5G networks are moving in the direction of network diversification, broadbandization, integration, and intelligence. With the popularization of various intelligent terminals, mobile data traffic will show explosive growth towards 2020 and beyond. In the future 5G network, reducing the cell radius and increasing the number of low-power nodes is one of the core technologies to ensure that the future 5G network supports 1000 times traffic growth. Therefore, ultra-dense heterogeneous networks have become a key technology for improving data traffic in future 5G networks.
In the future, wireless networks will be deployed with a variety of wireless nodes that are more than 10 times larger than existing sites. Within the macro station coverage area, the distance between sites will remain within 10m, and support for 25,000 users per 1km2. At the same time, the ratio of the number of active users to the number of sites may reach 1: 1, that is, one-to-one correspondence between users and service nodes. The densely deployed network has shortened the distance between terminals and nodes, which has greatly improved the power and spectrum efficiency of the network.At the same time, it has also expanded the network coverage, expanded the system capacity, and enhanced services in different access technologies and various coverages. Inter-level flexibility. Although the ultra-dense heterogeneous network architecture has great development prospects in 5G, the decrease in the distance between nodes and the increasingly dense network deployment will make the network topology more complicated, which is prone to incompatibility problems with existing mobile communication systems. In 5G mobile communication networks, interference is a problem that must be solved. Interference in the network mainly includes: co-channel interference, interference from shared spectrum resources, and interference between different coverage levels. The interference coordination algorithms of existing communication systems can only solve the problem of a single interference source. In 5G networks, the transmission loss of adjacent nodes is generally not much different, which will cause the strength of multiple interference sources to be similar, which will further degrade the network performance, making the current It is difficult to cope with coordination algorithms.
Precise and effective sensing of neighboring nodes is a prerequisite for large-scale node collaboration. In an ultra-dense network, dense deployment causes a sharp increase in the number of cell boundaries, coupled with irregular shapes, leading to frequent and complex handovers. In order to meet the needs of mobility, new handover algorithms are bound to emerge; in addition, dynamic network deployment technologies are also the focus of research. Due to the sudden and random opening and closing of a large number of nodes deployed by users, the network topology and interference have a wide range of dynamic changes; and the small number of service users in each small station also easily leads to the spatial and temporal distribution of business. Dramatic dynamic changes occur.
Ad hoc network
In traditional mobile communication networks, manual deployment is mainly used to complete network deployment and operation and maintenance, which consumes a lot of human resources and increases operating costs, and network optimization is not ideal. In future 5G networks, network deployment, operation, and maintenance challenges will be faced. This is mainly due to the existence of various wireless access technologies in the network and the different network node coverage capabilities. The relationship between them is complex. Therefore, the self-organizing network (SON) 's intelligence will become an essential key technology for 5G networks.
The key problems solved by the self-organizing network technology mainly include the following two points: (1) self-planning and self-configuration in the network deployment stage;(2) network maintenance phase of self-optimization and self-healing.Self-configuration that is, the configuration of new network nodes can be plug and play, with low cost, easy to install and other advantages.The purpose of self-optimization is to reduce the business workload, to achieve the effect of improving the quality and performance of the network, the method is to measure through UE and eNB, in the local eNB or network management parameters self-optimization.Self-healing means that the system can automatically detect, locate and troubleshoot problems, greatly reducing maintenance costs and avoiding impact on network quality and user experience.The purpose of self-programming is to dynamically plan and execute the network while meeting the needs of system capacity expansion, business monitoring or optimization results
Content distribution network
In 5G, services such as audio, video, and image services for large-scale users have grown dramatically, and the explosive growth of network traffic will greatly affect the quality of services for users accessing the Internet. How to effectively distribute high-traffic business content and reduce the delay for users to obtain information has become a major problem for network operators and content providers. Relying only on increasing bandwidth does not solve the problem. It is also affected by factors such as routing congestion and delay in transmission, the processing capacity of the website server, and the like. The emergence of these problems is closely related to the distance between user servers. Content distribution network (CDN) will play an important role in supporting future 5G network capacity and user access
The content distribution network is a new layer added to the traditional network, namely the intelligent virtual network. The CDN system comprehensively considers the connection status, load status, and user distance of each node. By distributing related content to the CDN proxy server near the user, users can obtain the information they need nearby, which can alleviate network congestion and reduce response time To improve response speed. The CDN network architecture constructs multiple CDN proxy servers between the user side and the source server, which can reduce latency and improve QoS (quality of service). When the user sends a request for the required content, if the source server has previously received a request for the same content, the request is redirected by DNS to the CDN proxy server closest to the user, and the proxy server sends the corresponding content to the user. Therefore, the source server only needs to send the content to each proxy server, which is convenient for users to obtain content from a nearby proxy server with sufficient bandwidth, reducing network delay and improving user experience. With the advancement of cloud computing, mobile Internet and dynamic network content technologies, content distribution technologies are gradually becoming more specialized and customized, and they are facing new challenges in terms of content routing, management, push, and security.
