WO1999003209A2

WO1999003209A2 - Coupling communications signals to a power line network

Info

Publication number: WO1999003209A2
Application number: PCT/GB1998/002009
Authority: WO
Inventors: Robin Paul Rickard; Matthew Grant Fradley
Original assignee: Northern Telecom Limited
Priority date: 1997-07-11
Filing date: 1998-07-08
Publication date: 1999-01-21
Also published as: AU8233398A; WO1999003209A3

Abstract

Communications signals are transmitted over a power distribution network comprising a substation (10) having a plurality of power lines (40, 41) leading from it for distributing electricity to a plurality of premises S1, S2 and a plurality of communications stations (TRX) coupled to the power lines. Communications signals are coupled to the network at busbars (25) of the substation for transmission over the plurality of lines (40, 41) to the communications stations. Signals can be coupled to the two ends of the busbars and one of the feeds modified with respect to the other whereby to control the combined power that is delivered to the lines.

Description

COUPLING COMMUNICATIONS SIGNALS TO A POWER LINE NETWORK

TECHNICAL FIELD

This invention relates to an apparatus and method for coupling communications signals onto a power line network.

BACKGROUND OF THE INVENTION

It has been proposed to transport telecommunications signals over a power line. International Patent Application WO 98/06188 A1 (Northern Telecom) describes such a network. Delivering a telecommunications service in this manner is attractive as it overcomes the need for installing cabling to each subscriber, which is one of the greatest costs in providing a new telecommunications network. Existing power lines are used to carry the telecommunications signals to subscribers.

Power line transmission is one of several ways for delivering communications to subscriber premises and is therefore sensitive to competition with existing copper wires and more recent alternatives such as optical fibre/coaxial cable delivery and fixed radio access techniques. Therefore, there is a desire to provide a system that is capable of delivering an acceptable quality of service at a cost that is attractive to a subscriber.

In the above-mentioned WO 98/06188 A1 communications signals are injected onto one or more of the power lines leading from an electricity substation at a position on the line near to the substation. This incurs considerable expense in excavating to locate a buried power line and then working on the line to install a coupling unit to allow communications signals to flow to/from the line.

SUMMARY OF THE INVENTION

The present invention seeks to provide an improved way of coupling communications signals on to a power line distribution network.

A first aspect of the present invention provides method of transmitting communications signals over a power distribution network comprising a substation having a plurality of power lines leading from it for distributing electricity to a plurality of premises and a plurality of communications stations coupled to the power lines, the method comprising:

- coupling the communications signals to the network at busbars of the substation for transmission over the plurality of lines to the communications stations.

Coupling communications signals to the line at the busbars allows a large number of subscribers to be served in a cost-effective way. This avoids the need to make connections to the power lines at other positions where there will be the additional time and expense of excavating to locate the power lines and to provide a housing for the coupling equipment. The term "at the busbars" includes coupling signals both on and directly adjacent to the busbars. There are a number of different ways in which the coupling can be achieved.

There may be a single or multiple connection points to each busbar. The connection point can be on the busbar itself or to a power line leading from the busbar. The connection can be achieved by a number of different methods, such as by physically contacting the busbar by clamping a connector around the busbar, or by replacing a fuse unit in the one of the power lines leading from the busbar by a replacement fuse unit having a coupling device incorporated into it. A suitable replacement fuse unit is described in International Patent Application PCT/GB 98/00173. Where a single coupling point to each phase line is used, this is preferably provided at a central point on the busbar.

It has been found that, unexpectedly, a single connection point can result in an unequal split of power between the power lines leading from the busbars, even though the busbars are short in comparison with the RF carrier wavelength (typically <5%.) The variation in power levels between the power lines can be of the order of 20dB. This requires a high level of transmit power to reach the furthest subscribers on the most weakly-coupled power line. A high transmit power results in a large radiated power from the line. Multiple connection points to the busbars are more expensive and may pose a safety hazard because cable runs across the busbar array need to be minimised.

Providing connection points at two points on the busbar, and preferably, the two ends of the busbar, has been found to offer a more equal split of communications signal power between the power lines leading from the busbars. This permits a longer transmission range for a given injected power level, or a lower transmission power for a given transmission range. This is an important consideration as it reduces radiation from the network, and avoids the use of repeaters. This further reduces costs. Also, communications cable runs can be conveniently routed to avoid the power distribution frame which is preferable for safety reasons.

