20 major 5G key technologies

5G network technologies are mainly divided into three categories: core networks, backhaul and preamble networks, and wireless access networks.


 
Core Network

Key technologies of the core network include: Network Function Virtualization (NFV), Software Defined Network (SDN), Network Slicing, and Multiple Access Edge Computing (MEC).
 
NFV is to softwareize network functions through IT virtualization technology and run on general hardware devices to replace traditional private network hardware devices. NFV runs network functions on a general-purpose hardware device or white box in the form of virtual machines for configuration flexibility, scalability, and mobility, and hopes to reduce network CAPEX and OPEX.


Network devices to be virtualized by NFV mainly include: switches (such as Open vSwitch), routers, HLR (Home Location Register), SGSN, GGSN, CGSN, RNC (Radio Network Controller), SGW (Serving Gateway), PGW (Packet Data) Network gateway), RGW (Access Gateway), BRAS (Broadband Remote Access Server), CGNAT (Carrier Level Network Address Translator), DPI (Deep Packet Inspection), PE Router, MME (Mobile Management Entity), etc.

NFV is independent of SDN and can be used alone or in combination with SDN.
Software Defined Network (SDN) is a network design that separates the network infrastructure layer (also known as the data plane) from the control layer (also known as the control plane). The network infrastructure layer and the control layer are connected via standard interfaces, such as OpenFLow (the first open protocol for interconnecting data and control planes).

SDN decouples the network control plane to a common hardware device and centrally controls network resources through software. The control layer is usually implemented by an SDN controller. The infrastructure layer is usually considered as a switch. The SDN connects to the SDN controller and switch through a southbound API (such as OpenFLow), and connects the SDN controller and application through the northbound API.

 


SDN enables centralized management, increased design flexibility, and the introduction of open source tools with the benefits of reducing CAPEX and OPEX and inspiring innovation.
5G networks will be oriented to different application scenarios, such as ultra-high definition video, VR, large-scale Internet of Things, car networking, etc., different scenarios for network mobility, security, latency, reliability, and even billing methods. The requirements are different. Therefore, a physical network needs to be divided into multiple virtual networks, each of which faces different application scenarios. Virtual networks are logically independent and do not affect each other.

 


Network slicing can only be achieved after NFV/SDN is implemented. Different slices rely on NFV and SDN to be created through a shared physical/virtual resource pool. Network slices also contain MEC resources and functionality.


 
Multi-access edge computing (MEC) is a cloud-based IT computing and storage environment at the edge of the network. It enables data storage and computing power to be deployed closer to the user's edge, reducing network latency and better providing low latency, high bandwidth applications.

MECs can introduce new applications through the open ecosystem, helping operators to provide richer value-added services such as data analysis, location services, AR and data caching.
Backhaul refers to the part of the wireless access network connected to the core network. Optical fiber is an ideal choice for backhaul networks. However, in environments where fiber is difficult to deploy or costly to deploy, wireless backhaul is an alternative, such as point-to-point microwave. In addition, the wireless mesh network is also an option for 5G backhaul. In R16, 5G wireless itself will be designed as wireless backhaul technology, namely IAB (5G NR integrated wireless access and backhaul).

Frontanul refers to the BBU pool connecting the remote RRU part, as described in the C-RAN section. The capacity of the preamble link mainly depends on the wireless air interface rate and the number of MIMO antennas. The 4G preamble link adopts the CPRI (Common Public Radio Interface) protocol, but since the 5G wireless rate is greatly increased and the number of MIMO antennas is multiplied, CPRI cannot meet the 5G era. For the capacity and latency requirements, the standards organization is actively researching and developing new pre-transmission technologies, including sinking some processing power from the BBU to the RRU unit to reduce latency and forward capacity.
 
In order to improve capacity, spectrum efficiency, reduce latency, and improve energy efficiency to meet 5G key KPIs, key technologies included in 5G radio access networks include: C-RAN, SDR (Software Defined Radio), CR (Cognitive Radio), Small Cells, ad hoc networks, D2D communication, Massive MIMO, millimeter wave, advanced modulation and access technologies, in-band full-duplex, carrier aggregation, low latency, and low power technologies.
The Cloud Radio Access Network (C-RAN) software converts the wireless access network functions into virtualization functions and deploys them in a standard cloud environment. The C-RAN concept was developed from a centralized RAN with the goal of increasing design flexibility and computational scalability, improving energy efficiency and reducing integration costs. In the C-RAN framework, the BBU function is virtualized, centralized, and pooled. The RRU and the antenna are deployed in a distributed manner. The RRU connects to the BBU pool through the pre-transmission network. The BBU pool can share resources and flexibly allocate processing from each RRU. signal.

