Wireless Communication Technologies for IoT in 5G: Vision, Applications, and Challenges

developing rapidly and achieving many breakthrough results. The advent of 5th generation mobile communication networks, the so-called 5G, has one of the most exciting and challenging topics in the wireless study area. The power of 5G enables it to connect to hundreds of billions of devices with extreme-high throughput and extreme-low latency. The 5G realizing a true digital society where everything can be connected via the Internet, well known as the Internet of Things (IoT). IoT is a technology of technologies where humans, devices, software, solutions, and platforms can connect based on the Internet. The formation of IoT technology leads to the birth of a series of applications and solutions serving humanity, such as smart cities, smart agriculture, smart retail, intelligent transportation systems, and IoT ecosystems. Although IoT is considered a revolution in the evolution of the Internet, it still faces a series of challenges such as saving energy, security, performance, and QoS support. In this study, we provide a vision of the Internet of Things that will be the main force driving the comprehensive digital revolution in the future. The communication technologies in the IoT system are discussed comprehensively and in detail. Furthermore, we also indicated indepth challenges of existing common communication technologies in IoT systems and future research directions of IoT. We hope the results of this work can provide a vital guide for future studies on communication technologies for IoT in 5G.


Introduction
The development history of mobile communication systems demonstrated that aim to meet the requirements of humanity, the data rate of mobile communication is constantly being improved and achieved breakthrough results. Mobile generations have evolved through 5 periods, starting from 1G to the current 5G [1]. Network generations from 1G to 3G have shown the continuous evolution of services and speeds. The 4G was proposed in the early 2000s. 4G was the first network generation entirely based on the IP packet switching method [2]. After about ten years of implementation, the former advantages of 4G have converted into disadvantages. Nowadays, 4G has access speed has become too low with high latency [3]. Humanity needs a solution to connect with data rates up to Gbps. The advent of the next-generation network called 5G in the early 2020s marks a comprehensive digital society. In particular, in 5G, a new concept is considered the Internet of Things (IoT) [4,5]. IoT is an integrated system of advanced technologies and solutions that allows devices, people, platforms, software, and solutions to be connected through the Internet [6,7].
According to Cisco, more than 500 billion devices will be connected to the Internet by 2030. These devices will be endogenously equipped with IoT modules that allow device-to-device (D2D) communications to each other, forming IoT networks [8]. IoT applications will be deployed in almost all humanity domains, including smart cities [9,10], smart transportation [11,12], smart agriculture [13,14], and smart homes [15]. In [16], we presented a detailed survey of IoT applications for humanity. We illustrate several typical IoT applications as in Figure 1.
However, the survey results also showed that IoT networks in 5G have a series of challenges such as performance improvement, support QoS, saving energy, privacy, and security [17,18]. Communication solutions including architecture, routing algorithm, protocol, and spectrum have been proposed to solve these problems. In this study, we conduct a comprehensive survey of communication technologies for IoT in 5G. The main contributions of this survey are listed as follows: The evolutionary history of network generations has proven that each generation is born to correct the weakness of previous generations and do some things that the previous generation could not do [19]. In the early 2020s, the Internet of Things concept was born simultaneously with the emergence of 5G [20]. Therefore, to define the vision of IoT in 5G, we need to clarify the advent context of IoT in 5G.
For the convenience of following the article, we have compiled the acronyms in Table 1. 1.1. Forming of IoT in 5G. The development history of mobile communication systems began in the early 1980s. During its development, mobile radio communication sys-tems always tend to integrate all systems. End-user devices are smarter more and more, lighter, save energy, support all types of data such as voice, video, and real-time multimedia applications. The data rate and bandwidth increase with costs decrease. The 1G-3G network generations are standardized and deployed widely worldwide, in references [1,21]; so, we will not consider these issues to focus on presents the 4G network generation.
The 4 th mobile network generation (4G) is formed after 3G and before 5G. Besides the provided services of 3G, it also provides added services such as broadband Internet access, IP phone (VoIP), video conferencing, online games, high-definition Internet TV, 3D TV, and cloud computing. The two technologies were standardized for 4G as Wimax and LTE [21]. One difference with previous generations, 4G unsupports the traditional circuit switching mechanism but relies entirely on IP protocol with the packet switching mechanism. Aim to speed up data transmission, spectrum modulation technologies of previous generations are replaced by OFDMA technology, combined with MIMO multipoint transceiver mechanism and smart antenna. [22]. As a result, the bit rate in 4G is significantly higher than in 3G.
With many advantages mentioned above, 4G has become a pioneering technology and commercialized in many countries. In Vietnam, 4G was deployed in 2016 [23]. However, after many years of deployment, 4G has revealed the limitations of this network generation. According to Cisco, over 500 billion devices will be connected to the Internet in the future. This is beyond the provided capacity of 4G [24]. Moreover, the delay of 4G is too large for the real-time applications, approximately 10ms, and the data rate of 4G is relatively low, approx. 3 (Mbps). With the number of devices increasing hundreds of times today, 4G will consume a huge amount of energy.
