The smart city consists of a number of key enablers, which are smart education, smart healthcare, a smart economy, smart factories, smart warehouses, smart transportation, etc. ^[1][2][1,2]. The key enablers of smart cities require uninterrupted and efficient power supplies to fulfill their requirements ^[1][2][3][1,2,3]. Moreover, to effectively pursue all the functionalities of a smart grid system, a suitable wireless communication is required that can provide a high data rate, reliability, flexibility, low latency, massive connectivity, security, and the necessary adaptive functionalities ^[1][4][1,4]. Among the various applications of the smart grid, massive connectivity and monitoring, secured communication for resource allocations and operations, and time-critical operations are the vital factors in the case of an effective, uninterrupted, and reliable power supply for smart cities ^[5][6][5,6]. Future 6G wireless communication is a suitable solution for the communication of smart grid systems by providing all the required key performance indicators (KPIs) in terms of communication to successfully perform the abovementioned smart grid applications ^[7][8][9][7,8,9].

1.1. Basics of the Smart City

The term “smart city” signifies a developed urban locality that ensures sustainable development with regard to the economy, as well as enhanced living standards by prevailing various key features such as industry, economy, education, health care, transportation, government, people, and living standards ^[10][19]. The smart city is a unique way through which to solve urban problems ^[11][20]. Smart applications are integrated into smart cities—whose daily activities are related to the modern technologies in the world ^[12][21]. Cities should have interconnected technology and management, along with peaceful and interactive living standards ^[11][12][20,21]. Therefore, the smart city can promptly overcome the worst scenarios and destructive crises ^[13][22]. According to Figure 1, the smart city is based on ten major key enablers. Furthermore, all the key enablers are equally important for the feasible advancement of smart cities ^[1][14][1,23]. The major smart city key enablers are briefly described as follows:

1.1.1. The Smart Grid

The smart grid is a future electrical network encompassing four major domains: bulk generation, transmission, distribution, and consumer segments. Moreover, uninterrupted communication between the domains is essential for the smart grid system ^[15][24]. Furthermore, the Internet of Things (IoT) and artificial intelligence (AI) are also integrated with the power grid in the case of the smart grid system ^[15][24]. In addition, the smart grid is not only a sustainable, cost and power-effective power supply solution, but it also provides high-quality, low losses, as well as a safe and secure power supply ^[16][17][18][25,26,27].

1.1.2. Smart Education

Distance learning is a common method of teaching in the current education system ^[19][20][28,29]. Moreover, future distance learning, or education systems, require a higher quality of Internet service, along with real-time interactions and lower power consumption ^[20][29]. Moreover, the improvement of the quality of remote learning and education will enhance the number of quality human resources, improve employment productivity, support influential institutions, and secure sustainable economic advancement ^[21][13].

1.1.3. Smart Transportation

Smart transportation is another key enabler for a sustainable smart city ^[22][30]. Smart transportation requires different communication technologies, such as vehicle-to-everything (V2X), vehicle-to-grid (V2G), vehicle-to-infrastructure (V2I), vehicle-to-vehicle (V2V), vehicle-to-pedestrian (V2P), vehicle-to-cloud (V2C), vehicle-to-device (V2D), etc. Smart transportation avoids collisions, saves energy, and improves road-safety, as well as increases the efficiency of traffic ^[23][31]. Thus, smart transportation can play a crucial role in the economic advancement of the smart city.

1.1.4. Smart Healthcare

Smart healthcare is a crucial key enabler for smart cities ^[24][25][32,33]. Smart healthcare can provide intra-hospital observation, remote patient monitoring, virtual healthcare, long-distance surgery, remote learning for doctors and nurses, remote diagnoses, emergency services, and remote consultations ^[26][17]. These applications required new technologies like virtual reality (VR) and augmented reality (AR) which require real-time, reliable, low latency, and flexible communication networks ^[1][26][1,17]. Smart healthcare ensures precise and high-quality medical care for the residents of smart cities.

1.1.5. Smart Farming

Smart farming introduces information and communication technology (ICT) in the agricultural and farming sectors of smart cities ^[27][34]. The IoT, sensors, actuators, unmanned aerial vehicles (UAV), robots, geopositioning, big data, and other ICT applications are to be introduced into the cultivation process of the agricultural sector ^[27][34]. Several evolving technologies are to be introduced into smart farming like smart machines, crop sensors, cloud computing, as well as the IoT for the smart control, analysis, and planning of farming ^[28][35]. The process of smart farming consists of smart analysis and planning, smart sensing and monitoring, smart control, and storing big data in cloud devices ^[29][36]. Smart farming not only reduces labor requirements, but it also enhances the caliber of production ^[30][37].

