OFCOM report on PLT's (Home Plug Networking)

12 Conclusions and recommendations This study has examined the likelihood and extent of RF interference from in-home PLT devices over the next 5 to 10 years via two main routes; radiated emissions directly from the PLT user's home and interference from indirect PLT powerline radiation via a shared mains connection. In the case of both of these routes we have concluded that, provided uptake increases in line with our market forecasts, there will be a high probability of interference to HF users by 2020 if PLT device features do not change from those currently implemented. However, our results also indicate that interference mitigation techniques currently being developed by PLT device vendors and being discussed in international standards bodies will be sufficient to reduce this interference to negligible levels for most HF users. The exception to this is the safety critical aeronautical bands which will require notching by default. In addition our study has considered the potential for interference from emerging PLT devices at VHF. While this is less of a concern than at HF due to reduced power levels, we have still found that action does need to be taken to protect VHF users including safety critical aeronautical systems. Our conclusions across each of the interference mechanisms investigated can be summarised as follows: ? Cumulative effect of ground wave over very large areas - The likelihood of interference from PLT devices via this route is low. Instead the observed interference signal at the victim receiver is dominated by the closest PLT devices. ? Cumulative effect of sky wave over very large areas - Based on results from NATO, the likelihood of interference from PLT via the cumulative effect of sky wave is low. The exception to this is in quiet rural areas where interference is more noticeable. ? Cumulative effect of line of sight interference to airborne HF users - Based on results from the ITU, there is a significant chance of interference to these HF users. Both power control and notching combined with a maximum transmit power level of -63dBm/Hz are required to bring PLT signals below the interference threshold. ? Cumulative effect of radiated emissions to ground based HF users - The likelihood of interference via this route is similar across the three HF victim receiver types. The probability of interference is high with notching or power control alone but can be reduced to negligible if both techniques are used together. IARU bands are already notched by default in most PLT devices on sale today, but it is worth noting that earlier PLT devices did not include IARU notches and mayneed to be updated. ? Cumulative effect of line of sight interference to airborne VHF users

- Based on our modelling of aeronautical instrument landing systems we have found that a small variability in PLT transmit power level or antenna gain of household wiring could dramatically change the probability of interference. This probability is borderline if VHF PLT devices are widely adopted in the near future, rising to high in the longer term. We therefore recommend that these bands are notched by default. ? Cumulative effect of radiated emissions to ground based VHF users - Of the VHF victim receivers examined sensitive narrowband FM receivers and, to a lesser extent, FM broadcast receivers give most cause for concern and some bands may require notching. Nonetheless such receivers operating near the limit of their sensitivity may still have a low but measurable probability of interference. ? Interference from indirect PLT powerline radiation via a shared overhead cable - The likelihood of interference via this route is low to medium from now up to 2020 if PLT devices do not change. However, the introduction of power control is likely to reduce interference via this route to negligible levels. It should be noted that there is uncertainty over the filtering effects of UK consumer units which could affect this conclusion. ? Interference from indirect PLT powerline radiation via a shared mains connection close to the victim receiver - The likelihood of interference via this route is medium to high from now up to 2020 if PLT devices do not change. However, the introduction of power control is likely to reduce interference via this route to negligible levels. It should be noted that there is uncertainty over the filtering effects of UK consumer units which could affect this conclusion. Based on these conclusions we recommend the following: ? The maximum transmit power of PLT devices should be harmonised. Our conclusions are based on a quasi peak transmit power level of -50dBm/Hz which is the maximum transmit power level for UPA devices. However, currently PLT product manufacturers can vary the transmit power of PLT chipsets and the Homeplug AV standard operates at the lower level of