D2D communication
In 5G networks, network capacity and spectrum efficiency need to be further improved. Richer communication modes and better end-user experience are also the evolution directions of 5G.Device-to-device communication (D2D) has the potential prospect of improving system performance, enhancing user experience, reducing base station pressure and improving spectrum utilization.Therefore, D2D is one of the key technologies in the future 5G network.
D2D communication is a direct data transmission technology based on cellular system.The data of D2D session is transmitted directly between terminals without forwarding through the base station, and the relevant control signaling, such as session establishment, maintenance, wireless resource allocation, billing, authentication, identification and mobility management, is still the responsibility of the cellular network.The introduction of D2D communication in cellular network can reduce the burden of the base station, reduce the end-to-end transmission delay, improve the spectrum efficiency and reduce the terminal transmission power.When the wireless communication infrastructure is damaged, or in the blind area of wireless network coverage, the terminal can realize end-to-end communication and even access the cellular network with D2D.In 5G networks, D2D communications can be deployed in both authorized and unauthorized bands.
M2M communication
M2M (machinetomachine, M2M), as the most common application form of the Internet of Things, has achieved commercial applications in the fields of smart grids, security monitoring, urban informatization, and environmental monitoring. 3GPP has formulated some standards for M2M networks, and has set out to start researching key M2M technologies. M2M is mainly defined in two broad and narrow senses. In a broad sense, M2M mainly refers to machine-to-machine, human-to-machine, and mobile network and machine-to-machine communication. It covers all technologies that enable communication between people, machines, and systems. In a narrow sense, M2M refers only to machines and Communication between machines. Intelligent and interactive is a typical feature of M2M that is different from other applications. The machines under this feature are also given more "wisdom".
Information Center Network
With the increasing proliferation of services such as real-time audio and high-definition video, the traditional TCP / IP network based on location communication cannot meet the requirements for data traffic distribution. The network shows a development trend centered on information. The idea of information-centric network (ICN) was first proposed by Nelson in 1979 and later strengthened by Baccala. As a new network architecture, ICN aims to replace existing IP.
The information referred to by ICN includes real-time media streaming, web services, multimedia communications, etc., and the information center network is the total collection of these pieces of information. Therefore, the main concept of ICN is the distribution, search and transmission of information, and it is no longer maintaining the connectivity of the target host. Different from the traditional TCP / IP network architecture centered on the host address, ICN uses an information-centric network communication model, ignoring the role of the IP address, or even using it as a transmission identifier. The new network protocol stack can implement functions such as information name resolution, route cache information data, and multicast delivery information at the network layer, which can better solve the problems of scalability, real-time, and dynamics in computer networks. The ICN information transfer process is an information transfer process based on a publish-subscribe method. First, the content provider publishes its own content to the network, and the nodes in the network understand how to respond to a request for related content. Then, when the first subscription sends a content request to the network, the node forwards the request to the content publisher, the content publisher sends the corresponding content to the subscriber, and the node with a cache will cache the content that has passed. When other subscribers send requests for the same content, the neighboring cached nodes directly respond to the corresponding content. Therefore, the communication process of the information center network is the matching process of the requested content. In the traditional IP network, the "push" transmission mode is adopted, that is, the server dominates the entire transmission process, ignoring the user's status, resulting in the user receiving too much spam. The ICN network is just the opposite. It uses the "pull" mode. The entire transmission process is triggered by the user's real-time information request, and the network uses the information cache to achieve rapid response to users. In addition, information security is only related to the information itself, not the storage container. In response to this characteristic of information, ICN networks use information-based security mechanisms that are different from traditional network security mechanisms. Compared with traditional IP networks, ICN has the advantages of high efficiency, high security, and support for client mobility.
