GB2383724A

GB2383724A - Space time coded data transmission via inductive effect between adjacent power lines

Info

Publication number: GB2383724A
Application number: GB0130043A
Authority: GB
Inventors: Bahram Honary; Paul Anthony Brown; Javad Yazdani
Original assignee: Lancaster University
Current assignee: Lancaster University
Priority date: 2001-12-15
Filing date: 2001-12-15
Publication date: 2003-07-02
Anticipated expiration: 2021-12-15
Also published as: GB0130043D0; GB2383724B

Abstract

Space time coded data is transmitted via electricity mains by sending data encoded with some redundancy ("time diversity") from one modem down one conducting line which interacts inductively with adjacent conducting lines to provide further replication of itself ("space diversity"). The effects of induction on parallel conductors will cause all receiving modems to receive some of the signal, necessitating an identifier (<I>eg</I>. header) of the intended recipient to form part of the signal. A handshake procedure employing test signals is used to establish the nonlinearity of the frequency response and time domain response of the link, and appropriate coding and selected combinations of frequency channels and time slots involving different conducting lines are chosen to optimise communication (fig 3), thus providing diversity in an analogous manner to mobile telephone communication via multiple antennas. A filter is used to minimise "mains hum", <I>ie</I>. signals at electricity supply frequencies.

Description

COMMUNICATIONS SYSTEM The present invention relates to a communications system.

There is a continuous and increasing demand to deliver digital communications to homes or places of business. In many instances the existing communications infrastructure, and in particular communications connections to homes or business premises, is not capable of supplying sufficient bandwidth to satisfy this demand. There is a demand for high bandwidth connections to the Internet which currently cannot be satisfied without substantial and costly upgrading of communications connections to homes and business premises (for example by laying fibre optic cabling).

The vast majority of homes and business premises are connected to electricity mains. It has been speculated previously that it is possible to superimpose or overlay communications signals onto electricity mains in order to receive and transmit digital information from a home or business premises. This superimposing of signals is known as power line communications (PLC), and could potentially be a solution to the problem of providing high bandwidth communications connections to homes or businesses premises. However, the tree and branch nature of electricity distribution networks has given rise to substantial difficulties in the transmission of communications signals across the network.

It is an object of the present invention to provide a communications system which overcomes or substantially mitigates the above problem.

According to a first aspect of the invention there is provided a communications system comprising an electricity distribution and/or transmission network formed from at least two substantially adjacent conductors connected to a plurality of electricity delivery points, the system further comprising first and second modems located at or adjacent electricity delivery points of the electricity distribution network and configured to transmit data at communications frequencies and to filter

out signals at mains electricity supply frequencies, wherein the first modem is arranged to transmit coded data onto the at least two substantially adjacent conductors so that the inductive interaction of the conductors provides space time coding of the data.

The term'electricity transmission and/or distribution network'is intended to include any electricity network and is not limited to any particular range of voltages.

Preferably, the electricity distribution and/or transmission network is formed from three substantially adjacent conductors, and the first modem is arranged to transmit coded data onto two or three substantially adjacent conductors.

Preferably, the electricity distribution and/or transmission network is formed from four substantially adjacent conductors, and the first modem is arranged to transmit coded data onto two, three or four substantially adjacent conductors.

Preferably, the data is coded using a channel coding apparatus.

Preferably, the channel coding apparatus comprises trellis coding means or feedback coding means, or feed-forward coding means with training information, or blind coding means.

Preferably, the coded data is passed via a serial to parallel converter before transmission.

Preferably, following the serial to parallel converter, the coded data passes via a pulse shaper and a modulator before transmission.

Preferably, the second modem is arranged to use a decision metric to retrieve the coded data from received signals.

Preferably, the second modem is connected to two substantially adjacent conductors of the electricity distribution and/or transmission network.

Preferably, the second modem is connected to three or more substantially adjacent conductors of the electricity distribution and/or transmission network.

Preferably, the first modem is arranged to transmit test signals over a range of frequencies and time intervals, the second modem is arranged to receive the test signals and upon receiving the test signals to transmit data describing properties of the received signals, and the first modem is arranged to receive the transmitted data and use the data to select appropriate combinations of frequencies and time intervals for the subsequent transmission of coded data.