-55dBm/Hz. This ambiguity makes future interference levels difficult to predict and levels will vary depending which devices become dominant. From our extension of ITU results for airborne HF users we note that a maximum PLT power level of -63dBm/Hz (in addition to notching and power control) is desirable to protect these safety critical services. ? Power control and smart notching should be implemented. Power control and smart notching are anticipated to be available in Q2 and Q3 of 2010 respectively. Our results show that these features need to be introduced in these timescales to keep interference at manageable levels and recommend that these features are implemented in PLT devices in a timely manner. ? Notches in the IARU bands should be implemented. Most PLT vendors already notch the HF IARU bands by default and our results show that this is an essential interference mitigation approach. However, we have not yet seen notching of amateur radio bands at VHF. We therefore recommend that notching of amateur radio and other similar sensitive narrowband receiver bands are implemented both at HF and VHF. ? Notches in the HF aeronautical bands should be implemented. Our results show that notching and power control will be required to protect the HF aeronautical bands from PLT interference over the next 5-10 years. As the incoming signals are likely to be at low levels and of short duration we recommend that permanent notching rather than smart notching is applied. We have also assumed a protection radius of 100m in our analysis and this should be discussed with the operators of aeronautical ground stations. ? Notches in the VHF aeronautical radionavigation bands should be mandated. Our results for ILS localiser receivers indicate that the probability of interference is likely to become high during the next

5-10 years. We therefore recommend that these safety critical bands are notched bydefault. Additional options include: ? Interference between PLT and VDSL should be monitored. There have been suggestions of interference between VDSL and PLT devices where the cables are installed next to each other. Our assessment is that this could lead to interference only in very specific installation situations, but we recommend that Ofcom monitor this situation with BT as VDSL is deployed in the UK. ? The filtering effects of UK consumer units should be measured. Our results in relation to conducted interference are based on measurements that exist of PLT signal loss through consumer units, and lead us to recommend action on power control and notching. It would beprudent to test the filtering effects of a range of typical consumer units in the UK to ensure that consumer units with a lesser filtering effect are not present in large numbers. ? Measurements of radiated emissions from in-home PLT devices deployed in the UK. We found a lack of measurements of radiated emissions specifically from in-home PLT devices. One such study has been carried out recently in Canada measuring emissions outside a single PLT home. However, it would be useful to repeat this in the UK at distances covering the near field, far field and ground wave propagation to understand any dependency on UK-specific wiring practicesand to examine the effect as the number of PLT homes increases.
Reply to
Bob
Loading thread data ...

One problem is that the report assumes that interference to ham radio HF operation will be cured by installing notch filters for the ham bands. They are used on larger industrial systems:

I have yet to see a unit with such a filter on a HomePlug device.

The UK opposition:

Incidentally, some users are finding that PLT/BPL devices tend to rip on DSL signals when the power and phone wires are loosly coupled.

I have several slow DSL 1500/384 kbit/sec installs, where the HomePlug device is in the same cabinet and same power plug as the DSL modem. No effect that I can see. However, on one 6000/512 Kbits/sec DSL, there seemed to be some interaction, so I removed the HomePlug device.

Reply to
Jeff Liebermann

Not quite "Cumulative effect of radiated emissions to ground based HF users - The likelihood of interference via this route is similar across the three HF victim receiver types. The probability of interference is high with notching or power control alone but can be reduced to negligible if both techniques are used together." "can" and "negligible" to me means that there will be will still be some people who will get interference especially since some amateurs still like to experiment with equipment.

"Smart Notching has been investigated within ETSI with guidelines for signal detection and notch depth published in 2008 [8]. Plugtests, carried out in 2007 by ETSI using Sony and DS2 smart notching demonstrators, have shown the feasibility of smart notching and have given the encouraging results that shortwave radio stations were received as well when smart notching was activated as when there were no PLT devices active [9]. It is worth noting that the IP for smart notching via this route is owned by Sony and so there could be an additional cost to PLT vendors if this feature was required."

"It should be noted that Smart Notching does not necessarily operate effectively down to ambient noise level, but it has been demonstrated to the ?minimum planned signal level? of other systems such as short wave radio."

They noted that some form of monitoring with BT systems should take place as this seems to be an installation specific problem and perhaps some form of "advice" on installation issued. I am in the process of moving house and all the phone sockets in the new house are next to mains sockets so when I have time I will experiment.

