Release Date：2022-07-25  Author：By Fei Teng, Ni Yanzi  Click：

5G deployment continues to accelerate worldwide, and the mission of 5G is to drive digital transformation of industries. Adapting 5G to industrial needs is key to the success of 5G, and also an enabler for transformation of operators. 
2021 is the first year of commercialization of 5G industrial application. ZTE launched the 5G precise RAN solution at the beginning of 2021. This solution runs through precise planning, precise slicing, precise identification, precise scheduling, precise measurement and precise O&M, which enables 5G to shift from best effort to deterministic communication. After a year of practice, ZTE upgraded the precise RAN solution to 2.0 with a series of innovative enhancements addressing the pain points of the industry. These enhancements include: top-level private network design and template solutions for 10+ industries; more precise slice resource reservation; EdgeQoS providing intelligent, flexible and precise service guarantee; and the ToBeEasy system providing easy O&M of private networks. With the E2E deterministic capabilities enhancement, the precise RAN 2.0 solution enables a more efficient 5G private network (Fig. 1). 

Precise Planning 2.0 
In precise planning 1.0, B2C and B2B planning differ in the coverage design standard (from overall performance to a specified spot), requirement model (from one dimension to three dimensions), and wireless scenario (from assumed to specific). Precise planning 2.0 has made two major improvements:
—The B2B simulation and big data platforms support the dedicated B2B features.
The precise B2B network planning should be focused on single user and single service, while the traditional B2C network planning only on overall performance estimation. ZTE has launched the simulation platform and big data platform to meet B2B planning requirements.
ZTE's global simulation platform supports simulation based on high-resolution (1-5m) electronic map and ray tracing model, indoor/outdoor collaborative simulation, and correction with measurement report (MR) or drive test data, and accurately predicts the network coverage. It also supports the construction of channel models, and the loading of user locations and tracks, and predicts the attainable rate of each terminal through a Monte Carlo simulation. The supported simulation scenarios include industrial parks, aviation coverage, drone inspection, sea coverage, and tunnels. 
ZTE's big data platform can implement precise and intelligent network planning and service orchestration. The big data platform collects multi-dimensional network data, and predicts the grid-level network coverage and user throughput to quickly determine whether the network meets the B2B service requirements. If not, the platform could give suggestions for network enhancement. 
—Top-level design of private networks for 10+ major industries
ZTE has accumulated extensive experience in 5G private network design. For each of the 10+ major industries, including power grid, coal mining, manufacturing, port, and steel-making, ZTE can provide a dedicated private network design, covering a series of innovative and customized products and technologies (e.g. cloud-network cabinet series, explosion-
proof base stations, miniaturized BBU for rail transportation, QCell without pBridge, site-level computing engine NodeEngine, and CampSite solution for movable private network). And leveraging a series of enhanced capabilities of 5G such as large uplink, low latency, high reliability and high precision positioning, the private 5G solutions could be replicated in an industry.

Precise Slicing 2.0 
Slice resource guarantee is the basis of service guarantee. Compared with slices with shared resources, slices with reserved resources ensure that the service experience will not be affected under resource constraints, and they provide deterministic resources to guarantee the SLAs of B2B applications. 
However, with the development of 5G in vertical industries, more and more private network slices need to be deployed; and the limited cell resources make it hard to reserve resources for all the slices. Precise slicing 2.0 supports calendar-based resource reservation, which means that the resource for slices can be allocated at different times to allow the deployment of more slices inside one cell. This significantly improves the network efficiency while guaranteeing the service experience, increasing the commercial competitiveness of a private network.

Precise Guarantee 2.0
Compared with the non-differentiated parameter setting and best-effort service of B2C, the B2B "slice +5QI" based scheduling could provide differentiated SLA guarantee. However, it is based on static and complicated service subscription and parameter configuration, and requires that the QoS subscription, SIM card, CPE and industrial equipment should be strictly bound, which means that any mismatch or change will affect the service guarantee, and any service addition/adjustment must be modified through the core network. Therefore, precise guarantee 2.0 incorporates the AI-based EdgeQoS to realize automatic service identification and precise closed-loop scheduling so as to achieve service guarantee easily.
Precise Identification
ZTE's EdgeQoS has innovatively introduced the AI-based intelligent service identification function to realize the dynamic identification of local services in parks, providing different applications in parks with the corresponding flow-level guarantees. In this way, only default subscription is required, and SIM cards, CPEs, and industrial equipment can be matched freely. As long as services are started, the EdgeQoS will automatically analyze and recognize the services, which greatly simplifies operators' subscription flows and service deployment. The efficiency of industrial application deployment can be improved by more than 10 times. 
Precise Scheduling
After the services are identified, the EdgeQoS matches them with the right steering and scheduling strategy, and performs closed-loop scheduling optimization based on the flow-level KPI measurement, guiding base stations to adjust radio parameters. And multi-dimensional enhancements are included in the above process:
—Precise service identification and scheduling: The EdgeQoS can identify the flow type and also the packet characteristics of a service, and then match the service with the corresponding resource scheduling function to improve the utilization efficiency of radio resources. For example, for video streaming service, the system can identify the I frame and P frame packets, and further identify their packet period, packet size and packet arrival time, so as to guide the base station to execute different scheduling strategies for I frame and P frame. Through such precise service identification and packet identification, a more refined service experience and network efficiency can be guaranteed. 
—Service orchestration to avoid I-frame collision: Industrial applications include a large number of video services, such as video surveillance, machine vision, and video based remote control. Generally, video service streams are composed of I frames and P frames, and I frames require much higher throughput. When multiple video services are concurrent in the network, it is likely that multiple I frames are transmitted at the same time, that is, I frame collision occurs, and the instantaneous requirement for the network uplink bandwidth can reach several hundred