Preferably, the first modem is arranged to take into account the quantity of data to be transmitted when selecting the combinations of frequencies and time intervals.

Preferably, the first modem is arranged to transmit test signals on different conductors of the network, the second modem is arranged to receive the test signals and upon receiving the test signals to transmit data describing properties of the received signals, and the first modem is arranged to receive the transmitted data and use the data to select appropriate conductors for the subsequent transmission of coded data.

Preferably, the first modem is arranged to transmit test signals on more than one conductor simultaneously, and is arranged to use data received from the second modem to select appropriate combinations of conductors for the subsequent transmission of coded data.

Preferably, the first modem is arranged to dynamically switch between conductors when transmitting test signals, and is arranged to use data received from

the second modem to select appropriate dynamic switching rates for the subsequent transmission of communications data.

Preferably, the communications data includes an identifying characteristic which identifies that modem which is the intended recipient of the communications data.

Preferably, the electricity distribution and/or transmission network connects buildings to an electricity sub-station.

Preferably, the network is a three-phase network comprising three live conductors and a neutral conductor.

Preferably, the electricity distribution and/or transmission network connects electricity sub-stations.

Preferably, the electricity distribution and/or transmission network is located within a building, and provides electricity to power sockets.

Preferably, the network is monitored regularly for changes of transmission properties of the network at communications frequencies.

According to a second aspect of the invention there is provided a method of communicating via an electricity distribution and/or transmission network formed from at least two substantially adjacent conductors connected to a plurality of electricity delivery points, first and second modems being located at or adjacent electricity delivery points of the electricity distribution network and configured to transmit data at communications frequencies and to filter out signals at mains electricity supply frequencies, wherein the method comprises transmitting coded data onto the at least two substantially adjacent conductors so that the inductive interaction of the conductors provides space time coding of the data.

The second aspect of the invention may incorporate any of the preferred features of the first aspect of the invention.

A specific embodiment of the invention will now be described by way of example only with reference to the accompanying figures, in which: Figure 1 is a schematic illustration of an electricity distribution network; Figure 2 is a second schematic illustration of an electricity distribution network; Figure 3 is a table illustrating the selection of time and frequency combinations for data transmission, as used by an embodiment of the invention ; Figure 4 is a schematic illustration of a series of transmitters and receivers; and Figure 5 is schematic illustration of a channel encoder and a possible waveform shaper.

An electricity mains generally comprises a distribution network, typically a polyphase cable arrangement interconnected in a tree and branch format. The physical dimensions of the distribution network are generally much less than the free space wavelength at which electricity is distributed (for 50Hz or 60Hz mains electricity the free space wavelength is 6000 or 5000 kilometres respectively). This means that steady state propagation characteristics of mains electricity are dominated by the resistance per unit length of the polyphase cables, and as a result the frequency response of the distribution network tends to be linear.

At communications frequencies the dimensions of the electricity distribution network are comparable with the wavelength of the communications carrier signal. For example, at a frequency of 1 MHz the free space wavelength is around 300m, and a quarter wavelength will be around 75m. Since a mains cable may be up to around 250m long, and will include branches which are generally tens of metres long, the wavelength of the communications carrier signal components will be comparable to the length of parts of the electricity distribution network. This means that the propagation characteristics of the carrier signal will be significantly affected by the

electricity distribution network, giving rise to non-linear frequency responses for these networks. For example, a given wavelength of communications carrier signal may be reflected from a branch of the network 180 degrees out of phase, and thus give rise to destructive interference with the communications carrier signal. The nonlinear response of the electricity distribution network introduces a problem when it is desired to convey broadband data signals, since these of necessity include high frequency carrier signal components.

The frequency response between two terminals in a tree and branch electricity distribution network is likely to be different to the response between two different terminals. Referring to figure 1, a signal is to transmitted from terminal A to terminal D. The signal will be modified in the frequency domain by the tree and branch network during propagation from terminal A to terminal D. A signal transmitted from terminal D to terminal A will undergo the same frequency mapping as the signal transmitted from terminal A to terminal D.