Reply to
Bob

Please note that Smart Notching requires a *TRANSMITTER* to determine what frequencies to notch. The full name is Adaptive Smart Notch, which installs a notch in the broadband noise transmitted by the device if it detects a qualifying transmitted signal. That's tolerable for some ham operation, which usually transmits at or near the same frequency as reception. It's useless for receive only HF signals, such as DRM stereo, SWL, packet radio monitoring, HF APRS gateways, and others, where there's no transmitter available to convince the device to shut up.

From 5.3.4 in the Ofcom report. Smart notching is estimated to be available from Q3 2010. This provides a monitoring function in the chipset which will detect the presence of transmissions to which PLT devices may cause interference and apply a frequency notch as appropriate.

The stuff radiates well beyond your house. I pickup junk from BPL systems that are about 500 meters away. Probably farther if I had a decent antenna.

There's no interfernce between BPL/PLT and POTS phone service.

Reply to
Jeff Liebermann

If there is no "Transmission" then what do receivers detect? I haven't seen any specs regarding the Rx sensitivity for these "smart notch" filters but they did say "encouraging results that shortwave radio stations were received" so they must have agreed some form of "threshold" level.

My broadband is on the POTS line and the "mains sockets" refer to the electric supply in the UK. My broadband/phone at the moment interferes with my newly installed burglar alarm causing it to give a line fault indication every time I use the phone and the installer is going to have to reset the noise threshold level on the line so that I don't get plagued by text messages. I will be interested to see if there is a similar effect using "Homeplug" devices.

Reply to
Bob

Nothing. More bluntly, if there's no transmission, adaptive smart notching simply doesn't work.

Translation: You don't get to see the test results because they were all over the map, probably random results, and required considerable guesswork to call the results "encouraging". I don't have the spec but my guess is that it shouldn't be difficult to make a fairly strong RF signal run a frequency counter. I'll see if I can find the specified field strength.

Congrats. You have DSL, not BPL or PLT.

Different issue and a rather muddled description. There is nothing in your description that involves the power line. My guess(tm) is that you don't have a proper DSL filter installed seperating the POTS (plain old telephone service) audio from the DSL signal, or it's mis-installed, or mis-wired. In the USA, alarm circuits require a RJ-31X/RJ-38X filter.

If your alarm circuit doesn't have one of these, you're doing something wrong.

Reply to
Jeff Liebermann

My alarm circuit does have a hard wired filter, however according to the installer there was a known issue with "noise levels" on broadband connections as speeds increased, I currently get about 7Mb download, which meant the threshold on the alarm cct. needed to be increased. He was here this afternoon and reprogrammed the controller settings and now everything works as it should including all text notifications. My muddled description. My phone/broadband sockets, qty5, and electric skts/wiring are close together and as far as I can tell for 4 rooms the cables run run parallel to each other and are only a couple of inches apart so when I have time I will install a couple of Homeplug units and see if I pickup any interference to either the broadband connection or an increase in noise to the alarm. The modem/router is at the end of the phone cable run.

Reply to
Bob

I have probably misunderstood the concept but surely once a transmissions is received then that frequency is notched out for a given period of time and the "notch" is not automatically removed on cessation of the transmitted signal.

Reply to
Bob

Not quite what I expected. "In [Schw_08d] two criteria or thresholds are given, when a broadcast receiver is defined to be receivable: · Criterion (1): 14 dB above the noise floor. It is 3 dB lower than the desired SNR of an AM receiver to understand voice (derived from the acoustical measurements above) and around 11 dB lower than that required by a normal DRM transmission (out of [ITUR_03]). · Criterion (2): the absolute threshold of -95 dBm which is derived from equation 13 For activation, a processing time of 15 seconds is conceded to PLT modems. If the two criteria were passed once, the notch shall be kept for at least 3 minutes. This timing hysteresis is a tradeoff between consumer acceptance listening to a fading broadcast service and the processing capabilities of PLT modems creating a notch. The timings were found by the STF332 [Schw_07a] during the plugtests where the concept was verified [Schw_08c]

page 81

Reply to
Bob

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.