiency of the wireless network. Central processing units will be fast enough to manage and switch the resources (bandwidth, time, power) among multiple end-users, and data processing will be conducted at the base-band processing units (BPUs). Figure 3 depicts some of the major components in the 6G system architecture, that will cause a major paradigm shift towards the realization of 6G standards. The air interface is the main component that causes a major improvement in the wireless generations. Orthogonal frequency division multiplexing (OFDM) played a major role in the development of 4G, as code division multiple access (CDMA) was the key player in 3G. Similarly, the development of the new air interface will be an essential component of 6G system architecture.

Fig. 3

6G wireless network elements

Full size image

AI and ML is another crucial component of the 6G system architecture. AI and ML will play an important role in the self-organization, self-healing, self-configuration of 6G wireless systems. Spectrum congestion has also pushed the 6G to adopt a new spectrum for communication. Therefore, this new spectrum will also be an active component in the 6G system architecture. Since 6G will accommodate a wide range of communication devices ranging from IoTs to live HD video transmission, 6G will need to be in line with all previous technologies. Therefore, a flexible and multi-radio access technologies (RAT) system architecture will be an essential component in the 6G network.

Air interface

Since 6G will concentrate on the current terahertz frequency range with extremely wide bandwidths available, it will bring up new obstacles to interact efficiently at these frequencies. Getting a secure transmission infrastructure that has an adequate range and isn’t power-hungry will be the answer here. The availability of incredibly wide bandwidths would change the emphasis from spectrally optimized solutions to improved coverage solutions. In these new frequency spectrums, the tradeoff between spectrum performance, power efficiency, and coverage will play a key role in developing devices. This will lead to the design of a modern air interface where more consideration can be paid to single-carrier systems. The OFDM scheme would be revisited for lower frequency ranges where spectral efficiency will be important as it does not use the energy effectively because of the cyclic prefix, which is just the duplication of information and does not hold any additional information. Furthermore, a high peak-to-average power ratio (PAPR) makes the power amplifiers complex and expensive.

Many researchers have proposed the non-orthogonal multiple access (NOMA) as a promising new scheme for the B5G/6G mobile networks [40,41,42]. In NOMA, all of the users are allowed to access the complete resource (frequency band) simultaneously. Some researchers have suggested the rate-splitting multiple access (RSMA) as a new access technology for 6G communication systems [43,44,45]. Both NOMA and RSMA rely on the successive interference cancellations (SIC) to decode the information for the user. RSMA uses the SICs to decode the common message firstly and then decode the private message. Both schemes need to be matured enough before practical deployment. A new AI-based software-defined air interface is presented in  [46], where the authors proposed an intelligent air interface switching system for user QoS enhancement.

3GPP release 15, reveals the specifications for the 5G-New Radios (NR), in which multiple waveform configurations and two sets of frequencies are defined. By adopting the variable numerologies (symbol duration, sub-carrier spacing, and pilot spacing), we can give the transmitter leverage to self-organize and self-configure according to the channel conditions and service required. This is often useful on different measurements. For instance, by reducing the symbol length, low latency can be achieved, and increasing the spacing of the sub-carrier can be helpful in reducing the phase noise in mmWave and sub-mmWave. In high mobility situations, optimizing the sub-carrier width can also be helpful for Doppler shift compensation.

New spectrum

mmWave is already a candidate for 5G, but it is not exploited to its full potential as the beamforming algorithms are not mature enough. It requires improvements in the networks when personal BSs and satellite connectivity can get merged into cellular communication. In the previous generations, the spectrum is divided for multiple services, for instance, television (TV) services, military communications, and cellular communication [47].

Therefore, the idea of using an unlicensed spectrum is proposed, i.e., to use the mmWave, THz band, and visible light spectrum, simultaneously [48,49,50,51,52]. These bands are never used for any communication. The problem with the higher frequency band, though, is that the signal is attenuated very rapidly about the distance traveled. For example, a 3G or 4G BS can have a coverage of about several miles whereas a 5G or 6G BS coverage may limit to only a few hundreds of meters. To resolve this issue in mmWave and THz communications, the idea of using massive multiple inputs and multiple outputs (MIMO) and beamforming emerged, which is described in the next subsections.

Artificial intelligence/machine learning

By offering pervasive, secure, and close proximity-instant wireless networking for humans and computers, 6G wireless communication networks would be the core of society’s digital transition. A broad variety of emerging developments, such as self-driving cars and voice assistants, have been made possible by recent advancements in ML research. B5G/6G wireless networks have increased complexity, requiring smarter methods for handling any losses and handling network features, detecting anomalies, and understanding KPI trends. This can be done by introducing solutions for ML and SDN. In order to preserve a certain level of KPI, ML/AI will boost the decision-making process. The operation and implementation of RAN for 6G needs a new strategy. Incorporating AI in wireless algorithms (e.g., for channel estimation, for channel state information (CSI) feedback, and decoding, etc.) may bring a change in the direction of these algorithms [53]. Application of ML, DL [54], and AI algorithms to the communication network, we can instantly manage the resources as per the user requirements. The probability of choosing the best solution is improved in this way and the network can maintain its optimum state.

Advanced beamforming with very large scale antenna (VLSA)

The idea of beamforming is to steer the beam to only the desired direction or user. Since energy is not spread in all directions, the transmission range is thus improved by concentrating the beam in one direction.

