ZTE says AI will be unavoidable in next wave of virtualized networks

Telco AI (artificial intelligence) is gaining ground as a technique for operators to improve optimization of their networks, improve understanding of their customers, and take steps towards full automation. ZTE has been a particularly strong advocate, and last week it presented its latest developments to a seminar on network intelligence, held by the SDN/NFV Industry Alliance in Beijing.

The SDNFVIA was founded in 2015 by CAICT (China Academy of Information and Communications Technology) to promote innovative SDN and NFV commercial deployments. One of its key concerns, of course, has been the issue of management and orchestration (MANO) in virtualized networks, a vital area, but one split between rival approaches such as that of ONAP (Open Network Automation Protocol) and ETSI’s Open Source MANO.

For many, one of the transformational aspects of 5G will be full automation of MANO in virtualized networks, and ZTE’s findings give some pointers towards that nirvana. Wang Rui, the company’s MANO product OSS chief engineer, outlined the challenges for managing next generation NFV/SDN networks, which will incorporate new features such as multilayer networks, dynamic change, and virtual reality support.

At this point, ZTE argues, the need for AI to manage telco networks will become clear. Current networks, even virtualized ones, are fairly predictable and their variables can be handled by existing analytics and SON (self-optimizing network) approaches. But as they become fully dynamic and real time, that will change.

Wang Rui says AI can help telcos refine the models and rules suitable for network management in a big data environment, and assist decision making. “The introduction of AI provides the SDN/NFV network with vast potential for use in operation and maintenance, resource utilization and efficiency,” he said. “In our practice, the AI correlation algorithm and data cleaning improve the extraction efficiency of root cause analysis (RCA) rules by 70%, which greatly reduces the dependence on staff skills.”

Last month, ZTE said MNOs needed to accelerate their network AI efforts and announced plans for a platform which will incorporate SON capabilities as well as algorithms to support new interfaces based on natural language processing and facial recognition.

The Chinese vendor aims to offer an end-to-end platform which covers a wide range of telco-specific use cases from intelligent automated networks to new consumer services, and which incorporates the algorithms along with the chips and terminal hardware. The elements promised include ‘self-researching AI chips’, robot modules and intelligent terminals such as smartphones and smart home controllers.

“Complemented with high computing power, precision algorithm and data analytics capability, AI technology will lead to the evolution of highly intelligent autonomous, automatic, self-optimizing and self-healing networks,” ZTE said in its release. “At this stage, operators and vendors are still proactively exploring and seeking more efficient, stable and accurate AI algorithms and solutions to reduce the operation labor cost and effectively improve operating income. The platform can help operators introduce new technologies and build next generation intelligent network more conveniently amidst the ongoing advancement of AI technologies.”

ZTE’s inclusion of smartphone and controller devices in its AI portfolio indicates that the algorithms – which, before the days of cheap mass storage and compute power – required a supercomputer to run, can now be applied to a mobile gadget. Intel and Qualcomm have both recently demonstrated neural processing engines running on chips targeted at gateways or mobile devices. They effectively take snapshots of broader machine learning models which are created and modified in the cloud, and run them locally to reduce latency and improve context awareness (see Wireless Watch September 4 2017).

All this comes at a time when market research company IDC has released a report predicting that worldwide spending on cognitive and AI systems is forecast to reach $57.6bn in 2021, up from $12bn in 2017, which in turn was up 59% over 2016. Telcos are a fairly small part of this picture, something ZTE and its supporters believe may change. According to IDC, the five industries which will hit $1bn in spending on AI this year are retail, banking, discrete manufacturing, healthcare and process manufacturing.

About half of all spending on cognitive and AI technology will be on software, including applications and platforms, between 2016 and 2021, while hardware will be the smallest and slowest growing area of spending.