The two (or more) connection points can be fed by a signal which is equally split among the connection points. Advantageously, the connection points can be fed by signals which differ from one another in terms of their relative amplitude and/or phase. In this manner it is possible to preferentially transmit along a particular one of the power lines by arranging the combined signal power from the connection points to be greater along a particular one of the lines. One reason for doing this is to preferentially transmit along a power line to which a wanted subscriber is coupled. The weighting of signals can be changed according to the subscriber that it is required to communicate with.

This minimises signal power (and hence radiated power) from parts of the network that it is not necessary to transmit along at that point in time. A further reason for preferentially transmitting along a particular line can be because that line is longer or has higher losses than other lines.

The above technique can be applied to both the downstream (base station to subscriber) and upstream (subscriber to base station) directions of communication. When used in the upstream direction, it has the effect that the base station preferentially receives from a particular one of the power lines.

A busbar array has a plurality of busbars, each serving a particular electrical phase. In order to communicate with subscribers on different phases, it is preferred to connect to the busbars of each phase line.

One of the potential problems with transporting RF signals over a power line is that of unwanted radiation of RF energy from the power line and other parts of the electricity distribution network. This is because the electricity distribution network was not designed to carry RF signals. One significant point where radiation can occur is at the substations where electricity is transformed from high voltage (eg 6.6 or 11kV) to low voltage (400V). The busbars at the substation are typically mounted as a grid array on the substation wall and while they are shielded from view they are often electrically unscreened. This is because screening is considered unnecessary at the 50Hz mains frequency. At RF frequencies the busbar array can function as an antenna, radiating the RF signals which it receives via the distribution cables into the surrounding area. This is undesirable as it can cause interference with equipment operating at these frequencies.

A further improvement is to couple the set of communications signals to the busbars with a phasor relationship such that radiation of communications signals from the power line is minimised. This allows communications signals to be coupled onto the power line at a higher power level while still maintaining the radiated power within a particular limit. Radiated power is a particular problem when the communications signals occupy frequency bands that are not specially allocated for power line communications and therefore where there could be an interference problem with other users. It has been found that an increase of some 10 or 20dB in transmitted power level is possible with this technique compared with coupling a set of in-phase communications signals on to the conductors. This allows a base station to serve subscribers who are located further away from the base station, which maximises the coverage of the base station and minimises the number of additional base stations or repeaters which are needed. Preferably the phasor relationship of the communications signals is arranged such that the vectorial sum of the set of communications signals is substantially equal to zero. The set of communications signals may comprise two or more signals.

The set of communications signals can be derived in several ways. A first way comprises receiving an input communications signal, dividing the input signal into the set of communications signals and choosing the phasor relationship of the set of communications signals by phase-shifting at least one of the divided signals. The phase-shifting can be achieved by a suitable weighting device.

An alternative way comprises directly generating a set of communications signals, such as by direct digital synthesis, one for each conductor of the set of conductors, which generated signals have a phasor relationship such that radiation of communications signals from the power line is minimised. The generated signals are modulated according to an input communications signal which it is intended to transmit.

One or more of the set of communications signals may be scaled in amplitude. This has the advantage of coping with differential loading effects on the power line conductors.

Advantageously the method further comprises monitoring a quantity which is indicative of the success of the step of setting the phasor relationship of the communications signals such that radiation of the communications signals is minimised and varying the phasor relationship of the set of communications signals according to the monitored quantity. The monitored quantity can be radiation of communications signals from the power line or a quantity which is indicative of the loading of the conductors of the power line.

The phasor relationship of the set of communications signals can be varied by applying perturbations to phasor settings for one or more of the communications signals.

The network can comprise underground or overhead lines or a combination of these. Radiation of RF signals can be a particular problem with some types of overhead lines which do not have a co-axial like structure or benefit from the shielding of being buried underground.

The communications signals can be data packets carrying data for computer applications such as home working, file transfer, internet access and a wide range of other applications.