1.1.6. Smart Social Networking

The popularity of social platforms and people from different categories are connected through online social networking (OSN) ^[31][38]. The data from OSN provide social, cultural, and economic information, which are utilized by the commercial industries, authorities, policymakers, and governments of smart cities to realize the market trends and behavioral patterns of smart city residents ^[32][39]. Therefore, the effective progression of smart cities is linked to individual behavior within online social networking (OSN) platforms ^[33][34][40,41].

1.1.7. Smart Factory and Smart Warehouse

The smart factory is defined as an intelligent or digital factory ^[35][42]. A fully connected manufacturing system is introduced in the case of the smart factory, which can operate without any human force by data generation, transmission, reception, and processing for production purposes ^[36][43]. Smart factories introduce novel technologies in the case of production, such as a cyber physical system (CPS), AI-based decision making, intelligent data exchange models, human–machine interactions, cloud-enabled manufacturing and services, and comprehensive connections for the IoT for the effective production of goods and the minimization of human forces ^[35][36][42,43]. The smart warehouse provides high-quality services for the residents of smart cities ^[37][44]. Moreover, the smart warehouse enhances the productivity of supply chain management and also provides advantages to the stack holders of the communities within smart cities ^[38][45].

1.1.8. Smart Environment

The smart environment is a vital key enabler for the advancement of the smart city ^[39][46]. The smart environment changes the city and also reshapes the environment of the city for the settlement of the residents ^[40][47]. Moreover, various environmental challenges such as waste management and the pollution of the smart city can be effectively handled by the smart environment ^[41][42][48,49]. Furthermore, AI-based intelligent resource management in the smart environment is also a vital factor for smart cities. All of the functionalities of the abovementioned key enablers of the smart city, as well as smart power supply and management, are required to be uninterrupted. As mentioned earlier, the smart grid is a viable solution for fulfilling the power supply requirements of the smart city key enablers.

The smart grid holds immense significance for the advancement of the sustainable smart city. The key features of the smart grid communication network are its reliance on high speed, reliability, low latency, low jitter, flexibility, massive connectivity, higher area capacities, computational offloading, and secure data communication networks to effectively and intelligently manage complex power systems. Unlike traditional power grids, smart grids employ bidirectional communication, which greatly enhances their capabilities ^[43][102]. However, the specific communication demands and appropriate techniques can vary based on the particular environment and circumstance.

Figure 1 illustrates the flowchart regarding the operation logic and framework for implementing a smart grid by 6G. The goal and objectives of the smart grid communication network for the smart city applications are defined, such as massive connectivity and monitoring, secured communication for operation and resource management, and time-critical operations. The 6G-enabled smart grid infrastructure is designed based on these major requirements. The requirements are considered based on the goal and objectives of the smart grid for smart city applications. Afterward, all the requirements based on the objectives are verified. Subsequently, 6G-enabled smart grid infrastructure is deployed and integrated into the control and monitoring system. Afterward, the smart grid functionalities are tested and validated. The monitoring and management of a smart grid by 6G-enabled wireless communication is also performed accordingly, as is depicted in Figure 1.

At present, low carbonization is one of the prime concerns for the power grid systems of smart cities ^[44][103]. The integration of 6G and a smart grid is a suitable solution for enhancing the coordinated operation of the smart grid by a damping technique for low carbonization. A 6G-enabled smart grid can provide real-time monitoring and a control of energy flows. Moreover, the integration of 6G and a smart grid can also provide enhanced energy efficiency, reduction in carbon emissions, and the support of sustainable energy practices. In addition, this synergy provides real-time grid condition monitoring, demand responses, wide-area monitoring and control, advanced control, and efficient load management by the massive, high-speed, and low-latency 6G wireless communication. Grid resilience, cybersecurity, and supportive policies play critical roles in ensuring a sustainable and reliable smart grid ecosystem by utilizing sustainable, secured, and reliable 6G wireless communication. The technology of 6G could have a significant impact on the various applications of smart grid systems.