In general the frequency mapping of the signal between any two terminals in the tree and branch network will be unique. The number N of bi-directional point-topoint frequency response mappings for a tree and branch electricity distribution network with'n'branches may be summarised mathematically as follows:

where m= (n-l) Applying equation 1 to figure 1 (i. e. n=5, m=4) gives N = 4+3+2+ 1, i. e. there are ten unique frequency domain mappings.

From equation 1, a single phase electricity distribution network with 50 branches will have 1225 individual frequency response mappings.

In many countries electricity is distributed using a three-phase cable (i. e. three conductors carrying electricity at different phases, and a neutral conductor bonded to earth). Where the electricity distribution network comprises three-phase cables, the total number of frequency response mappings increases three fold. Typically, a three phase power cable comprises three conductors each sheathed in coloured PVC (red, yellow, blue) together with a neutral conductor sheathed in black PVC. The neutral conductor is connected to earth at a supply substation. Since the conductors run adjacent to one another a substantial inductive coupling arises between the conductors. Consequently, a communications signal injected for example onto the red conductor will very quickly spread onto the yellow or blue conductors. In addition to the inductance, there is a capacitance between the conductors which will introduce non-linearities into communications signals. Discontinuities in the supply network together with terminations of the conductors at houses combine with the capacitance and the inductance to give rise to a complicated frequency response mapping for the electricity distribution network.

In addition to the nonlinear frequency response, the electricity distribution network will also have a nonlinear time domain response. This may be characterised in a manner analogous to that described in relation to frequency response mapping.

The inventors have realised that the non-linearity introduced by the frequency response and time domain response of the electricity distribution network may be determined, and that combinations of transmission frequencies and time slots may then be selected in order to provide efficient transmission of data between different points on the network. In addition, the inventors have realised that an appropriate coding may be used to code the data.

Referring to figure 2, a mains cable which runs under a street comprises a

three phase cable arrangement, i. e. three power carrying conductors (red, yellow, ZD blue), and a neutral (black) conductor. The black neutral conductor is connected to earth at an electricity substation (not shown). Each power carrying conductor carries an alternating current (AC) with a potential difference between the conductor and

neutral typically of 230V rms., the currents carried by each conductor being mutually separated by a phase of 120 degrees. A first house 1 is connected to the red conductor and the black neutral conductor of the three phase cable, a second house 2 is connected to the yellow conductor and the black neutral conductor, and a third house 3 is connected to the blue conductor and the black neutral conductor. The fourth house 4 is connected to all three phases of the power supply network; this may be due to high power consumption requirements in the fourth house.

The nonlinear frequency response and time domain response are determined for a range of frequencies and times, and this information is used to provide efficient communication.

Referring to figure 2, in one embodiment of the invention each house 1-4 is provided with a modem which is connected to the electricity mains at the house, and is arranged to receive communications frequency signals from the mains. The electricity substation (not shown) is also provided with a modem. The modems are arranged to transmit test signals to each other to determine those parts of the frequency spectrum, and time slots, which may be used to provide efficient communication. For example, the modem at house 2 may transmit a handshake signal at a specific carrier frequency onto the yellow and neutral conductors of the threephase electricity supply network. Through the inductance of the conductors, the signal will pass onto the red conductor and the blue conductor. The signal will be received by the modem at the substation, and by the modems at each of the houses 1, 3,4. Each modem will transmit a return handshake signal indicating the signal to noise ratio of the received signal. The modem at house 2 receives these signals and records the signal to noise ratios indicated by each of the modems. The modem at house 2 then transmits a handshake signal at a different carrier frequency, and again records the signal to noise ratio of the signals received by the modems at the substation and the other houses 1,3, 4. This is repeated over a wide range of carrier frequencies, for example in steps of 50kHz. A similar exercise is carried out by transmitting a signal in different time slots and determining which signals are received with the least distortion. Carrier signals are also transmitted for combinations of

frequencies and time slots, to determine which combinations provide the best ratio of signal to noise.

Once all of the test signals have been transmitted by the modem at house 2, and all the responses received from the other modems one the electricity distribution network, the modem at house 2 will store the set of frequency ranges, time slots, and combinations of frequency and time slot, which provide the best ratio of signal to noise for transmission to a given modem (for example the modem at house 1). An example of data relating to combinations of frequency and time slot is shown in figure 3. Referring to figure 3 it can be seen that certain combinations of time and frequency have been marked, indicating that these provide good signal to noise ratios. This data is stored an used for future data transmissions. Different combinations of frequency bands and time slots will be determined for communication between each modem.