A further aspect of the present invention provides a power distribution network comprising a substation having a plurality of power lines leading from it for distributing electricity to a plurality of premises and a plurality of communications stations coupled to the power lines, and means for coupling the communications signals to the network at busbars of the substation for transmission over the plurality of lines to the communications stations. Preferred features may be combined as appropriate, and may be combined with any of the aspects of the invention as would be apparent to a person skilled in the art.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the invention, and to show by way of example how it may be carried into effect, embodiments will now be described with reference to the accompanying drawings, in which:-

Figure 1 shows an electricity distribution network which is adapted to carry telecommunications signals;

Figure 2 shows one way of coupling to the busbar array of figure 1 in more detail;

Figures 3 and 4 show alternative ways of coupling to the busbar array of figure 1 ;

Figure 5 shows distribution of signal power to power lines which is possible using the arrangement of figures 3 and 4;

Figure 6 shows the arrangement of figure 3 with the addition of weighting control for the two paths;

Figure 7 shows a way of controlling the arrangement of figure 6;

Figure 8 shows the arrangement of figure 4 with the addition of control of the signals fed to each phase line;

Figure 9 shows the phasor relationship for a set of three communications signals which can be coupled using the arrangement of figure 8;

Figure 10 shows a first arrangement for applying a set of communications signals to a power line with a phasor relationship;

Figure 11 shows a second arrangement for applying a set of communications signals to a power line with a phasor relationship; Figure 12 shows arrangements for monitoring effectiveness of the arrangements of figures 10 and 11 ;

Figure 13 shows an example of a control unit for use with the arrangements of figures 10 and 11.

DESCRIPTION OF PREFERRED EMBODIMENTS

Figure 1 shows an electricity distribution network which is adapted to carry telecommunications signals. Mains electricity enters the network from a high voltage 6.6kV or 11kV transmission line 15 and is transformed by transformer 20 at substation 10 into a 400V supply which is delivered via busbars 25 and distribution cable 40 to customer premises S1 and S2. A substation 10 typically has between 4 and 8 such distribution cables of the kind shown as 40, 41 leading from it, each distribution cable serving a number of subscriber premises. A distribution cable can extend for several hundreds of metres. Distribution cable 40 comprises a set of phase lines blue (B), red (R) and yellow (Y) and a neutral line (N) which are electrically insulated from one another. A full distribution network will usually include more of the premises shown as S1 , S2 and will typically include a more elaborate tree-and-branch distribution network. Subscriber premises may receive a single phase electricity supply (230V) or a three-phase electricity supply (400V). Domestic subscriber premises usually receive a single-phase supply and neighbouring subscriber premises are usually coupled to different phase lines. In figure 1 subscriber S1 is shown coupled to the blue phase line and subscriber S2 is coupled to the red phase line. This helps to distribute the load of the network evenly across the three phases.

A base station BS couples communications signals, such as voice or data, onto the electricity distribution network. As shown in more detail in figures 2-7, the communications signals from base station BS are coupled to the busbars 25 at substation 10, the busbars acting as a star point for serving all of the distribution cables 40, 41. Communications signals propagate over the cables to transceiver stations TRX located at subscriber premises S1 , S2. Subscriber premises coupled to one or more phase lines of distribution cable

40 by a branch line 26. A coupling unit 27 at the subscriber premises S1 couples communications signals to and from the branch cable 26 to serve transceiver unit TRX. In the upstream direction, communications signals are transmitted from the subscriber transceiver station TRX towards the base station BS.

Communications transceiver stations TRX are located at various distances from the base station BS along lines 40, 41. The quality of communications experienced by these transceiver units depends on the signal to noise ratio (SNR) at the transceiver stations. Factors affecting the signal to noise ratio are: (i) attenuation between the base station and transceiver unit;

(ii) noise level on the power line; and, (iii) the power level at which the base station transmits.

Communications signals can be transmitted between a phase line and neutral or earth. Communications signals can be transmitted using a variety of line coding or modulation techniques, such as Frequency Shift Keying (FSK) modulation. Radio frequency (RF) carriers carry the communications information. Frequency bands in the range 1 to 30 MHz can be used, as these offer greater bandwidth for carrying communications signals, but it is also possible to use other frequencies. The bit rate of such signals can be as high as 1 -2Mbit/s.

