Depth data processing device and method for base station

Abstract

The invention discloses a depth data processing device and method for a base station, and belongs to the technical field of telecommunication. According to the invention, a deep data processing device for deep learning and processing signaling control messages, IMS protocol messages and data messages is deployed between a base station and a UPF/SGW/SGW/core network, so that the mobility management capability and IMS conversation capability of mobile communication are kept; a management plane component sets formats of data messages transmitted by interfaces in a control plane component and presets a global configuration table and a distribution table which are used for loading transparent transmission and local forwarding processing information, the control plane component deeply learns messages and generates a forwarding table, and a forwarding plane component is used for realizing local processing and extremely-low-delay forwarding of GTP messages on a base station side. The access of more mobile terminals is realized within the limited base station spectrum bandwidth, and the access to a local network or a base station operator network is realized with lower delay, so that the local access quality of the mobile terminals is greatly improved, the spectrum utilization rate of the base station is greatly improved, and the mobile network is promoted to reach carbon neutralization and carbon peak-reaching indexes.

Description

technical field [0001] The invention discloses a device (Deep Data Processing Unit, DDPU) and method for deep data processing of a base station, relates to mobile communication network technology, and belongs to the technical field of electrical communication. Background technique [0002] Multi-access edge computing and mobile network local data centers migrate intensive computing tasks to mobile network edge servers near the base station, reducing the congestion and burden on the mobile core network and mobile transmission network, enabling data to be calculated and processed at the near end, and slowing down the mobility. Bear the pressure of network bandwidth, realize low latency, large bandwidth, improve data processing efficiency in the Internet of Everything era, and be able to quickly respond to data requests from mobile terminal users on the base station side and significantly improve service quality. [0003] The GTP protocol (GPRS Tunneling Protocol) provides a pr...

AI Technical Summary

Problems solved by technology

How to quickly forward GTP messages on the base station side is the key issue to increase the number of mobile terminal access. Due to the limitations of the von Neumann architecture, the server equipment has a gap of 1000 times in the forwarding and processing speed of the message compared to the actual demand. Unable to meet the characteristic requirements of low network latency

Conventional layer-2 or layer-3 routing and switching devices such as switches, hubs, splitters, and TAPs do not have the capability of in-depth processing and classification of packets, do not have the ability to analyze and process signaling packets of 4G / 5G mobile communication protocols at the base station side, and do not have IMS The analysis and processing of protocol signaling messages does not have the function of simultaneously realizing local forwarding and local low-latency communication after deep processing of data messages on the base station side

In addition, traditional splitters and TAP devices perform forced splitting processing on signaling packets / IMS protocol packets, leading to loss of control signaling, which will inevitably lead to dropped calls of mobile terminals, affecting the attachment of mobile terminals to the network; there are still some improvements The technical solution for message forwarding and processing speed needs to reconfigure the base station, terminal, core network, UPF / SGW, and mobile bearer network. By reconfiguring the UPF / SGW or core network side to distribute packets, this will undoubtedly increase the complexity of implementation. Spend

Images