zuruck zur Themenseite

Articles and background information on the topic

Schildknecht

Meinrad Happacher | Meinrad Happacher,

Profisafe via 5G

How efficient is a 5G infrastructure in terms of functional safety? The third and final part of this series of articles describes laboratory experiments carried out with Profisafe via Profinet within a 5G network using typical industrial components.

© Schildknecht

The tests were carried out at the '5G-Industry Campus Europe' in Aachen. The 5G-Industry Campus Europe is Europe's largest 5G research network and spans a total of four factory halls and 1 km2 of open space where industrial use cases are being investigated. The factory halls of the WZL, the Fraunhofer IPT and the FIR at RWTH Aachen University are comparable to industrial production environments with their mechanical equipment and processes. The test setup used for the 5G test essentially consists of three components:

  • the 5G communication infrastructure,
  • an S7 PLC and an IO device (both Siemens) running a Profisafe application with functional safety
  • and a 5G device, IPCs and two Dataeagle devices from Schildknecht, which tunnel the Profinet protocol from the PLC to the IO device via the 5G network.

The radio data systems

Schildknecht 's Dataeagle devices are data radio systems that wirelessly connect the PLC and the IO module, enabling wireless Profinet communication between these two devices while converting the layer 2 Profinet frames into layer 3 IP packets. In addition to monitoring, the wireless data systems pre-process the Profinet fieldbus telegrams to maintain fieldbus communication regardless of which connection is used between the two nodes in order to meet the requirements of the control technology (e.g. transmission time requirements, number of retransmissions). In addition, they keep the state of the Profinet protocol stable using pre-processing algorithms to meet hard cable-based timing requirements of the controller. Without such a procedure, fieldbus errors can occur, leading to machine downtimes of minutes if a data packet reaches its destination outside the Profinet/Profisafe time window.

Advertisement

Figure 1: The testbed of the first scenario with a 5G device (UE) and a central PLC connected by cable via the 5G campus network (UE to local breakout).

© Schildknecht

The 5G infrastructure

Ericsson's 5G Industry Campus Network Europe in Aachen was used as the communication infrastructure for the 5G testbed. The infrastructures at each location of this network contain all the nodes required to build a fully capable, state-of-the-art 5G network. This 5G campus network is based on the B40 band (2.3 GHz) as an anchor band and uses the NR band N78 (3.7 GHz) with a bandwidth of 100 MHz for data transmission. The current 5G infrastructure is based on Release 15 of the 3GPP 5G specification. The infrastructure is also equipped with a 5G RAN and core network, which is currently not optimized for latency-optimized communication (URLLC).

Figure 2: The testbed of the second scenario with a 5G device (UE) and a central PLC (UE), which is also connected via 5G (UE to UE).

© Schildknecht

On the mobile side of the test setup, a Qualcomm X55 chipset-based 5G device is used to establish the wireless network connection. The IPCs have the functionality to perform layer 2 tunneling, as this is not possible directly on the UE (user equipment). In addition, 5G technology only supports IP connectivity (layer 3). The user equipment establishes the 5G connectivity to the base station. In the testbed, this function is performed by a 5G device.

The two testbeds 'UE to Local Breakout' and 'UE to UE' are intended to simulate two real cases in a production environment and are shown in Figures 1 and 2: The measurement data can be read directly from the Dataeagle radio data system (on the left in the picture) via its maintenance and configuration interface; here, the diagnostic function is used to determine the latency of the communication link.

The requirements for the test cases

For the best possible safety communication in conjunction with Profinet, a 5G system must meet the following requirements: The Profinet update time of the PLC must be set to 1 ms, which means that a Profinet packet is generated and transmitted to the IO module every millisecond.

The standard TDD pattern (Time Division Duplex) with dynamic scheduling must be used in the 5G system; optimization in the 5G radio access network, such as pre-sheduling, is deliberately omitted.

The data exchange time

Figure 3: The complementary cumulative distribution function (CCDF) of the data exchange time when using WLAN (blue) compared to 5G (orange: testbed 1 and green: testbed 2).

© Schildknecht

The data exchange time (DX) serves as a key indicator for the performance tests. This can be understood as the application time. The watchdog time (safety monitoring time) is also defined in Profisafe. The data exchange time should be less than the safety monitoring time, as the application switches to the safe state if this is exceeded. In both testbeds, the data exchange time must be measured on the side of the IO module (see Figures 1 and 2 on the previous page).

Figure 3 shows the distribution of the measured data exchange time for an exemplary set of Profisafe data transmitted via different wireless communication systems (5G and WLAN). As expected, the visualization of the transmitted 5G data confirms the comparatively low variance of the exchange time. However, the most important transmission characteristic is the maximum transmission time, which is determined by the Profisafe protocol to determine the connection status. Even when measured over a long period of time, 5G shows no outliers in the distribution diagram, resulting in a maximum transmission time of 25 ms (orange and green curve, above) in the data set under consideration. For comparison: When using IEEE 802.11 WLAN, the maximum transmission time was 200 ms.

Thomas Schildknecht is CEO at Schildknecht.

© Schildknecht

A comparative measurement with Profinet via IEEE 802.11 WLAN in the 2.4 GHz band and the 5G test setup described above shows that both 5G and WLAN meet the requirement for "communication of functional safety". The measurement for WLAN was carried out in a shielded laboratory without any other interfering radio systems. This can be seen from the fact that 100% of the telegrams were received. The latency time for WLAN is lower than in the 5G campus network due to the non-applied co-existence mechanisms. In a real production environment, however, a higher latency was observed with IEEE 802.11 WLAN, as 802.11 WLAN has to coexist with other license-free communication technologies.

All IEEE 802.11 devices operate in the unlicensed spectrum, while 5G devices operate in different licensed bands. However, no coexistence mechanisms are required in licensed bands that limit the performance of IEEE 802.11 in the license-exempt spectrum. In Europe, the harmonized standard (HS) EN 300 328 defines the rules for operation in the 2.4 GHz ISM (Industrial, Scientific, and Medical) band.

Martin Stümpert is 5Gang project coordinator at Ericsson.

© Schildknecht

The coexistence mechanism used by IEEE 802.11 is described as Listen-before-talk (LBT) with truncated exponential backoff. According to this mechanism, the radio channel is classified as unoccupied if there is no signal above certain threshold values. An IEEE 802.11 device then waits for a random duration to reduce the probability of two or more devices transmitting at the same time. In the latter case, this is called a collision, which renders both transmissions unreadable and requires a retransmission.

For the measurement results, this means that in practice there is an undefinable non-deterministic influence on the WLAN latency time. It should therefore be noted that the latency observed over WLAN is unpredictable and non-deterministic. This is the key approach for the benefit of 5G campus networks: Here there are no interfering external influences, 5G does not need to implement coexistence measures and the latency is predictable and deterministic.

  • Xing Icon
  • LinkedIn Icon
Advertisement
Back to topic page
Advertisement

You might also be interested in

Advertisement
Advertisement
Advertisement
Advertisement

EUCHNER

Customized solutions for smart automation

In a world of increasing automation, functional safety is essential. Plant operators not only have to comply with legal requirements, but also ensure efficiency and availability. For more than 70 years, EUCHNER has been supplying customized...

read more...
Advertisement
Advertisement
Advertisement
Subscribe to our newsletter
Advertisement
Back to home