A first way of coupling communications signals to the busbars is shown in figure 2. Communications signals from a base station pass through a coupling unit 46 which acts as a high-pass filter to allow communications signals to pass, and to block mains electricity. An output of each coupling unit 46 connects to a point on the busbar array 25. In figure 2 the communications signals are connected to the array near to a central position on each of the longitudinal conductors (R, Y, B) forming the array. The connection can be achieved by clamping an electrical connector directly to the busbar conductor, or, as shown in figure 2, by replacing a fuse in the distribution line 42, the fuse being adjacent to the busbar conductor. The replacement fuse unit 50 includes means for coupling communications signals to/from the power line 42. A suitable fuse is described in more detail in International Patent Application PCT/GB 98/00173. Communications signals injected at fuse 50 propagate along line 42 and also propagate in both directions along line R to other distribution lines 40, 41 , 43, 44, 45. Communications signals are injected onto each of the phase lines Y, B in a similar manner to that just described for line R. Signals can be injected at other positions along the busbar array, such as at one end.

In figure 3 each busbar conductor (R, Y, B) is fed at two points.

Communications signals from base station BS are split into two portions by a 3dB splitter 60 and each portion is applied to a respective end of a conductor. The two portions can be injected at other positions along the array, but it has been found that injecting at the ends provides best performance. Also, cables can be routed to avoid the distribution frame for increased safety. The signals can be injected in series with the ends of the busbar conductors, as shown in figure 3, or, as shown in figure 4, by using replacement fuse units 57, 58 in lines 40, 45.

Figure 5 is a graph showing the effect of coupling signals to the busbars using the arrangements of figures 3 and 4. The graph shows power against position along the longitudinal busbar array. The power contributed by the left- hand injection point falls towards the right-hand side of the array whereas the power contributed by the right-hand injection point falls towards the left-hand side of the array. The variation in power need not be linear, as shown here.

The total power contributed by both injection points is fairly constant across the full length of the array.

Figure 6 shows a further improvement. Communications signals to/from base station BS are split, as before, into two paths for feeding two points on the array. One of the paths is modified with respect to the other path by a set of complex weights W^ W₂, W₃, one weight per path. Each weight can modify the amplitude and/or phase of the signal in that path. Weighting the signals in one of the paths has the effect of modifying the distribution of combined signal power across the busbar array. In this way it is possible to arrange the power along a particular one of the lines 40, 41 , 42, 43, 44, 45 to be maximised.

The weight values are controlled by weight control signals 75. While figure 6 shows weights in only one of the two paths, it is possible to use a further set of weights in the other path, such as at position 76. It will be appreciated that an alternative to the arrangement shown in figure 6 is to provide a single communications signal from base station BS, to split this into two paths by a splitter 60, to apply weighting to one or both of the split paths and then to divide this weighted signal into three parts, one for each of the phase lines, thereby reducing the number of weights.

Figure 7 shows one way of controlling the values of weights W_{1 f} W₂, W₃. A weight controller 80 has an associated memory 81 which stores weight values. Values may be associated with a particular subscriber, distribution line (40, 41 ,..) or some other parameter. The weight values may be calculated by a skilled operator at a time of setting up the system, or during a subsequent inspection, or they may be derived automatically and may change according to measurements made by the system itself in response to changing conditions on the distribution network.

A particularly advantageous way of using the weighting is to target injected power on the busbar towards a particular one of the power lines. Sequence controller 82 is part of the base station and determines the sequence in which communications stations (TRX at S1 , S2..) are polled. The sequence controller controls operation of the transmitter TX and receiver RX parts of the base station and also issues a control signal, with this timing information, to weight controller 80. The weight control signals 75 can thus be changed in synchronism with the subscriber that is being communicated with in order to target injected power at busbar array 25 along a particular line, e.g. to maximise power along line 40 to which subscriber S1 is located at the time that the base station is communicating with S1.

For packet data or time division duplex (TDD) communications where the same frequency band is used for both directions of communication conditions should be reciprocal such that the same weight set can be used for both directions of communication. This allows an optimum weight value to be calculated based on an upstream (base station receiving) communication flow and to use that same weight value on the downstream flow. The weight value can be calculated using a received signal strength indication (RSSI) from the receiver RX, as shown in figure 7, or from some other indication of quality of the communications path such as bit error rate. An optimum weight value can be derived by applying changes or perturbations to the weight value and monitoring the effect that this has on the monitored parameter. Weights typically have in-phase (I) and quadrature (Q) components which can be individually varied in turn to derive the optimum weight value. The weight value can be derived during a continuous transmission from a subscriber station, such as a test transmission initiated by the base station.

It is possible to use the same weighting value for every subscriber on a particular power line (40, 41..) or an individual weighting value can be derived for each subscriber on the line. By deriving a weight value for each subscriber it is possible to account for the distance of the subscriber along the line, and to transmit at a higher or lower power level by varying the amplitude of the weight values in one or both paths.