The limitations of 4G were indicated that the advent of 5G is an inevitable trend. Humans need a new network  [25] as follows: Aim to achieve these goals, in Table 2, many breakout technologies and solutions need to be implemented synchronously. However, like previous generations, the improvement of the radio access layer has always been a significant challenge to meet the goals of 5G. In this study, we approach 5G from an Internet of Things perspective. The concept of IoT was first mentioned in 5G. IoT is an advanced technology that allows things, machines, devices, solutions, and people to connect through the Internet. IoT is expected to become popular in all areas serving people, such as smart agriculture, smart transportation, smart cities, health, rescue and disaster recovery, retail, management house, and green energy. A very diverse survey of IoT applications is presented in [26].
1.2. The Architecture of IoT in 5G. IoT in the 5G framework consists of main four-layer architecture, as shown in Figure 2, and is related to data collection, processing, analysis, and sharing of information between equipment and communication networks.  A specific illustration of this architecture is presented in Figure 3. The sensor devices of IoT applications interact with the IoT gateway based on low-power communication networks such as SigFox, LoRa, or NB-IoT. These IoT gateways collect information from IoT devices and then transmit it to the Cloud through the 5G backhaul communications. In the middleware layer, the collected data is processed and stored, combining autonomous decision-making systems or human controls to make under layer tasks.   The company predicts that IoT devices will generate over 55% of global data by 2025. In order to accelerate the application of IoT in various areas serving humanity, Intel is developing an IoT ecosystem at all layers of the IoT architecture with the key technologies and solutions [27] as follows: (i) In thing layer, Intel providers unique performance scalability with four processor families for IoT applications. Besides, processors of Intel run a variety of operating systems such as Linux, Microsoft, and Google (ii) In network layer, Intel supports many networking interfaces and protocols to provide the necessary connectivity. Besides, Intel also provides Gateway solutions for the IoT (iii) In middleware layer, Intel server technology is extensively used in the network and cloud infrastructure. Moreover, Intel is focusing on three IoT computing projects, including edge computing, cloud computing, and AI and computer vision (iv) In application layer, Intel provides foundations to support the IoT application in various other domains, such as Figure 4 1.3.2. Samsung. According to Samsung, the total number of IoT devices is expected to increase to 21.5 billion by 2025.
The number of devices also increases to 34.2 billion if it includes smartphones, laptops, and tablets. Furthermore, Samsung also forecasts that the global IoT market will archive around $1.600 billion. Relying on the expectation that all devices will be connected to the Internet, Samsung has built IoT ecosystems in 5G to realize aspirations such as smart homes, smart cities, smart factories, healthcare, smart agriculture, and logistics [28]. Some of the recent developments in the field of IoT are as follows: (i) In application layer, Samsung is providing IoT solutions that allow users to control home appliances. The Samsung electronic devices such as TVs, washing machines, and refrigerators can be controlled by remote based on a Samsung smartphone (ii) In middleware layer, Samsung is implementing research projects related to optimal computing solutions, specifically edge computing, cud computing, and AI vision (iii) In thing layer, products and devices designed for Samsung IoT platforms, including phones, tablets and wearables, digital signage, and automation solutions. In particular, Samsung designed the unique IoT modules, called Samsung ARTIK modules, which can be customized based on the size, ability, and capabilities of the Samsung products. Moreover, the Samsung ARTIK Smart IoT platform combines open-source modules and cloud services with an ecosystem of tools and partners that is motivation to drive the development of the IoT in 5G. Figure 5 is an illustration of the Samsung Artik 530 development kit Ericsson. According to Ericsson, the expected IoT numbers of connections would increase over 3.5 times from about 1.7 billion in 2020 to approx 6 billion by 2026. Erricson also forecasts there will have over 24 billion interconnected IoT devices Internet by 2050. Consequently, almost everything is around us as home appliances, vehicles, traffic lights, personal devices, learning devices, and health monitoring would be connected to the Internet. This will be a very exciting area both in academic and industrial research in the coming years. With the ambition to connect anything, anywhere, Ericsson is driving the growth of the IoT through its major contributions in the domain of realtime network performance and cloud computing solutions [29]. Some researches dedicated by Ericsson for IoT in 5G are as follows: (i) In application layer, besides developing IoT solutions and applications for a wide range of fields such as healthcare and smart agriculture, Ericsson developed an IoT Accelerator Developer Portal to support the development of IoT solutions for the community of application developers worldwide, as presented in Figure 6 (   Figure 7. Along with the achieved breakthrough study results by top telecommunication corporations, a series of research labs around the world are also driving the research process to find promising solutions for IoT in 5G aim enhance data rate, exploit spectrum more efficiently, extend communication distances, optimize energy consumption,   It could be applied in a series of different domains from popular applications in life such as payment utilities, smart retail, manage home appliances to expert apps such as selfdriving vehicles, monitoring traffic status, collision warning between vehicles and monitoring, and controlling green energy systems, smart cities management. In the agriculture area, IoT can also be applied in applications such as forestry management, farm management, monitoring forest fire, tracing, and tracking products. In the industrial area, actuators and robots with the support of AI technology can perform tasks day and night replace humans with extremely high productivity and accuracy. It realizes the dream of smart and green factories.

