Preparing your network for low latency services
By Gavin Hayhurst, Head of Product Marketing
As 5G continues to roll out across the world, all eyes are on the promise of exciting services such as remote surgery, autonomous driving, immersive entertainment, augmented reality, and smart grid applications. And there is a good reason. According to Strategy Analytics, the revenue opportunity for autonomous cars alone will grow from $800 billion in 2035 to $7 trillion by 2050. To truly begin to monetize 5G, communications service providers must prepare their networks to ensure their investments will effectively support these lucrative Ultra Reliable Low Latency Communications (URLLC) applications.
To successfully launch these services two critical features are required: low-latency and high-reliability. For example, AR, VR, and Mixed Reality (MR) will place high demands on reliability, availability, and low latency. In the case of remote surgery, CSPs are not only in charge of delivering timely and reliable audio and video streaming, but will need to support things like haptic feedback, which gives remote surgeons the sensation to ‘feel’ what the robot’s arms are touching during surgery. According to 5G Americas, haptic feedback requires a tight delay requirement, with the end-to-end round trip time lower than 1 millisecond. This can only be supported by an extremely reliable system because of the life-threatening implications of any degradation in service.
With 4G LTE, latency is currently in the 4-millisecond range, but URLLC services require latency of 1-millisecond and 99.999 percent reliability. Achieving these requirements requires major capital investment and stringent planning to make sure that the network you are building will perform to these requirements.
A challenge for creating a URLLC-capable network is the growing catalog of diverse and competing use cases – each adding complexity to the network and their own capacity and latency demands. Before CSPs get to point of delivery, they need to ascertain whether their network can meet the latency requirements of not just one URLLC service, but each additional one as they are launched to the market.
This is where latency modeling comes to the fore. Latency modeling gives CSPs the assurance that as you build out your 5G network that it is future-proof and can meet the latency requirements of the different use cases. Once your 5G network is live, latency modeling can help determine whether your network can support a new service – and flag areas of the network that cannot meet the requirements. This will ensure that you do not assign a network slice to certain parts of the network where it may not meet the criteria for critical applications.
In addition, latency modeling will give you a whole-network view of latency, so when the load on the network changes, you can understand what the potential impact to latency will be. From here, CSPs will be able to dig deeper into their latency capabilities and answer questions such as, where do I have network coverage that can support autonomous vehicles, and where can I not? And what do I do about this? Do I need to change my equipment? Do I need to change my configuration? Do I need to add more small cells, or swap them out for different ones?
Until now, latency modeling has not been a high priority for most operators because, while download speeds and online browsing experiences will be better with 5G, it does not have life-impacting consequences. When we get to services like autonomous driving or remote surgery, however, latency becomes a matter of life and death.
Now is the time for network operators to start to assess their networks in terms of latency, and to ensure future expansion designs will meet the criteria for key services they plan to introduce. The latest release of ASSET, our radio network planning tool, introduces comprehensive latency modeling for both 5G and LTE across cells, terminal, services and backhaul, helping engineers to understand where network latency is good and where improvements are required.
As you continue to plan and rollout your 5G network, don’t miss the opportunity to get it right so that you can be the first to successfully launch and support the delivery of URLLC-enabled applications to the market.