5G probably won’t be speaking TCP/IP as we know it
A new working group has been set up within ETSI to provide a central repository and thoughtspace for those who think that 5G will require new protocols that are more optimised to the properties of cellular networks.
An ETSI statement said that The Next Generation Protocols Industry Specification Group (NGP ISG) has been set up to, “provide a forum for interested parties to contribute by sharing research and results from trials and developments in such a way that a wider audience can be informed. An action plan to engage other standards bodies will be developed so that parallel and concerted standardisation action can take place as a further step in the most appropriate standards groups.”
That sounds like ETSI expects to see its proposals develop alongside, or in some cases within, bodies such as the IETF that are also looking at developing or enhancing protocols that are more optimised to mobile networks.
Forward leaps in the technology of the local access networks will not deliver their full potential unless, in parallel, the underlying protocol stacks used in core and access networks evolve.
The issue is that some think that current protocol stacks, notably TCP/IP, will not be sufficient to meet 5G network latency and throughput demands.
ETSI’s release put it thus: “The telecommunications industry has reached a point where forward leaps in the technology of the local access networks will not deliver their full potential unless, in parallel, the underlying protocol stacks used in core and access networks evolve. The development of future 5G systems presents a unique opportunity to address this issue, as a sub-optimal protocol architecture can negate the huge performance and capacity improvements planned for the radio access network.”
The underlying reliance of mobile access networks on TCP for the delivery of data across the network is something of a historical accident, and one that means that capacity and throughputs can often be strangled because of incompatibilities between TCP and the mobile network. In other words,even where cellular conditions may be acceptable for transport, kinks in the way the protocol responds to errors mean that performance and user experience suffers. (For more on this see this short presentation from Natasha Rooney, SDO lead on the GSMA’s Web Working Group and IETF contributor.)
ETSI’s NGP ISG has been assembled to try and find the best way of overcoming these issues. It will look at areas such as addressing and security as well as requirements from use cases such as ultra low latency, video and content distribution. It will also encompass requirements from network operators, including challenges with encrypted content.
Andy Sutton, Chair of NGP ISG said, “The TCP/IP protocol suite has undoubtedly enabled the evolution of connected computing and many other developments since its invention during the 1970’s. NGP ISG aims to gather opinions on how we can build on this momentum by evolving communication systems architectures and networking protocols to provide the scale, security, mobility and ease of deployment required for the connected society of the 21st century.”