Survey of Recent Communication Solutions
Advances in the semiconductor, electronics, and automation industries are driving the development of communication solutions for IoT in 5G. These solutions are smarter, more reliable, robust, high data rate, and energy saving. As a result, various low-power communication technologies have been proposed for IoT in 5G, such as SigFox and LoRa. Survey results have demonstrated that low power technologies are suitable for IoT 5G networks due to their unique characteristics such as wide coverage, low power, high energy efficiency, and suitable data rate. In this section, we present the recently proposed communication solutions for IoT in 5G. We divide these proposals into four categories based on technology. The detailed survey results are presented in the following subsections and are summarized in Table 3.

SigFox. SigFox technology was introduced in the 2010s
to connect low-power devices such as electricity meters and smartwatches, which need to be continuously operated on and have extremely low data rates. SigFox uses the industrial, scientific, and medical radio band, which uses 868 MHz in Europe and 902 MHz in the US with a channel bandwidth of 100 MHz. SigFox uses a wide-reaching signal that passes freely through solid objects, called ultra narrowband and    The figures indicated that the maximum number of sensors that can transmit data at the same time is approximately 100. The results indicated that, as the number of sensors increases above 100, the network performance could be decreased. Moreover, this study also proposes solutions to improve performance, large-scale, and high-density of sensors in SigFox IoT networks.
In [33], Mikhaylov et al. (2019) evaluated the performance of SigFox communication technology in the real world. Specifically, they deployed a SigFox-based communications network at 311 different locations in Brno city, Czech Republic. Then, they conduct tests to evaluate the performance and characteristics of the radio channel. The experimental results show that the packet delivery ratio achieved over 94% in the urban environment in the real world.

ZigBee.
ZigBee is a communication technology that uses the IEEE.802.15.4 standard and operates in the industrial, scientific, and medical radio frequency bands. It is a low-power wide-area communication solution for IoT networks. ZigBee technology in IoT networks has advantages compared to other communication technologies because of its simplicity, flexibility, and low cost. The transmission distance of ZigBee is about 100 m, with a data rate that is about 250 kbps, depending on power output and environmental features. ZigBee is typically used in extreme-low data rate networks, short-range, and long-lasting battery life such as home automation, medical device data collection, and industrial equipment control.
In [40], Pirayesh et al. (2021) proposed a ZigBee receiver based on MIMO against jamming attacks for IoT networks. This work designed a prototype of the ZigBee receiver based on MIMO technology and a learning mechanism to mitigate the unknown interference. The experiment results demonstrated that the proposed system could provide an average of over 26.7 dB jamming mitigation capability compared to the traditional ZigBee receiver.
In [41], Farha et al. (2021) introduced a new security schema based on a timestamp against replay attacks for ZigBee networks. This solution improves energy consumption significantly. Besides, to enhance feasibility, this solution uses powered devices to provide energy for powerconstrained devices with the current timestamp. The proposal is designed to be suitable for all ZigBee networks. The experiment results indicated that the proposed solution improves significantly against ability reply attacks in the ZigBee-based IoT networks.
In [42], Ali et al. (2019) designed the smart sensors that combined two communication modules include ZigBee and LoRa, to measure temperature and humidity factors for IoT applications. The collected data from sensors are sent to the central receiver unit by using the ZigBee or LoRa transceiver modules. The choice of the communication module can be controlled remote or based on the Cloud. The practical design and experiment figures indicated the benefits of the low-power, long range communication solutions for IoT applications.  (2) improve NB-IoT protocol through the improvement of the k -means algorithm to cluster NB-IoT devices and perform a scheduling strategy based on the priority. The experiment results indicated that the proposed uplink traffic scheduling schema enhanced performance compared to existing uplink traffic scheduling schemas.
In [45],  introduced a method to design the physical layer of the NB-IoT device. The focus of this work presents the characteristics and the scheduling of downlink and uplink physical channels at the NB-IoT base station and end-user device to help readers without having to read all the 3GPP specifications.

