2nd 6G-PDN Workshop: 6G Programmable Deterministic Networking with AI

We plan for a half-day workshop with an open call for papers to be presented in two technical sessions. In addition, we will invite experts from the field to provide their view stemming from IEEE 802.1 TSN, IETF DetNet and 3GPP TSC to foster the exchange of different viewpoints.

Over the past decade the mobile communications system has transformed into a fundamental infrastructure that supports digital demands from all industry sectors. 6G is envisioned to accelerate the path started in 5G for catering to the needs of a wide variety of vertical use cases, both current and emerging. That will however require major enhancements to the current 5G capabilities, in terms of well-established key performance indicators (latency, reliability, bandwidth etc.), as well as new key value indicators (sustainability, trustworthiness etc.) that are currently being formed. Creating a deterministic network, understood as end-to-end time sensitive, reliable, and predictable communications, remains a challenge for cellular networks, especially in an industrial context. Wired deterministic communication standards have already emerged including Time Sensitive Networking (TSN) and Deterministic Networking (DetNet) and 5G has specified mechanisms for interworking with those standards. However, the available support of 5G in conjunction with TSN and DetNet is not sufficient for future end-to-end time-critical applications. Network programmability and predictability is key in supporting emerging 6G networks in achieving their promises of increased performance and flexibility at a lower cost. Deep network programmability, that is the ability to program the network fabric both vertically (control and data plane) and horizontally (end-to-end from the radio to edge and core network), is expected to characterize the new generation of mobile networks (6G), currently under development, towards supporting extreme performance requirements and service-specific operations. Enriching next-generation mobile networks with data plane programming capabilities can bring significant benefits with regards to network slicing and multi-tenancy, dynamic traffic engineering, offloading network functions to the data plane, etc. Coupled with edge solutions and advanced machine learning algorithms for prediction and network automation can help to overcome network performance limits, e.g., in terms of latency and deterministic delivery.