With private networks connecting to many IoT devices, testing the device’s user interface requires updating test processes.
Many IoT use cases rely on private 5G networks because they offer greater network control, better security, more reliable performance, and dedicated coverage and capacity as opposed to using a public network. With these advantages, private networks play an important role in specialized use cases for vertical markets.
Based on current GSA data (Figure 1), manufacturing is the top industry vertical for private 5G, followed by mining and education. Other industries are expected to grow as long as one key hurdle — UE operation — can be overcome.
Figure 1. Vertical markets implementing private 5G networks (image: GSA).
Device performance in private 5G is a challenge because, while operator and private 5G networks have similar building blocks, UE is very device-centric based on use case. Additionally, 5G introduces control user plane separation (CUSP), which enables vendors to combine RAN and core network hardware components with software from other sources. With so many varieties in vendors, testing only against 3GPP specifications compliance is not enough.
You should properly test IoT devices against different configurations and combinations and ensure the key performance indicators (KPIs) are properly measured. For engineers, understanding all elements of how users can use the UE, as well as the environments in which they are being deployed, are necessary to ensure devices meet performance parameters.
3GPP Release 16 opens doors
3GPP Release 16 paves the way to private 5G networks. It lets 5G become a substitute for private wired Ethernet, Wi-Fi, and LTE networks by including multiple capabilities for industrial environments.
3GPP also provides standards and guidance on private 5G network deployment. Network architecture and deployment environment affect how you need to test an IoT device’s UE.
The most “private” architecture is a non-public network (NPN), which is an enterprise with a dedicated, on-premises network. 3GPP categorizes NPNs in two ways:
- Stand-alone non-public network (SNPN): this design does not rely on network functions from a public land mobile network (PLMN). An SNPN-enabled UE must be configured with a subscriber identifier (SUPI) and credentials for each subscribed SNPN identified by the combination of PLMN ID and NID (Network identifier).In addition, 3GPP Release 16 specifies the ability for a UE to obtain PLMN services while on a stand-alone non-public RAN. This is related to when the UE has a subscription and credentials to obtain services from both PLMN and SNPN.
- Public network integrated NPN (PNI-NPN): in this model, a PLMN ID recognizes the network, while a closed-access group (CAG) ID locates appropriate cells. A CAG cell broadcasts the designated CAG identifiers per PLMN, which must be supported by UE operating on the network. Only devices that have access credentials for that specific CAG ID can latch on to such cells, thus providing access restriction.
Hybrid private 5G networks use a mix of public mobile network components and dedicated on-premises elements. UE for hybrid networks has its own set of performance parameters, depending on network configuration. Three hybrid designs exist:
- Radio access is shared with the public network; everything else is private.
- The user plane is private, but the control plane and radio access are shared.
- Network slice option; one virtual slice is dedicated to the private network while all other elements reside on a public network.
Because private 5G networks use unlicensed and shared spectrum, device integration can become complex. Systems integrators, who have become key players in private 5G, must verify that UE operates according to specification, elements are integrated properly to guarantee end-to-end quality of service (QoS), and connectivity between UE and network is reliable.
Ensuring UE performance in private 5G
QoS and connectivity take on an added layer of complexity in many private 5G use cases. For example, in a smart factory, there can be robots with hundreds of sensors and machinery with multiple actuators operating in an environment with considerable interference sources. Such a setting has created the need for stress testing to determine how the UE will operate under such extreme conditions.
Given the proprietary nature of many private 5G networks, the prevalence of Open RAN architecture, and data sensitivities, security is a main priority. Many UE manufacturers employ practical security testing, which uses a network simulator to conduct necessary tests, such as functional security measurements. They thoroughly test all security-related functions inside UE or other systems under test to ensure correct behavior and operational robustness (Figure 2).
Figure 2. A typical test configuration for cybersecurity covers functionality, vulnerability, fuzzing, and penetration.
Stress tests and security are primary considerations but hardly the only issues for engineers. Private 5G networks have unique requirements that are more specific and varied than open public networks. Not only are there a tremendous amount of frequency/band combinations that must be considered for sunny-day testing but attention needs to be given to ensure devices that are supposed to work exclusively in an NPN environment do not connect to macro networks and unauthorized UE do not connect to an NPN. For this reason, other tests must be conducted to ensure performance, including:
- Connectivity — 5G IoT devices need proper testing to verify call connection, cell selection/reselection, access control, and any mobility implications in NPN environments. There are new features of 5G NPN that allow the device to selectively connect to the correct network. Verify that a private 5G network is truly only catering to private 5G devices.
- Compatibility — many devices used in a private network support cellular, Wi-Fi, and short-range wireless technologies, such as Bluetooth and Zigbee. Ensuring UE can seamlessly transfer from one technology to another is essential to private 5G network performance.
- Interference — given most private 5G network use cases, interference testing is critical. In addition to supporting multiple technologies, devices must operate in less-than-ideal real-world environments and in mission-critical scenarios. Engineers must have confidence product performance will not degrade due to interference before they are shipped to customers.
Creating a test environment
Implementing a test process to support private 5G UE requires a practical approach. The test environment must simulate real-world scenarios to efficiently verify that the UE will perform when deployed into a private 5G network. Design your test system with intuitive software to more efficiently support various and ever-changing test conditions and evolving standards, which will help to control test costs.
Conclusion
Private 5G networks play a significant role in the fourth industrial revolution. Engineers responsible for developing UE in these use cases must implement test processes that follow 3GPP standards and create real-world scenarios that precisely mirror the specific private 5G network environment. Such an approach will provide greater confidence that the UE will meet established KPIs.