The technology of 6G wireless communication is a feasible solution for massive connectivity and the time-critical monitoring of smart grid systems. Furthermore, 6G can provide seamless communication among a massive number of devices, thereby facilitating real-time data exchange, comprehensive time-critical monitoring, and effective energy management. For time-critical monitoring, along with massive connectivity, the important KPIs are connection density, area traffic capacity, data rate, reliability, latency, energy efficiency, and spectral efficiency. PD-NOMA offers notable performance and advantages that facilitate enhanced data rates, spectral efficiency, massive connectivity, as well as low latencies to support massive connectivity along with the time-critical monitoring of smart grid systems ^[45][104]. In PD-NOMA, distinct transmission powers are assigned to multiple users according to their respective channel conditions. The superimposed signal, containing data for all users, is then transmitted to them ^[46][105]. Based on the reception and channel condition, the users employ SIC or direct decoding to decode their respective signal. The SIC decodes the signal of individual users sequentially from the superimposed signal based on their assigned power ^[47][106]. Therefore, PD-NOMA, along with OAM-MIMO (NOMA-OAM-MIMO), can provide massive connectivity, high data rates, and improved spectral efficiency. Moreover, IRS, THz, and BF can be incorporated with the D2D/cooperative relay-based NOMA-OAM-MIMO communication for smart grid UDNs to provide improved reliability, low latency, less power consumption, enhanced area capacity, and an enhanced coverage area of the UDNs of a smart grid ^{[48][49][50][51][52][53][54][55][56][57]}[81,83,84,90,91,93,95,96,97,101]. Furthermore, NOMA-enabled Q-learning UDNs can also provide high area capacities and massive connectivity for the UDNs of a smart grid system ^[58][59][94,107]. In addition, energizing the massive number of connected devices is another challenging issue. Thus, NOMA-incorporated wireless power transfers, such as NOMA-SWIPT, are required. They are needed to energize the IoTs, UAVs, robots, and edge/cloud devices, as well as provide energy efficiency ^[60][14].

The grid monitoring algorithms (e.g., PLeC/AB-PLeC/FIB/BW-PLeC) can be processed by AI-based edge/cloud computing devices 8. The energy devices (e.g., IoTs, UAVs, and robots) can communicate with the edge/cloud devices via the abovementioned novel techniques of 6G wireless communication system for time-critical smart grid monitoring ^[61][62][108,109]. The fault identification/location, as well as the deployment of the phasor measurement unit (PMU), are vital factors for grid monitoring ^[61][108]. For medium voltage grids, along with low latency monitoring, three different algorithms (path length constraint (PLeC), the application-level betweenness, along with PLeC (AB-PLeC), as well as the flow interference and bandwidth constraint (FIB)) were proposed in ^[61][108]. These algorithms are mainly focused on the strategic deployment of high-capacity links, thus aiming to strike a trade-off between deployment expenses and the achieved latency. The PLeC along with bandwidth and path length constraint (BW-PLeC) algorithms are introduced to achieve a low-latency communication framework ^[62][109]. These algorithms aim to enhance the performance of the conventional power-line communication technique by incorporating alternative enhanced speed communication links at crucial locations of the smart grid. The primary goal is to meet the specific delay requirements while minimizing deployment costs. All the requirements of the monitoring algorithms can be fulfilled by the abovementioned novel communication techniques of 6G.

The interconnection, reconfigurability, the wide variety of energy device types, and information networks introduce crucial implications for data security and reliability in smart grid systems ^[63][64][65][110,111,112]. Therefore, a massive amount of data are generated in case of bidirectional smart grid communication, and proper security and privacy are required for the operational and other confidential instances of data ^[17][66][67][26,113,114]. The operational integrity of the smart grid could be compromised by covert cyberattacks or the manipulation of data ^[68][115].

Thus, advanced encryption, authentication, and privacy mechanisms are required to protect the critical data transmitted between smart grid devices and control centers. In traditional artificial intelligence of things (AIoT) frameworks, vast amounts of identical energy-oriented data from the IoT devices of each user are transferred to cloud/edge devices for storage or decentralized processing ^[69][116]. However, this approach poses significant risks in terms of privacy infringement and potential data misuse. Federated learning (FL) is an attractive AI paradigm that ensures privacy while preserving data ^[69][70][116,117]. The energy data owners (EDOs) can collectively train the shared AI model, except in cases of divulging the energy-oriented data by an edge-cloud-assisted FL framework ^[70][71][99,117]. It facilitates efficient and fortified energy data communication for the respective smart grid users by integrating FL and the blockchain ^[71][99]. Furthermore, to address the challenge of limited knowledge regarding confidential multidimensional user data in real-world cases, a bi-layer deep re-enforcement-learning-inspired incentive algorithm is also introduced in ^[71][99]. This algorithm incentivizes active participation from energy data owners (EDOs), as well as encourages high-quality contributions to the shared model.

