The reason why edge computing is important is that even when 5G is truly commercial, ultra-large bandwidth (eMBB) application scenarios can be realized, but the emergence of huge data volume means that it is necessary to find a connection point in the cloud and terminal transmission process, preprocess the data and choose whether to go to the cloud.

Edge computing application evolution

"Today‘s 4G network can basically solve 80% of the transmission problem, compared to the 5G network, which is only 30 milliseconds less." For existing applications, most of the current network transmission is not too much of a problem, and 5G is actually facing the adaptation problem of new applications.

"When 5G is in the Internet of Things, the big problem will be coverage. Once the base station is not covered in place, it is difficult to ensure 100% data transmission, and there may even be problems such as flash disconnection during operation. In addition, when a packet is transmitted from one end to the other, there will be a lot of retransmission, and it must be all passed directly and it will also involve the underlying protocol. So there are a lot of practical problems to solve. The person added that this is also the reason why edge computing is playing an increasingly important role in the 5G era.

5G will have overall changes in network architecture. In short, 5G networks are reconstructed network architectures through SDN (software-defined networking)/NFV (network functions virtualization). It itself has infrastructure equipment or servers at the edge of the network, which can directly migrate the capabilities of the data center to the edge of the network in a software-defined way, and dynamically adjust the capabilities of edge computing according to the demand of the network.

At present, the processing of IoT data by current users is placed in the data center, before transmission or at the edge, and the actual appeal is one-third each. However, the agency expects that in the future, as more analytics and AI capabilities are available at the edge layer, these analytics capabilities will be balanced between the edge layer and the core layer.

Edge computing – the peripheral nerve of intelligent transportation

In the foreseeable future, the rapid development of two industries is very clear, one is the communication industry, the other is the transportation industry, because the rapid development of society brings more information and physical communication, information exchange depends on communication, physical communication relies on transportation. Therefore, as a combination of the most advanced communication technology and transportation technology, the field of intelligent transportation has always been a focus of social attention.

Over the past decade, intelligent transportation has enabled exponentially growing urban vehicles to obtain real-time road conditions, China‘s world-renowned high-speed rail to maintain high-speed operation, and ocean-going sailors to easily communicate with their families in real time. However, many of the technologies expected by the industry, such as autonomous driving and unattended rail transit, have not yet been fully realized.

With the development of new technologies, we are gradually entering a fully connected "intelligent society", and "edge computing", a new technology in the field of Internet of Things, is applied in the field of intelligent transportation. This has ushered in the hope of solving many problems that have long plagued the development of the industry. Edge computing refers to the deployment of computing power and services at the edge of the network through the Internet of Things network, providing communication and computing services to nearby terminals, sensors, and users, and solving the challenges of massive heterogeneous connectivity, real-time services, business intelligence, data interoperability, and security and privacy protection of IoT systems in highly distributed scenarios. In layman‘s terms, in the future intelligent transportation application environment, "cloud computing" is equivalent to the brain of intelligent devices, processing relatively complex processes; And "edge computing" is equivalent to the nerve endings of smart devices, carrying out some "knee-jerk" responses.

The advent of edge computing makes intelligent transportation more secure. Whether it‘s road, rail, sea or aviation, safety is the most important thing in the transportation industry. For example, the most important reason why the autonomous driving technology that major technology companies have spared no effort to invest in recently has not been applied is also its inability to ensure absolute safety on the road. The arrival of "edge computing" has brought great help to solve these problems. As in humans, our first response to danger usually does not pass through the brain, but is "subconscious." For example, when an autonomous car is in danger and needs to stop in time, if it also needs to upload data to the "cloud", calculate the stop command, and then transmit it to the car, and the car will react. Then it is better to let the vehicle itself also have some computing power to deal with this problem. At the same time, we can also imagine a scenario where sudden natural disasters, signal interference or technical failures cause autonomous cars and trains in a certain area to fall into a network state. Then, they can only rely on the computing power given to them by edge computing to make a "knee-jerk" response to ensure their safety.