The advantage of C-RAN is that it can improve computational efficiency and energy efficiency, and it is easy to implement more advanced joint optimization schemes such as CoMP (coordinated multipoint transmission), multi-RAT, dynamic cell configuration, etc., but the challenge of C-RAN is pre-transmission network design and deployment. The complexity.
Software Defined Radio (SDR), which enables some or all of the physical layer functions to be defined in software. Note the difference between software-defined radios and software-controlled radios, which only refer to physical layer functions controlled by software.

Some traditional physical layer functions such as modulation, demodulation, filtering, channel gain, and frequency selection can be implemented in SDR. These software calculations can be performed on general-purpose chips, GPUs, DSPs, FPGAs, and other dedicated processing chips.
Cognitive Radio (CR), which makes behavioral decisions in real time by understanding the state of the wireless internal and external environment. SDR is considered to be an enabling technology for CR, but CR includes and enables a variety of technical applications, such as dynamic spectrum access, ad hoc networks, cognitive radio anti-jamming systems, cognitive gateways, cognitive routing, real-time spectrum management. , collaborative MIMO, etc.
Small Cells are small base stations (small cells). Compared to traditional macro base stations, Small Cells have lower transmit power and smaller coverage, usually covering the range of 10 meters to several hundred meters. Usually, Small Cells are based on the coverage. It is divided into microcells, Picocell and family Femtocell.

Small Cells' mission is to continuously supplement the macro station's coverage blind spots and capacity to improve network service quality in a lower cost manner. Considering that the 5G wireless frequency band is getting higher and higher, the 5G millimeter wave frequency band will be deployed in the future. The wireless signal frequency band is higher, the coverage is smaller, and the user traffic demand in the future multiple scenarios is constantly rising. In the latter 5G era, a large number of Small Cells will be deployed. These Small Cells will form a super-intensive Heterogenous (HetNet) network with the macro station, which will bring unprecedented complexity challenges to network management, frequency interference and the like.
Self-organizing network (SON) refers to a network that can automatically coordinate neighboring cells, automatic configuration, and self-optimization to reduce network interference and improve network operation efficiency.

SON is not a new concept. It was proposed in the 3G era, but in the 5G era, SON will be a crucial technology. As mentioned above, network densification in the 5G era presents unprecedented complexity challenges for network interference and management. It requires SON to minimize network interference and management, but even SON is afraid to cope with super-dense 5G networks. The CR (cognitive radio) technology mentioned above is needed to help.
Device-to-device communication (D2D) means that data transmission does not pass through the base station, but allows one mobile terminal device to communicate directly with another mobile terminal device. D2D originated from the 4G era and is called LTE Proximity Services (ProSe) technology. It is a short-range communication technology based on 3GPP communication system, which mainly includes two major functions:
• Direct discovery, direct connection discovery function, the terminal finds that there are terminals directly connected to it;
• Direct communication, direct communication, and data interaction with surrounding terminals.

In the 4G era, D2D communication is mainly used in the public security field. In the 5G era, IoT applications such as car networking, autonomous driving, and wearable devices will be greatly developed. The application range of D2D communication will be greatly expanded, but it will face security. And resource allocation fairness challenges.
One of the main ways to improve the wireless network speed is to use multiple antenna technology, that is, multiple antennas are used at the base station and the terminal side to form a MIMO system. A MIMO system is described as M x N, where M is the number of transmit antennas and N is the number of receive antennas (eg, 4x2 MIMO).

If the MIMO system is only used to increase the rate of one user, that is, multiple parallel data streams occupying the same time-frequency resource are sent to the same user, called single-user MIMO (SU-MIMO); if the MIMO system is used for multiple Users, multiple terminals use the same time-frequency resources for transmission at the same time, called multi-user MIMO (MU-MIMO), MU-MIMO can greatly improve the spectrum efficiency.

Multiple antennas are also applied to the beamforming technique, which adjusts the amplitude and phase of each antenna to give the antenna a specific shape and direction of the radiation pattern, so that the wireless signal energy is concentrated on the narrower beam and the direction is controllable. Increase coverage and reduce interference.