One problem with injecting RF signals at busbars is unwanted radiation of RF energy from the busbars and other parts of the electricity distribution network. In some trials of equipment that applies communications signals to power lines in an in-phase relationship, transmit power has been limited to 50mW, this figure corresponding to government regulations for spurious, i.e. out-of-band radiated RF power from radio transmitters.

Figure 8 shows a further modification where the set of communications signals that are applied to the busbars (one signal per busbar phase line) have a phasor relationship that minimises radiation. This can allow a 10 to 20dB increase in transmit power while still maintaining radiated power from a line within regulated limits. A further set of weights 70 are inserted in the lines that feed communications signals from the base station. These weights could alternatively be positioned at points 71 , 72.

The following description relates to the operation of this radiation-reducing technique.

As the power level at which base station BS transmits is increased, the amount of radiation emitted by the distribution network also increases. By applying the communications signals to the phase lines of the distribution network such that they vectorially sum to zero, it is possible to minimise the amount of radiation from the network, and therefore transmit at a higher power level. This allows a base station to serve subscribers located at a greater distance from the station than would have been possible without this arrangement while still maintaining radiated power within acceptable limits. Figure 9 shows a phasor diagram which illustrates the phasor relationship between three communication signals, one for each of the phase lines B, R, Y. The set of communications signals are offset from one another by 120°. This phasor relationship should apply to all frequency components which form each of the communications signals. As an example, where each communications signal comprises two sine wave components at different frequencies, the phasor relationship between the set of such communications signals should maintain a 120° phase-shift between both frequency components of the signals. Applying the communications signals to the network with this relationship means that radiation from the lines, when viewed in the far-field is minimised. This can be applied more generally to situations where n communications signals are applied to n phase lines, the communications signals being offset from one another by 360%. All or only some of the phase lines in the power line may carry communications signals.

Figure 10 shows one arrangement which can be used at base station

(BS) for applying communications signals to the power line with a phasor relationship. Transmitter TX outputs a modulated communications signal upon a RF carrier. This is divided by splitter 200 into three branches, one for each of the three phase lines.

Each branch includes a weighting device W_B, W_R, W_γ which can phase- shift a signal at RF. The devices should be capable of applying a constant phase-shift across all frequency components within the bandwidth occupied by the communications signal. Operation of the weighting devices is controlled by a control unit 220. This controls the magnitude of the phase shift which each device provides. The weighting devices may also be arranged to scale the signals in amplitude. The result of this process is that the power line carries a set of signals having the same information content but which are shifted in phase and scaled in amplitude with respect to one another to minimise radiation from the line.

Figure 11 shows an alternative way of applying communications signals to the power line with a phasor relationship. Input 300 receives a communications signal at base band. This is fed to signal generating devices 310, 320, 330. The signal generating devices generate an RF signal having a particular phase and amplitude. The generated signal is preferably generated in a modulated form according to incoming data from input 300. For example, where modulation is by frequency shift keying (FSK), a carrier at a particular frequency representative of a particular data level '00', '01' etc. is generated.

This technique is often called direct digital synthesis (DDS) as the signal is typically generated by retrieving a stored digital representation of the signal from an electronic memory and converting it to analogue form. As in figure 4, the generated signals are passed through a high-pass filter unit 230 before being applied to phase lines B, R, Y. Operation of the generating devices 310,

320, 330 is controlled by control unit 220 which controls the phase and amplitude of the generated signals. The result of this process is that the power line carries a set of signals having the same information content but which are shifted in phase and scaled in amplitude with respect to one another to minimise radiation from the line.

The phasor relationship of the communications signals which is required to minimise radiation from the line should remain fairly static. For example, where three communications signals are applied to three lines, the three signals should be offset from one another by 120°. However, varying conditions on the electricity network can require a different phasor relationship for the communications signals to ensure that radiation is minimised. Differential electrical loading of the phase lines is one such condition. Figure 12 shows two ways of monitoring the network to ensure that an optimum phase relationship is established for the communications signals.

The upper part of figure 12 shows an antenna 450 located in the far-field of busbars 25. This senses radiation from the busbars and feeds a signal 460 to controller 220 for adjusting values of the weighting or signal generating devices at base station BS. Antenna 450 should be located near the busbars, or some other point of the network which requires monitoring, so as to sense the combined effect of the radiations from the lines rather than the individual contributions of a particular line.