Intelligent reflecting surfaces (IRSs)

Intelligent Reflecting Surfaces (IRSs) can be the potential area for beamforming in 6G [55]. IRSs are composed of thin electromagnetic materials, which can reflect/configure the incoming electromagnetic rays in an intelligent way by configuring the phase of reflected rays by a software [55]. Indeed, IRSs use at a large number of low-power and low-cost passive elements to reflect the incident signals with configurable phase shifts without the requirements of additional power, encoding, decoding, modulation, demodulation requirements. IRSs are installed on the important points and locations such as high-rise buildings, advertising panels, vehicles  (cars, airplanes, unmanned aerial vehicles (UAVs)), and even the clothes of the pedestrians. The main advantage of the IRS is that it can enhance the signal-to-interference-plus-noise-ratio (SINR) with no change in the infrastructure or the hardware of the communication network. Also, there is no need for extra power required for the installation.

IRS can reduce the hardware complexity at the receiver and the transmitter by reducing the number of antennae installed at them, thereby, reducing the radio frequency (RF) chains at the transmitter and the receiver. IRS can replace the conventional relays system due to its advantages in terms of power, spectral efficiency, and reduced hardware complexity [56]. IRS can be used in the deep-fade and non-line-of-sight (NLOS) communication environment. The principle by which SINR is enhanced at the receiver is optimally controlling the phases of the incident ray at multiple elements of the IRS, to produce useful beamforming at the receiver [56]. Degradation factors such as noise and interference have no impact on the IRS. All these features of the IRS make it a promising technique for the B5G/6G communication systems.

Orbital angular momentum (OAM)-aided MIMO

A new dimensional property of the electromagnetic waves (EW) was discovered in the 1990s termed as the orbital angular momentum (OAM). This discovery promised the transmission of multiple data streams over the same spatial channel. An EM wave carrying the OAM has the phase rotation factor of \(\exp (-jl\phi )\), where l is OAM state number represented in integer and \(\phi\) is transverse azimuth angle [57,58,59]. The main advantage of OAM over other beamforming techniques is that OAM can have an unlimited number of orthogonal modes, which allows the EW to multiplex multiple data streams over the same spatial channel, thereby, enhancing the spectral efficiency and transmission capacity. OAM support a high number of user in mode division multiple access (MDMA) scheme without utilizing extra resources (i.e., frequency, time, and power). The flexibility of OAM to be used in narrowband and wideband frequency hopping scheme makes it an attractive scheme for low probability of interception (LPI) applications. OAM-based MIMO systems have advantages over the conventional MIMO systems in terms of capacity and long-distance line-of-sight (LoS) coverage [60]. Therefore, OAM has great potential for applications in 6G wireless networks.

Coexistence of variable radio access technologies

6G can lead to a ubiquitous networking infrastructure where users would not only be left with the option of selecting the best communication network. Each node in this network would, however, be intelligent enough to sense the conditions of the channel and the specifications of QoS at any other node. For example, the use case and the availability of network will decide the network as cellular, wireless LAN, Bluetooth, and ultra-wideband (UWB), etc. 6G communication standard should, therefore, be designed in such a way that it will converge all of the wireless technologies. Communication with Wi-Fi, Bluetooth, UWB, VLC, UAVs, biosensors, and satellite communications can all integrate into 6G and should fall under one standard such that all of them can connect with each other. The Wi-Fi operating at 2.4 GHz has already entered deeply into IoTs as most of the appliances are now connected through this network [61,62,63].

By merging all these technologies, 6G would be able to utilize the massive infrastructure deployed by previous technologies, which otherwise can cost 6G a huge revenue. The features in the previous technologies, such as network densification, high spectral efficiency, high throughput, low latency, high reliability, and massive connectivity should be converged in 6G. 6G technology should also keep the trend of offering new services by applying the new technologies, such as AI/ML, VLC, quantum communications (QC), and quantum machine learning (QML). These services may include but are not limited to smart cars, smart homes, smart wearable, and 3D mapping [64].

Figure 4 gives an overview of the evolution of the wireless generation, with timelines, from 1G to 6G with respect to applications, KPIs, network characteristics, and technology. Figure 4a shows that a major leap in the application is observed with 4G. 4G introduced mobile Internet, mobile TV, and HD videos. AR/VR, ultra-HD (UHD) videos, wearable devices, vehicle-to-infrastructure (V2X), smart city, telemedicine, and IoTs concepts are introduced in 5G. 6G is projected to have applications such as space tourism, Tactile Internet, fully automated cars, holographic verticals, deep-sea sight, digital sensing, and Internet-of-bio-Nano-things (IoBNT). Figure 4b shows that how KPIs are changing with the evolution of wireless generations from 1G to 6G. Figure 4c shows the evolution of the network characteristics with wireless generations. All Internet protocol (IP) and the ultra-broadband concept is introduced in 4G. The concepts of cloudification, softwarization, slicing, virtualization, and wireless worldwide web (WWW) are introduced in 5G. Integration of intelligence with cloudification, softwarization, slicing, and virtualization will be introduced in 6G communication systems. Figure 4d depicts the evolution of technologies with the development of wireless communication generations. The initial stage of the wireless communication system is the development of the advanced mobile phone system (AMPS). Global systems for mobile (GSM) and general packet radio systems (GPRS) family is developed in 2G wireless systems. Code-division multiple access (CDMA) family shifted the wireless systems from 2G to the 3G. OFDM with the integration of turbo codes and MIMO systems is the key technology for 4G communication systems. 5G communication systems brought some new technologies such as cloud/fog/edge computing, massive MIMO, SDN, mmWave and sub mmWave (NR) along with low-density parity-check (LDPC) and polar codes. ML, AI, blockchain, THz communication, orbital angular momentum multiplexing (OAM Mux), spatial Modulation (SM)-MIMO and intelligent re-configurable reflecting surfaces are the new technological domains in 6G.

Fig. 4

Evolution of wireless communication, with timeline, from 1G to 6G based on (a) Applications (b) KPIs (c) Network characteristics and (d) Technological development

Full size image