In order to transmit data to house 1, the modem will select those combinations of frequency and time slot which are marked in figure 3. If only a small amount of data is to be transmitted, then only a few of the combinations may be required, for

example fla, tel ; fg, t2. If a larger amount of data is to be transmitted, then all the available combinations may be used, i. e. fio, tl ; f8, t2 ; f9, f4, 4 ; f7, t6 ; f4, t7 ; f5, tg ; f2, t9 ; fl, tlO. Using all of the available combinations will provide communication with the maximum useful available bandwidth. Generally, the available combinations of time and frequency are used to transmit the data and at least one replica of the data. This is done to ensure that data is not lost during transmission. The transmission of more than one representation of the data using combinations of frequencies and time slots is referred to as temporal and frequency diversity.

The data to be transmitted may be coded, the coding used being selected on the basis of properties of the connection between the modems at houses 1 and 2. For example the coding scheme could be one of the following: trellis coding, a feedback scheme, a scheme using feed-forward or training information but no feedback, or a blind scheme.

A signal transmitted from one modem will in general, due to the inductive interaction between the conductors of the electricity distribution network, be received at each of the modems connected to the network. This means that a signal intended for one modem only must include some identifying characteristic, for example a header identifying the intended recipient, or a specific frequency range or set of frequency ranges which is allocated to the intended recipient.

The modem at house 4 may wish to communicate with the modem at house 1.

Since there is substantial inductive interaction of the adjacent conductors of the electricity distribution network, it is not necessary that the modem at house 4 transmit data onto the conductor that is connected to the modem at house 1 (i. e. the red conductor).

In addition to determining the best combinations of frequency and time slot, the modems are arranged to determine the signal to noise ratio achieved when transmitting a signal on combinations of conductors, and the effect of dynamically switching between conductors. For example, the modem at house 4 will transmit a test handshake signal on combinations of conductors, and dynamically switching between conductors at a range of different switching rates. The modems at the houses 1-3 and the modem at the substation will transmit a return signal indicating the signal to noise ratio of the received signal. Combinations of conductors, and dynamic switching between conductors which provide good signal to noise ratios are recorded, together with favourable combinations of frequencies and time slots. This is in effect a table corresponding to that shown in figure 3, but with extra information.

The inventors have realised that using combinations of conductors to transmit data is analogous to the use of multiple transmission antennas by mobile telephone communications systems. Transmission of a coded signal from more than one antenna is commonly used in mobile telephone systems to avoid loss of data due to channel fading. The same objective is achieved here by transmitting the coded signal on more than one conductor. The space diversity of the electricity distribution network comes from the inductive interaction of the adjacent cables of the network.

The electricity distribution network provides the advantage that, unlike mobile telephone communications, the spatial properties of the network are static so that the signal to noise ratio will not vary substantially, thus providing improved performance.

Space-time coding of data has recently been introduced in the field of wireless communications systems, as described in references [1, 2 and 3]. Space time coding comprises transmitting a coded signal from a plurality of transmitters, and provides a means of enhancing the signal to noise ratio achieved. The signal may be received by a plurality of receivers (this is known as'many in many out') or may be received by a single receiver (this is known as'many in single out').

For the case of a wireless communications system composed of N antennas at a transmitting site and M antennas at a receiver site, data to be transmitted is encoded using a channel code. The encoded data goes through a serial-to-parallel converter and is divided into N streams of data. Each stream of data is used as the input to a pulse shaper, and the output of each shaper is then modulated. At each time slot t the output i of the modulator is a signal c't that is transmitted using the transmit antenna i for 1 < ! N. The N signals are transmitted simultaneously each from a different transmitting antenna and all these signals have the same transmission period T. The signal at each receiving antenna is a noisy superposition of the N transmitted signals corrupted by Rayleigh or Rician fading. Figure 4 shows a possible transmitter arrangement [2], and figure 5 shows a possible channel encoder and possible waveform shaper.

At the receiver, a demodulator computes a decision statistic based on the signals received at each receive antenna I 'M. The signal it received by antenna j at time t is given by

Where the coefficient a is the path gain from transmitting antenna i to receiving antenna and 7/, is the additive white Gaussian noise (AWGN) for the channel between all the transmitting antennas and the receiving antenna j at time t.