The lower part of figure 12 shows an alternative arrangement for determining the effectiveness of the phasor relationship. Sensing coils 400, 401 , 402 are arranged one per phase line to detect signals present on respective busbars B, R, Y. The coils can conveniently be inductive coils which are wound around the busbar or laid closely adjacent to it. Sensing is also possible by a capacitive link or by one or more antennas located close to the busbars. Outputs of the sensing coils are either fed directly to control unit 220 or are conveyed, one at a time, by switch 410 to control unit 220 in a multiplex fashion. Sensing coils 400, 401 , 402 or equivalent devices provide an indication of loading of each of the phase lines. As described above, differential loading of the phase lines requires adjustment to the phasor relationship of the communications signals applied to those phase lines.

Figure 13 shows an example of the control unit 220. A monitoring signal 420, 460 is applied to a channel filter 500 which passes only those frequencies which are of interest. A power detector 510, shown simply as a diode detector D and capacitor C provide a power measurement which is applied to an analogue-to-digital converter 520. The output of the A-to-D converter is fed to a microprocessor 530 which performs an adjustment of the control signals which are fed to the weighting devices or signal generating devices which establish the phasor relationship of the communications signals. Figure 13 shows microprocessor 520 outputting two sets of weight control signals for weights W_B, W_R. A weight can conveniently be controlled by an in-phase (I) and quadrature (Q) control signal. Microprocessor 530 can perform a perturbation process which works by applying steps in the size of the I and Q weight values and monitoring the success of the changes. The algorithm can work by successively changing I up, I down, Q up, Q down. After these four steps have been performed, the change in I or Q which had the best effect is adopted. The process continues until the best cancellation effect is achieved.

It is also possible to vary the control signals by performing a correlation technique.

Claims

1. A method of transmitting communications signals over a power distribution network comprising a substation having a plurality of power lines leading from it for distributing electricity to a plurality of premises and a plurality of communications stations coupled to the power lines, the method comprising: - coupling the communications signals to the network at busbars of the substation for transmission over the plurality of lines to the communications stations.

2. A method according to claim 1 comprising coupling communications signals to one or more of the power lines adjacent where the power line adjoins the busbar.

3. A method according to claim 2 comprising applying the communications signals via a fuse unit fitted in series with the power line, the fuse unit comprising means for coupling the communications signals to and/or from the power line.

4. A method according to claim 1 comprising applying communications signals in series with an end of the busbar

5. A method according to any one of claims 1 to 3 wherein the busbar comprises a longitudinal conductor having power lines leading from it at intervals along the conductor, and the method comprises applying communications signals to a centre of the busbar.

6. A method according to any one of claims 1 to 3 wherein the busbar comprises a longitudinal conductor having power lines leading from it at intervals along the conductor, and the method comprises applying communications signals at two spaced apart points on the conductor.

7. A method according to claim 6 wherein the two points are the two ends of the conductor.

8. A method according to claim 6 or 7 further comprising modifying the signals that are applied to one of the points on the conductor with respect to the signals that are applied to the other point whereby to control the combined power that is delivered to the lines.

9. A method according to claim 8 wherein the modifying comprises a phase- shift.

10. A method according to claim 8 or 9 wherein the modifying comprises a scale in amplitude.

11. A method according to any one of claims 8 to 10 wherein the modifying varies in time.

12. A method according to claim 11 wherein the modifying varies according to a communications station that it is desired to communicate with, the modifying being controlled so as to preferentially transmit along a power line to which that station is located.

13. A method according to any preceding claim wherein there are a plurality of busbars, the method comprising: - coupling a set of communications signals to a set of the plurality of busbars, each signal being coupled to a respective one of the set of busbars; and,

- setting the phasor relationship of the set of communications signals such that radiation of communications signals from the busbars is minimised.

14. A method according to claim 13 wherein the phasor relationship of the communications signals is arranged such that the vectorial sum of the set of communications signals is substantially equal to zero.

15. A power distribution network comprising a substation having a plurality of power lines leading from it for distributing electricity to a plurality of premises and a plurality of communications stations coupled to the power lines, and means for coupling the communications signals to the network at busbars of the substation for transmission over the plurality of lines to the communications stations.