March 24, 2021
The road to 5G has had its speedbumps along the way but the rollout is now at full throttle. With the deployment of 5G New Radio (NR), the realization of the next generation of wireless has led to a fork in the road, if you will. Current 5G implementation is not exactly what was expected a few years back. The reality of 5G has led to different priorities in designing UE, network elements, and the core network itself.
Fierce Wireless had a virtual conference – 5G Blitz Week – to discuss all things 5G. Anritsu was one industry leader to participate. Adnan Khan, Senior Manager – Wireless and Wireline Market Development, participated in a roundtable, during the event.
Also participating was AT&T, whose Senior Vice President of Wireless Technology and Experience Delivery, Igal Elbaz, served as a keynote speaker. During his talk, he stated that 5G is more than a radio access network (RAN) revolution. It’s a core network revolution.
Elbaz’s statement is accurate, as initial use cases are impacting the core network. As articulated by Anritsu Company GM and Vice President Robert E. Johnson in a Q&A session, those use cases are more in the industrial sectors than commercial applications. This view was echoed by speakers at 5G Blitz Week.
Low Latency in Industrial Use Cases
Low latency is a key 5G benefit for industrial applications. Let’s look at a smart factory. There are multiple cameras and AR technology throughout the facility to improve efficiency on the factory floor. This sector needs extensive bandwidth, low latency in the range of <1 ms, as well as multiple connections for reliability with micro- and macro-supported mobility among private/semi-private/public networks for session continuation. In some cases, there will be backup WLAN technology to provide a reliable and robust experience. A mix of low latency and high throughput is necessary to make industrial automation operate according to specification – and expectation – for the appropriate return on investment (ROI).
Mission critical connectivity use cases also need the low latency of 5G, in addition to high levels of reliability and security. One prime example is autonomous vehicles. Typically, four sensor groups form the AV toolkit in an autonomous car. They are video cameras, radar, ultrasonic sensors, and lidar. There must also be a high level of redundancy, requiring overlapping sensors to verify that what a car is detecting is accurate. For all these sensors to operate in the challenges of an automotive environment and in near real-time requires low latency.
Another use case is smart cities, which incorporate elements such as adaptive traffic signals and methods for improved first responder time of arrival. Smart city data is also used to create a digital twin, which helps make communities smarter, safer, and more efficient. City administrators and elected officials create virtual cities based on the data to plan transportation systems, prepare for environmental hazards such as flooding, and even warn citizens about high pollution areas.
Network Design Changes
Given all these use cases – and more to come – it’s no wonder that everyone involved in 5G, from carriers to chip makers, has dissimilar requirements. Every use case has a different network deployment. 5G is very complex and yet configurable at the same time. A variety of advanced features, like Ultra-Reliable Low-Latency Communication (URLLC), make 5G uniquely well positioned to meet all these requirements and capitalize on the market opportunities.
User plane latency is the first target to satisfy the requirements of low latency use cases. User plane latency is the time it takes to successfully deliver an application layer packet or message at the radio protocol layer from the service data unit (SDU) ingress point to the corresponding egress point (TR 38.913). Multi-access Edge Computing (MEC) creates the potential for high processing of large data amounts and a better means to optimize it by having it hosted at the edge of the network and closer to the end user. The goal is to bring the application as close to the user as possible.
From a test perspective, how do we measure latency in this network design? Over-the-Air (OTA) is the answer. Signals need to be measured from the gNodeB to the core network. Subsequently, tests must be conducted from the core network to the edge. Adding to testing complexity is what we refer to as a “data cocktail.” Its ingredients include AI, machine learning, autonomous vehicles, and cellular redirect.
Impact of High Throughput
Many consumer use cases shine a spotlight on high throughput – achieving 2 Gbps. Truthfully, consumers can’t tell the difference at this level of throughput. LTE, with 4x4 MIMO, can achieve 2 Gbps. Operators realize a huge benefit from the switch to 5G, however. 5G is much more efficient. Mobile carriers can send more bits per Hz, thereby using less spectrum. The result is a greater average revenue per user.
Consumer need for throughput is creating a shift from mobility to fixed wireless access. Gateways at the household are providing high throughput benefits, which have become particularly important during the pandemic. It is not uncommon for a parent to be on a business video conference from the home office while one child is enjoying a movie on a streaming service and another is immersing himself/herself in AR gaming.
5G Test Tools
When it comes to selecting the proper 5G test solutions, it’s important that the tools can verify the respective deployment and use case(s). One test instrument does not fit all applications, especially as it pertains to 5G.
For instance, millimeter (mmWave) brings challenges. The physics of mmWaves create considerable propagation loss. It requires OTA tests be made but that can be tricky. The reason is because OTA has propagation properties of its own. Test solutions must compensate for this phenomenon. For this reason, 5G device testing is becoming as complex as network testing.
Another test factor is virtualization and O-RAN, which create their own set of obstacles. Even though it leverages an open interface, engineers must test to ensure the interface is programmed correctly by the manufacturer. Testing must also be done to verify if an interface is compatible with hardware and software from different vendors. A test solution must have the protocol stack specifications properly implemented according to the specifications, as well as hardware that is robust and provides repeatable results. This becomes key when making latency measurements, as a millisecond delta can result in the difference in passing and failing.
Cost-of-test, while a contributing factor in any application, has greater importance with 5G. Many devices have a very low price point, making the cost of the entire design a priority. To meet this requirement, test tools must be built upon a flexible platform that can be easily and affordably expanded as use cases change and specifications evolve.
To learn more about 5G testing and solutions, visit our 5G technology page.
Comments
You can follow this conversation by subscribing to the comment feed for this post.