AI-5G Introduction part -1

Home               LTE              NB-IoT          5G(NR-NSA)

Symbiotic relationship of AI and 5G:

In this section we will learn about the Artificial Intelligence (AI) in 5G. we all know that the recent breakthroughs in AI and computing have encouraged researchers in the field of wireless communications to study and leverage  AI especially in the 5G context.

As we know 5G provides higher bandwidth networks promising super fast speed and at the same time it also provides reliability and responsiveness. On the other hand artificial intelligence (AI) is technology that can analyze massive amounts of data and learn from previous actions. So if both AI and 5G network are combined together, they have the potential to pack a significantly more powerful punch and work as wonders.

Here when both (5G and AI) are combined AI can provide its intelligence to decode 5G’s complexity whereas the 5G network will provide a mechanism that can help in transmitting the large amount of data that AI system will require to process. 

We know that 5G networks are expected to achieve system capacity improvements using an increased spectral efficiency, higher densification and channel bandwidth.

The greater complexity of 5G network is one of the reasons for this correlative relationship. Now the question arises, why are they so complicated? The basic radio technology, which uses higher frequencies and requires smooth connection mechanisms. In addition to this the networks requires to adjust resources in real time to adapt to changing conditions. At the same time there are multiple number of devices from smart phones to robots to that are need to be connected together. Also many applications like medical equipments and self driving cars will give rise to large amount of data and requires enormous analytical capabilities. All this gives raise to the complicities.

Leveraging ML and AI in 5G

The application of ML and AI in wireless is still at its early stage and is gradually growing for creating smarter wireless networks. Machine learning and artificial intelligence will play a vital role in assisting wireless operators. This will help to operate, manage and deploy the 5G network with the multiplication of IoT devices. At the same time design, the network topology and models along with user’s mobility and usage patterns in 5G will not be easy.

AI and ML will together help in building more intelligence in 5G networks and systems and at the same time will allow a move from managing networks to managing services. This symbiotic relation of AI and ML can be used to in several use cases that will help wireless operators transition from human management model to self driven automatic management transforming the maintenance processes and network operations.

The ITU (International Telecommunication Union) classifies 5g into following three categories of usage scenarios to account for more diverse services and resource hungry applications :

·     Enhanced mobile broadband(eMBB) : eMBB is service category that addresses applications which are bandwidth hungry such as massive video streaming and virtual or argumented reality (VR/AR)

·       Massive machine type communication(mMTC) :  This is a service category that enables massive sensing, monitoring and metering in support of massive deployments of the IoT

·      Ultra reliable and low latency communication (URLLC) : This is a service category that supports latency sensitive services including autonomous driving, tactile internet and drones.

AI algorithms can analyze multidimensional and cross domain data. This allows for precise 5G network planning, smart MIMO and dynamic optimization of cloud network resources and coverage. Moreover AI helps is bringing the order needed by the disorder or confusion of 5G networks. This helps them to work more efficiently and smoothly.

Stay tune to learn more about AI in 5G………………….  

--Pinal Dobariya

Referance:

https://arxiv.org/ftp/arxiv/papers/1811/1811.08792.pdf

https://newsroom.cisco.com/feature-content?type=webcontent&articleId=2012215

http://thenetwork.cisco.com/