Discussions, Challenges, and Open Issues
In this study, we have highlighted the revolutionary contributions of IoT in 5G in a wide range of fields to serve humanity. Low power communication technologies will play an essential role in supporting and driving IoT applications more public. The survey results have indicated that many applications have been presented in Table 4. The proposals are applied in a variety of domains such as environment, city, home, building, factory, and agriculture.
Communication technologies such as ZigBee, SigFox, LoRa, and NB-IoT have advantages such as low energy (iv) Security for cloud servers: with the huge amount of data is provided by IoT devices, cloud infrastructure will be a possible solution in storing and processing big data. Consequently, the security of cloud servers will be one of the significant challenges In [47], Zhou et al. (2021) presented a survey comprehensive of security logic bugs in IoT devices, platforms,  [49], Wang et al. (2018) presented potential risks at the application layer, including data collection, storage, and data processing in the cloud of cloud-based IoT systems. Attacks into cloud servers to gain control or execute tasks to affect autonomous devices in smart factories and farms. In [50], Hassija et al. (2019) presented a diverse survey of attacks and security threats and proposed several architectural solutions to mitigate attacks on IoT systems.
In our opinion, security is one of the most critical problems of communication solutions in the IoT 5G network. This issue will continue to be a research topic timely and attract both academic and industry researchers in the future.

Energy Efficiency.
Assuming that when IoT applications in 5G become popular, tens of billions of IoT devices will operate and transmit data continuously day and night. As a result, it will consume a huge amount of energy while energy resource is exhausted day by day. This is not feasible. Therefore, energy-efficient communication solutions are a real challenge.
In In our opinion, energy efficiency can be considered based on some of the solutions as follows: (i) Communication technology-based: integration of smart, flexibly, and low-power communication technologies such as NB-IoT and ZigBee. In [17], the authors presented a survey of the energy harvesting communication technology for autonomously power IoT devices. This technology promises green energy in the future (ii) Trade-off based: In reality, performance and energysaving have an antagonistic relationship. Therefore, a smart, flexible trade-off solution should be considered.
In [54], Couso et al. (2018) proposed a trade-off solution for inverters to balance energy saving and performance for IoT-based smart grid applications (iii) Cloud-based IoT networks: cloud will continue to be the backhaul infrastructure for IoT applications due to its robust storage, computing, and processing ability. However, cloud services have a high response time due to the edge computing solutions that are proposed. Consequently, an intelligent offload schema to optimal resource allocation between cloud and edge servers should be considered. In [55], Aljanabi et al. (2021) proposed a hybrid fogcloud offloading schema to optimal performance and energy for IoT applications

Conclusion
In this study, we introduced the vision, architecture, wireless communication technologies, and research timelines of IoT in 5G. Based on the analysis of the core components for IoT in 5G, we conducted a short survey of low power communication technologies for IoT in 5G. The survey results showed that the Internet of Things would be the future of humanities, where all things such as software, systems, and people are connected through the Internet. The advent of IoT in 5G led to the formation of a series of applications serving humanity, such as smart homes, smart cities, smart agriculture, smart factories, green energy, and IoT systems. Besides, we have provided a full picture of promising communication technologies for IoT in 5G such as SigFox, LoRa, Wi-Fi, and LoRaWAN. These solutions are suitable for the operating characteristics of IoT networks such as large coverage areas, high energy efficiency, and low energy consumption level, which support a large number of IoT devices. Moreover, the survey results also point out some challenges of communication technologies for IoT in 5G, including (1) privacy and security and (2) saving energy. In our opinion, the security and saving energy problems of communication technologies will continue to be exciting research topics in the future and receive attention from both academic research and industry. We hope that this study will play an important role as a guide for future research on communication technologies for IoT applications in 5G.

Data Availability
No data were used to support this study.

Conflicts of Interest
The authors declare that they have no conflicts of interest.