Massive MIMO is a larger number of antennas, and currently 5x is mainly used for 64x64 MIMO. Massive MIMO can increase the large wireless capacity and coverage, but faces the challenges of channel estimation accuracy (especially high-speed mobile scenarios), multi-terminal synchronization, power consumption and computational complexity of signal processing.
Millimeter wave (mmWave) refers to radio waves with an RF frequency between 30 GHz and 300 GHz, with wavelengths ranging from 1 mm to 10 mm. One of the biggest differences between 5G and 2/3/4G is the introduction of millimeter waves. The disadvantage of millimeter wave is that the propagation loss is large and the penetration ability is weak. The advantage of millimeter wave is that the bandwidth is large and the rate is high. The Massive MIMO antenna is small in size, so it is suitable for scene deployment such as Small Cells, indoor, fixed wireless and backhaul.
In the 4G era, OFDM technology is adopted. OFDM has the advantages of reducing inter-cell interference, anti-multipath interference, reducing the implementation complexity of the transmitter and receiver, and being compatible with multi-antenna MIMO technology. However, in the 5G era, due to the definition of enhanced mobile broadband (eMBB), large-scale machine type communication (mMTC) and ultra-reliable low-latency communication (uRLLC), these scenarios should not only consider anti-multipath interference, but also MIMO. The compatibility and other issues also put forward new requirements for spectrum efficiency, system throughput, delay, reliability, number of terminals that can be simultaneously accessed, signaling overhead, and implementation complexity. To this end, 5G R15 uses CP-OFDM waveforms and can adapt to flexible and variable parameter sets to flexibly support different subcarrier spacing and reuse 5G services of different levels and delays. For the 5G mMTC scenario, the Non-Orthogonal Multiple Access (NOMA) scheme has become the subject of extensive discussion because orthogonal multiple access (OMA) may not be able to meet its required connection density.
In-band full-duplex (IBFD) is probably one of the most desirable technologies in the 5G era. Neither FDD nor TDD is full-duplex because neither transmit and receive signals simultaneously on the same frequency channel, and in-band full-duplex can simultaneously transmit and receive in the same frequency band. The solution can increase the transfer rate by a factor of two.

However, the full double-industry in the belt brings strong self-interference. The key to realizing this technology is to eliminate self-interference, but it is worth mentioning that the self-interference cancellation technology is constantly improving. The latest research and experimental results have been made. The industry sees hope, but the biggest challenge is the complexity and cost.
Carrier Aggregation (CA), which increases the data rate and capacity by combining multiple independent carrier channels to increase bandwidth. Carrier aggregation is divided into three combinations: in-band continuous, in-band discontinuous, and discontinuous between bands, and the complexity is increased sequentially.

Carrier aggregation has been adopted in 4G LTE and will become one of the key technologies of 5G. The 5G physical layer can support aggregation of up to 16 carriers for higher speed transmission.

Dual connectivity (DC) means that the mobile phone can simultaneously use the radio resources of at least two different base stations (sub-master and slave) in the connected state. The dual connection introduces the concept of "diverted bearer", that is, the data is shunted to two base stations at the PDCP layer, and the PDCP layer of the primary station user plane is responsible for the PDU number, data splitting and aggregation between the master and the slave stations.

Dual connectivity is different from carrier aggregation, mainly in the layer where data offloading and aggregation are located.

In the future, 4G and 5G will coexist for a long time, dual connectivity of 4G wireless access network and 5G NR (EN-DC), dual connectivity of 5G NR and 4G wireless access network (NE-DC), 4G wireless under 5G core network Different dual connectivity forms such as dual connectivity (NGEN-DC) of the access network and 5G NR, dual connectivity of 5G NR and 5G NR will exist for a long time in the evolution of 5G networks.
In order to meet 5G URLLC scenarios, such as autopilot, remote control and other applications, low latency is one of the 5G key technologies. In order to reduce the network packet transmission delay, 5G is mainly implemented from two aspects: wireless air interface and wired backhaul. On the wireless air interface side, 5G mainly reduces the air interface delay by shortening the TTI duration and enhancing the scheduling algorithm. In terms of wired backhaul, the MEC is deployed to make the data and calculation closer to the user side, thus reducing the physics brought by the network backhaul. Delay.
mMTC is a big scene of 5G. The goal of 5G is the Internet of Everything. Considering the exponential growth of the number of IoT devices in the future, LPWA (Low Power Wide Area Network) technology is crucial in the 5G era.

Some LPWA (Low Power Wide Area Network) technologies are being widely deployed, such as LTE-M (also known as CAT-M1), NB-IoT (CAT-NB1), Lora, Sigfox, etc., with low power consumption, wide coverage, and low cost. The large number of connections is a common feature of these technologies, but these technical features are inherently contradictory: on the one hand, we reduce the power consumption, such as letting the IoT terminal go to sleep after sending the data, such as Reduce coverage to extend battery life (usually a few years to 10 years); on the other hand, we have to increase the transmission power per bit and reduce the data rate to enhance coverage, so weigh the pros and cons according to different application scenarios, Finding the best balance between these contradictions is a long-term issue of LPWA technology.

In the 4G era, two major cellular IoT technologies, NB-IoT and LTE-M, have been defined. NB-IoT and LTE-M will continue to evolve from 4G R13 and R14 to 5G R15, R16 and R17, which belong to the future 5G mMTC scenario. It is an important part of the 5G Internet of Everything.
Satellite communication access has been incorporated into the 5G standard. Compared with the 2/3/4G network, 5G is a “network of networks”, and satellite communication will be integrated into the 5G architecture to realize a seamless interconnection network composed of satellite, terrestrial wireless and other telecommunication infrastructures. Traffic flows dynamically across seamlessly interconnected networks based on bandwidth, latency, network environment, and application requirements.