|
|
1 |
(% class="row" %) |
|
|
2 |
((( |
|
|
3 |
(% class="col-xs-12 col-sm-8 test-report-content" %) |
|
|
4 |
((( |
|
|
5 |
---- |
|
|
6 |
|
|
|
7 |
Mobile networks require precise time synchronization and deterministic transport to coordinate radio equipment and maintain service quality. In 5G deployments, fronthaul traffic must meet strict phase accuracy and low latency, while the transport network simultaneously carries mixed best-effort traffic. Quality-of-service mechanisms must ensure that congestion does not degrade time-sensitive streams. |
|
|
8 |
|
|
|
9 |
[[~[~[Figure 2: 5G xHaul Overview Topology~>~>image:486324520409366529_5G-xhaul-usecas-v7.png~|~|alt="Figure 2"~]~]>>attach:486324520409366529_5G-xhaul-usecas-v7.png||target="_blank"]] |
|
|
10 |
|
|
|
11 |
This test validates an end-to-end 5G transport scenario using a dual-domain Segment Routing architecture with SRv6-to-SR-MPLS interworking, representing a migration scenario. |
|
|
12 |
This use case was implemented and evaluated on the physical test bed illustrated in the following figure. |
|
|
13 |
|
|
|
14 |
[[~[~[Figure 3: Use Case Physical Topology~>~>image:487484161209794561_5G-xhaul-Physical-v12.png~|~|alt="Figure 3"~]~]>>attach:487484161209794561_5G-xhaul-Physical-v12.png||target="_blank"]] |
|
|
15 |
|
|
|
16 |
We configured two Segment Routing domains, one SR-MPLS and one SRv6, with IS-IS as the underlay IGP and BGP peering to a route-reflector for route distribution. At the domain boundary, we enabled SRv6-to-SR-MPLS interworking on the domain gateway so traffic could cross the encapsulation boundary and forward end-to-end between the access and core segments. On top of this transport, we enabled SyncE and Full Timing Support (ITU-T G.8275.1) PTP on the routers, established a baseline, and measured time error at the timing test points; the results stayed within the limit of max|ππΈπΏ| = 1100 ns from the Grandmaster to the output of the Access nodes. Additionally, the relative time error between the outputs of the Access nodes was measured, and all were within the +/-130 ns limit. |
|
|
17 |
To validate behavior under congestion, we applied quality of service toolset to realize Time Sensitive Networking (TSN) Profile A, so that all DSCP CS6 traffic received priority treatment. We generated fronthaul eCPRI traffic between the O-RU and the O-DU marked with priority code point (PCP) 6. This was translated to DSCP CS6 in the outer IPv6 header upon encapsulation in the SRv6 domain. PTP synchronization packets were classified in the same traffic class. We then generated background best-effort traffic between the traffic generator endpoints to emulate internet traffic and saturate the links. Under saturation, we measured the change in PTP two-way time error compared to the baseline condition, during which no difference was observed. We still observed small latency (microseconds) and no packet loss for eCPRI or other DSCP CS6 traffic, such as PTP. High latency (milliseconds) and drops occurred only in the best-effort background traffic, which confirms that the prioritization and congestion handling worked as intended. |
|
|
18 |
In addition to time synchronization and QoS validation, we used proactive OAM to verify 5G transport service health beyond control-plane checks. On the same testbed, participating nodes from Arista, Ciena, Cisco, HPE, Nokia, Raisecom, and ZTE were involved in TWAMP-based performance monitoring and gNMI telemetry collection via Cisco Crosswork Automation. Additionally, Cisco Crosswork orchestrated a background L3VPN over the SRv6 locator on the HPE nodes via NETCONF. |
|
|
19 |
Due to time constraints, we were unable to run additional scenarios. Topology variations, device role swaps, and alternative traffic paths remain valuable areas for future validation. |
|
|
20 |
|
|
|
21 |
(% id="prev-next-links" %) |
|
|
22 |
|[[< Previous>>doc:Main.EANTC Transport & Cloud Networks Interop Test Report 2026.Use Case Test.WebHome]]|[[Next ~>>>doc:EVPN Service Automation and Assurance]] |
|
|
23 |
))) |
|
|
24 |
|
|
|
25 |
(% class="col-xs-12 col-sm-4 test-report-sidebar" %) |
|
|
26 |
((( |
|
|
27 |
{{box}} |
|
|
28 |
{{include reference="Main.EANTC Transport & Cloud Networks Interop Test Report 2026.Sidebar Nav"/}} |
|
|
29 |
{{/box}} |
|
|
30 |
))) |
|
|
31 |
))) |
