Putting fronthaul at the center of 5G networks
As we move into a communications world offering new applications that are both highly capable yet highly demanding, fronthaul systems face stiff new challenges.
With the advanced capacities of 5G, fronthaul needs to offer seemingly infinite capacity. For example, if virtual reality becomes as ubiquitous as expected by 2025, fronthaul networks could find themselves needing to serve each user with nearly 6,000 GB per month.
Then there is the huge demand to get latency far below current levels. Again, advanced end user apps like AR and VR are challenging Communications Service Providers (CSPs) to get latency down to 100 µs and below.
Data processing spreads across the network
Fronthaul can be spread in a series of layers across the network between edge data centers and central data centers. In this type of layered architecture, computing will be distributed between these data centers depending on the type of data handling needed.
Central data centers handle signal driven functions such as voice and functions that don’t require massive data. In the same way, the edge is needed to support the latency requirement and localized, massive data.
But how much computing will be located at the edge and how much will be centralized to optimize overall costs, as well as support the business? As network functions become more distributed, traffic per VNF will decrease, thus reducing transport savings to the point where it is no longer able to offset the increase in VNF costs. As we distribute IP Edge functions to more locations, costs keep reducing up to a point and then start increasing again.
Converged Edge Cloud with Open RAN architecture
Two other trends are RAN cloudification and Multi-access Edge Computing (MEC) and xApps at the edge. Cloud RAN software provides Cloud Flexibility intelligence, Cloud automation, dynamic scalability, improved radio performance and seamless evolution to 5G.
An open ecosystem at the edge enables new services and business models through ORAN/RIC based xApps and MEC API and OpenAPIs for a multi-vendor ecosystem.
Nokia’s AirFrame Fronthaul Gateway (FHGW), part of Nokia’s All-in-Cloud BTS radio network architecture, enables legacy CPRI-based radios to be easily connected to new Ethernet-based All-in-Cloud BTS architecture, with the CPRI radio interfaces remaining unchanged.
The gateway’s main tasks are traffic handling and connection management; radio parameter configuration; radio synchronization; and managing 4G and 5G radio protocols including CPRI, eCPRI and OBSAI.
Connect legacy radio to cloud RAN
A superb solution to connect legacy 4G/5G radios to cloud-based RAN, FHGW is the first solution built and tailored to fully support edge and far-edge cloud deployments. Offering an extremely compact footprint, the AirFrame open edge server chassis is available in either 2U or 3U high versions, enabling straightforward installation on existing base station sites.
The solution also offers a real-time, OPNFV compatible, OpenStack distribution, built to provide the performance and low latency required by Cloud-RAN and MEC.
Nokia AirFrame FHGW will be fundamental to helping achieve customers’ high performance expectations for 5G, at a cost and efficiency that CSPs demand.
Share your thoughts on this topic by joining the Twitter discussion with @nokia or @nokianetworks using #5G #cloud #AirFrame #MEC #fronthaul