The path gains a,, are generally modelled as samples of independent complex Gaussian random variables with mean zero and variance 0.5 per dimension. This is equivalent to the assumption that signals transmitted from different antennas undergo

independent Rayleigh fades. The noise quantities 7//, j = 1,..., M ; t = 1, 2,....,/are p samples of independent complex Gaussian random variables with mean zero (typical for white Gaussian noise) and variance Nol2 per dimension where No is the noise power. It is further assumed that ab, are constant during a time frame and vary from one frame to another (quasi-static fading) T < fading period. The other assumption is that the aij are independent from one time slot to another.

Thus the effect of the channel on a multiple input/multiple output (MIMO) communication system with N transmitters and M receivers, can be represented by the NxM channel matrix H:

Space time coding may be applied to the transmission of communications signals via a three phase power line transmission. Where this is done, N = M = 3.

Putting equation (2) in vector form :

Equation (3) represents the received signal vector rt, where et = [c1(l),c2(l),c3(l)]# (where []T is the matrix transposed of the transmitted code vector). The channel H will be a 3x3 matrix in this case.

Assuming that in each time slot (1, 2,..., l) the transmitter outputs 3 code symbols and that all the code symbols are equi-probable, the receiver decides in favour of the transmitted code symbol

if it minimises the decision metric

Each code symbol in a block is decoded separately by minimising the metric indicated in equation (5). The decoder then outputs the hard-decisions on the data.

The coding gain for a three transmitting antenna system is in the order of 25 dB [4].

The electricity distribution network may carry space time coded data to a modem which is connected to a single phase of the network. This is a'many in single out'configuration.

The electricity distribution network is similar to wireless communication systems in the manner in which it suffers from attenuation, multi-path propagation, reflection and a range of interference (noise). Thus, methods of space-time coding which have been developed for wireless communication systems [1, 2 and 3], may be applied to the electricity distribution network.

The channel code used to code the data may be of any suitable type. For example, trellis coding, feedback schemes, schemes using feed-forward or training information but no feedback, or blind schemes may be used.

The described embodiment of the invention relates to inter-building communication via an electricity distribution network, and to communication between a building and an electricity substation. It will be appreciated that the inter-building electricity distribution network shares many features with the network used to distribute electricity within a building. Thus, the invention may be applied to intrabuilding communication, for example communication between different floors of an office building using the mains electricity network within the building. Where the invention is applied to intra-building communication, signals may be transmitted onto the Live and Earth conductors, the Live and Neutral conductors and/or the Neutral and Earth conductors. A notable feature of the application of the invention to intrabuilding communication is that the properties of the mains electricity network within the building will be affected by the connection of appliances to the mains. For this reason the frequency and time domain mappings between communication points on the mains must be regularly monitored.

REFERENCES [1] V. Tarokh, N. Seshadri and A. R. Calderbank,"Space-Time Codes for High Data Rates Wireless Communications: Performance Criterion and Code Construction", IEEE Transactions on Information Theory, Vol 44, Number 2, pp. 744-765, Mar.

1998.

[2] V. Tarokh, H. Jafarkhani, A. R. Calderbank,"Space-Time Block Coding for Wireless Communications : Performance Results", IEEE Journal on Select Areas in Communications, Vol. 17, No. 3, March 1999.

[3] V. Tarokh, H. Jafarkhani, A. R. Calderbank,"Space-Time Block Codes from Orthogonal Designs"IEEE Transactions on Information Theory, Vol. 45. No. 5, July 1999.

[4] Y. Tang and M. Valenti,"Coded Transmit Macrodiversity Block Space-Time Codes Over Distributed Antennas", Presentation for the Virginia Tech. Symposium On Wireless Personal Communications, Lane Department of Comp. Sci. & Electrical Engineering-West Virginia University, USA, 2000.