The software-defined IoT (SDIoT) in 6G makes the CPS more vulnerable to cyberattacks ^[72][62]. A novel graphics-processing-unit (GPU)-based adaptive robust state estimator named LSTMKF was introduced in ^[72][62]. This state estimator integrates a deep learning algorithm, and is specifically based on long short-term memory (LSTM) with an extended nonlinear Kalman filter. By combining these techniques, the LSTMKF state estimator enhances security measures and effectively manages the load within the system ^[72][62]. Edge computing is required for the computational offloading of LSTMKF due to computational complexity. However, the lightweight false state injection (FSI) can poison the state information of the edge devices in the case of deep-learning-based resource allocation ^[58][94]. Thus, to enhance security and privacy, an FL with the blockchain can be integrated for edge computing-based distributed computational offloading, as well as in also enhancing energy efficiency ^[69][70][71][99,116,117]. Moreover, the edge/cloud devices can be energized by the NOMA-SWIPT-based technique, which is discussed in an earlier subsection ^[60][14].

The real-time operations of smart grid systems, such as fault detection, outage management, island detection and management, energy management, and load balancing, can be supported by 6G. The network slicing based on non-real-time, close-to-real-time, and real-time operations is an effective solution for smart grid systems ^[73][118]. The time-critical applications of smart grid systems require area traffic capacity, spectral efficiency, latency, user data rate, and reliability. The novel 6G-enabled D2D/cooperative NOMA-OAM-MIMO, along with IRS, BF, THz, AI-based edge computing, and the real-time smart grid operational algorithms (e.g., EDGE/PMU-based voltage stability/phase-angle based island detection/learning-to-infer), can fulfill the KPI requirements and can also perform the time-critical operations of smart grids ^{[74][75][76][77]}[119,120,121,122]. Furthermore, wireless power transfer (SWIPT/NOMA-SWIPT) can energize edge/cloud devices for time-critical operational purposes. Moreover, the emBB and mMTC are suitable for non-real-time applications, and mMTC is appropriate for semi-real-time applications and ultra-reliable-low-latency-communication (URLLC). Both are appropriate for time-critical (real-time) smart grid operations ^[73][118]. The time-sensitive communication service-based AI-enabled 6G network slicing approach is a possible solution for accomplishing the time-critical smart grid operations ^[73][118].

For the time-critical smart grid operations, there are several novel algorithms that have been proposed for time-critical energy management, island detection, and outage detection and recovery. The energy management of the smart grid system regarding the demand response is a time-critical issue that can be overcome by a hybrid EDGE algorithm ^[74][119]. Edge is evolved by combining a genetic algorithm (GA) along with an enhanced differential evolution (EDE) algorithm ^[74][119]. The phase angle-based island detection method is much more resilient, more secure, computationally simple, and takes less time to detect the island of a smart grid system ^[78][123]. The PMU-based voltage instability indexed method was introduced in ^[75][120]. The proposed index can also detect the outage of lines/gens in an unsupervised manner ^[75][120]. The AI-enabled network slicing and intelligent cloud/edge computing can also overcome the computational issue within the provided time constraint for the proposed technique. The real-time detection of line outages is an immensely challenging task, particularly in scenarios where unknown line outages rapidly accumulate, thus leading to large-scale blackouts ^[76][121]. Hence, a pioneering learning-to-infer approach has been devised to detect the time-critical multi-line outages in a smart grid ^[77][122]. Additionally, the integration of AI-edge computing and 6G wireless communication techniques can effectively tackle the computational complexity involved in the process. Moreover, 6G-based novel communication techniques, which are mentioned above along with AI-enabled network slicing and intelligent edge/cloud computing, can process the algorithms for computational offloading, as well as effectively perform the time-critical operations of smart grids.

Based on the above discussion, Table 1 summarizes the major applications and respective suitable techniques for 6G-based smart grid communication networks.