Should CISPR 22 Be Maintained As the Applicable Standard for the Supply of In-Home PLT

D-Link

Question 1:

Should CISPR 22 be maintained as the applicable standard for the supply of in-home PLT devices in Australia or should it be replaced by an alternative standard (e.g. EN50561-1 or ITU-T G.9964)?

We believe CISPR 22 is not the correct standard for inhome PLT. In Europe, where the deployment of PLT has been predominant, there has been an effort to harmonize regulation around PLT certification resulting in EN50561-1. Application of that norm has helped to clarify the regulatory aspect of PLT without creating any interference issue.

We believe that a standard such as EN50561-1 should be used as the baseline, as it collects the most up to date studies on PLT emissions. EN50561-1 only covers up to 30MHz though, but it mandates reducing the power when transmitter and receiver are close together, or when transmitter detects services using several bands. It also enforces permanent notches of different frequencies. To cover frequencies over 30Mhz, standard EN50561-3 could be used. Alternative, ITU-T G.9964 covers the whole spectrum up to 100Mhz, but it does not specify power levels when devices are close together, and the list of frequencies that are notched is more limited than in EN50561-1. Also, we don’t believe G.9964 provides enough protection to the services in the shared spectrum.

Regarding the specific concerns that ACMA has on the acceptance of EN50561-1:

The standard was written for European conditions where LV power distribution networks were primarily underground and any radiated interference would be suppressed.Underground wires are out of the buildings, where the in-home PLT signal strength is small. Main potential interference, if any, from PLT devices, will correspond to signals present inside the home, where the strength is significant. In this case, European and Australian networks are equivalent.
In Europe both the active and neutral lines are simultaneously switched on and off at the power-point (i.e. connected/disconnected). However, in Australia, mains wiring leaves the neutral line permanently connected and it could therefore act as an antenna and broadcast any interference on the line. This condition should not be relevant. If the concern is that Neutral line will radiate when connected, and in Australia Neutral line is permanently connected, and thus radiating permanently. Since Neutral line is connected at some point in time in Europe for an arbitrarily long time, and there hasn’t been any report of interference, then there is no such difference in behaviour.
If the PLT device detects a valid broadcast service, EN50561-1 allows the PLT device to continue to transmit for a further 15 seconds before having to reduce its power to minimise interference to the broadcast service. EN50561-1 already enforces several permanent notches for those frequently used services. Dynamic notches where the 15 second wait time applies are for infrequent services. Regarding the 15 second wait time, it has been taken as a safe number to ensure that there is a broadcast service running while avoiding detection false positives. We believe 15 seconds is a good trade off and should not significantly impact broadcast service.

Question 2:

If EN50561-1 is to be adopted, are modifications to the standard necessary to ensure that it is appropriate for Australian conditions?

ACMA should look into 30+MHz emissions (conducted and radiated) specified in EN50561-3. ITU-T 9977 should also be considered. It provides reference models and functionality of a mechanism to mitigate interference caused by in-home powerline devices to xDSL (implementing access Recommendations like ITU-T G.993.2 and ITU-T G.9701) and vice versa are specified.

Question 3:

Are you aware of any issues (in overseas jurisdictions) that have been associated with the adoption of EN50561-1 or the operation of in-home PLT devices generally?

Not that we are aware of. EN50561-1 is more restrictive below the 30MHz band in that it implements a dynamic power control mechanism, bigger notch depth, and dynamic notching that provide a wider protection to existing services operating below the 30MHz band. These features were not available in CISPR22 so the likelihood of interference has been reduced.

Question 4:

Are you aware of any empirical evidence or field trials in relation to interference between PLT devices and VDSL2 and G.fast services?

BT released a report to NBs (Notified Bodies) alerting on the possibility of MIMO PLT technology interfering VDSL2 lines based on a few measured homes. The conclusion of that report (with very few homes) was that devices transmitting in SISO using EN50561-1 (Line – Neutral) did not interfere with VDSL2.

Question 5:

Are you aware of any specific measures that have been successfully implemented (or are being developed) that will offer interference protection to VDSL2 and G.fast from in-home PLT devices?

Yes, There is an ITU effort (ITU-T G.9977) where the functionality of a mechanism to mitigate interference caused by in-home powerline communication devices to xDSL NT (implementingtransceivers in compliance with ITU-T Recommendations like G.993.2 and G.9701) are specified. Addressing various in-home network types and wiring topologies.The coordination of both the xDSL access network and in-home PLT network is facilitated by an arbitration function (AF), in order to reduce interference and optimize the performance of both networks and meet the throughput requirements of the end customer across both networks, by appropriately configuring relevant parameters of PLT devices and xDSL NT based on a coordination policy that is usually determined and provided by the operator(Scope of G.9977 Recommendation)

Question 6:

Are you aware of any impending developments in PLT technology and/or international standards that may reduce the risk of PLT interference?

EN50561-1, EN50561-3 and ITU-T G.9977 have been specified with that intent, by providing notching requirements to avoid interferences to radio services and DSL-PLC coordination protocol to avoid interferences to DSL services.

Moreover, PSD limits defined in EN50561-3 are very similar to those defined in ITU-T G.9964, but also adding notches to protect some Amateur and Aeronautical radio bands

EN50561-4 is going to start studying PLT MIMO certification.

Question 7:

What regulatory or non-strategies are appropriate to manage consumer awareness risks associated with the operation of PLT devices?

With regards to potential interference between PLT and VDSL2 or G.fast, any telco company providing such services should be aware of the PLT potential interference. As such they should have appropriate recommendation in the situation they have customer exposed to any drop in performance. Most of the time simple advice such as not having the PLT device at the same mains socket than the VDSL2/G.fast gateway helps to solve any issue.

Question 8:

Are there any other matters relating to the supply and use of in-home PLT devices that the ACMA should examine that have not been raised in this paper?