Claims

CLAIMS 1. A communications system comprising an electricity distribution and/or transmission network formed from at least two substantially adjacent conductors connected to a plurality of electricity delivery points, the system further comprising first and second modems located at or adjacent electricity delivery points of the electricity distribution network and configured to transmit data at communications frequencies and to filter out signals at mains electricity supply frequencies, wherein the first modem is arranged to transmit coded data onto the at least two substantially adjacent conductors so that the inductive interaction of the conductors provides space time coding of the data.
2. A communications system according to claim 1, wherein the electricity distribution and/or transmission network is formed from three substantially adjacent conductors, and the first modem is arranged to transmit coded data onto two or three substantially adjacent conductors.
3. A communications system according to claim 2, wherein the electricity distribution and/or transmission network is formed from four substantially adjacent conductors, and the first modem is arranged to transmit coded data onto two, three or four substantially adjacent conductors.
4. A communications system according to any preceding claim, wherein the data is coded using a channel coding apparatus.
5. A communications system according to claim 4, wherein the channel coding apparatus comprises trellis coding means or feedback coding means, or feed-forward coding means with training information, or blind coding means.
6. A communications system according to any preceding claim, wherein the coded data is passed via a serial to parallel converter before transmission.

<Desc/Clms Page number 17>
7. A communications system according to claim 6, wherein following the serial to parallel converter, the coded data passes via a pulse shaper and a modulator before transmission.
8. A communications system according to any preceding claim, wherein the second modem is arranged to use a decision metric to retrieve the coded data from received signals.
9. A communications system according to any preceding claim, wherein the second modem is connected to two substantially adjacent conductors of the electricity distribution and/or transmission network.
10. A communications system according to any of claims 1 to 8, wherein the second modem is connected to three or more substantially adjacent conductors of the electricity distribution and/or transmission network.
11.'A communications system according to any preceding claim, wherein the first modem is arranged to transmit test signals over a range of frequencies and time intervals, the second modem is arranged to receive the test signals and upon receiving the test signals to transmit data describing properties of the received signals, and the first modem is arranged to receive the transmitted data and use the data to select appropriate combinations of frequencies and time intervals for the subsequent transmission of coded data.
12. A communications system according to claim 11, wherein the first modem is arranged to take into account the quantity of data to be transmitted when selecting the combinations of frequencies and time intervals.
13. A communications system according to any preceding claim, wherein the first modem is arranged to transmit test signals on different conductors of the network, the second modem is arranged to receive the test signals and upon receiving the test signals to transmit data describing properties of the received signals, and the first

<Desc/Clms Page number 18>

modem is arranged to receive the transmitted data and use the data to select appropriate conductors for the subsequent transmission of coded data.
14. A communications system according to claim 13, wherein the first modem is arranged to transmit test signals on more than one conductor simultaneously, and is arranged to use data received from the second modem to select appropriate combinations of conductors for the subsequent transmission of coded data.
15. A communications system according to claim 13 or claim 14, wherein the first modem is arranged to dynamically switch between conductors when transmitting test signals, and is arranged to use data received from the second modem to select appropriate dynamic switching rates for the subsequent transmission of communications data.
16. A communications system according to any preceding claim, wherein the communications data includes an identifying characteristic which identifies that modem which is the intended recipient of the communications data.
17. A communications system according to any preceding claim, wherein the electricity distribution and/or transmission network connects buildings to an electricity sub-station.
18. A communications system according to claim 17, wherein the network is a three-phase network comprising three live conductors and a neutral conductor.
19. A communications system according to any of claims 1 to 16, wherein the electricity distribution and/or transmission network connects electricity sub-stations.
20. A communications system according to any of claims 1 to 16, wherein the electricity distribution and/or transmission network is located within a building, and provides electricity to power sockets.

<Desc/Clms Page number 19>
21. A communications system according to claim 20, wherein the network is monitored regularly for changes of transmission properties of the network at communications frequencies.
22. A method of communicating via an electricity distribution and/or transmission network formed from at least two substantially adjacent conductors connected to a plurality of electricity delivery points, first and second modems being located at or adjacent electricity delivery points of the electricity distribution network and configured to transmit data at communications frequencies and to filter out signals at mains electricity supply frequencies, wherein the method comprises transmitting coded data onto the at least two substantially adjacent conductors so that the inductive interaction of the conductors provides space time coding of the data.
23. A communications system substantially as hereinbefore described with reference to the accompanying figures.
24. A method of communicating substantially as hereinbefore described with reference to the accompanying figures.