Power line modelling for creating PLC communication system

Transcription

1 Power line modelling or creating PLC communication system P. Mlynek, M. Koutny, and J. Misurec Abstract The article presents a design o the power line communication model. This model is composed o communication model, model o power line and model. The communication model is realized as the OFDM system, power line are modelled rom transer unction o two-port network or rom transer unction o multipath signal environment. Noise model are modelled as white, which gets a spectral colouring by a ilter. On the resulting PLC communication model was shown comparison o dierent modulation technique and coding scheme. The dierent levels o mapping the carrier requencies in OFDM were simulated on the proposed model rom the viewpoint o comparing symbol error with signal to ratio and the intererence eect to symbols rearrange in the constellation diagram were simulated as well. Keywords power line, OFDM,, modelling, modulation, coding T I. INTRODUCTION HE PLC technology (Power Line Communication) uses a power lines or its data communication. PLC technology takes proits rom the advantage o not requiring any additional wiring. There is a need to use distribution lines o electricity or the control signals, IP telephony transer, and remote data acquisition [1],[] and [3]. It comes up almost simultaneously with electrical power network developing. This technology becomes more and more important. It is primary given with growing need o data channels using or s communication with meter equipments and control systems in energy industries. Its epansion is epected with AMR (Automated Meter Reading) and AMM (Automated Meter Management) systems coming. They are part o new metering and controlling trend co-called Smart Metering. The PLC technology should be as an alternative to other eisting data channels [4]. PLC systems all into two areas: wideband PLC and narrowband PLC. Wideband PLCs achieve the characteristics o wideband communication, enabling, or eample, ast Internet access or implementation o small LAN networks. Narrow- Manuscript received June 9, 010. The paper has been supported by the Czech Science Foundation project GACR 10/09/1846, Ministry o Education o the Czech Republic project No. MSM , and by VUT grant FEKT-S P. Mlynek 1, M. Koutny, J. Misurec 3 are with the Department o Telecommunications, Brno University o Technology, Purkynova 118, Brno 6100, Czech Republic ( s: mlynek@eec.vutbr.cz 1, koutnym@eec.vutbr.cz, misurec@eec.vutbr.cz 3 ) band PLCs today seem to be a little in the background. This is, o course, given only by the area o applicability. Power networks can also be used or other applications, which would be hard to implement in practice through another type o communication. Speciic services include central management o power consumption, tariing, remote meter reading, commanding, etc. There are a lot o ailings or widely using o this technology. An intererence o useul signal, smaller range o useul signal and equipments o energy network are the main. From the analysis o partial problems, there is a better to have a mathematical computer model o power line which it would enable a simulation o data transmission with power lines. II. THE USING OF PLC TECHNOLOGY The narrow-band PLC systems are used mainly in automation [3]. The automation systems based on this technology are implemented without an additional installation o communication networks. This means that high costs, which are necessary or the installation and realization o new networks within the eisting buildings, are substantially reduced by the use o PLC technology. The automation systems implemented by using the PLC can be used or various tasks: Control o various devices which are connected to internal wiring, or eample: lighting, heating, air conditioning, etc. Central control o various home systems, such as: windows and control o doors. Security unctions, sensor control, etc. Wideband PLCs achieve the characteristics o wideband communication, enabling, or eample, ast Internet access or implementation o small LAN networks. III. PLC COMMUNICATION SYSTEM For a creating o the complete PLC communication system, there is necessary to create model o channels as well as model and a transmitter and a receiver models. The complete PLC model will be created rom particular models. There will be possible to create analysis o a concrete power line based on the simulations o this system with various models o lines. The analysis will be possible to judge in term o possibility to using o various combinations o PLC technologies, security transer, modulations, coding etc. So that there will be obtained to best 13

2 parameters o data transer in mentioned systems. It is necessary to create the channel models or the PLC simulation. There are more possibilities o power line model creating. First o them is the power line model as environment with multipath signal propagation. The parameters o this line are obtained rom a distribution network topology or based on metering. Second o them is model, which applies chain parameter matrices to describing the relation between input and output voltage and current by two-port network. In design o PLC system, it is necessary to bear in mind the character o transmission medium and intererences which the PLC communication is inluenced. It is necessary to ind useable guard coders, modulation technique and encryption to ensure o security and oolproo communication with smallest error rate between data source and receiver. The PLC communication system is possible to divide to particular parts or purpose o modelling: PLC communication model, Source Models o power lines, Noise model. IV. OFDM MODEL FOR PLC COMMUNICATION The model o PLC communication is created by a transmitter, receiver and channel block. It serves or a creating o a source and destination o data communication or subsequent simulations o lines model which they are replaced by block o channel. The basic PLC communication model with OFDM system is shown in the Fig.1 [5], [6] and [7]. Cyclic prei (CP) Input data Coding Interleaving Mapping Pilot insertion S/P converter IFFT P/S converter Receiver Channel Removing CP Received data Decoding Deinterleaving De-mapping Channel estimation P/S converter FFT S/P converter Fig. 1 The basic model o PLC communication with OFDM system The coder adds the redundant inormation to the sequence o bits. I there is an error in bit chain the redundant inormation could be used or an error detection and correction by the help o detection and correction coders. The scheme o coding is designed or detection and the correction independent errors. The scheme is not designed or the bulk o errors. The interleaving technologies are used or elimination o evolve o bulk errors during the transer. The serial data transer is obtained rom block o coding. This block is connected to a block o mapping. There is a transer o bit sequence to a symbol sequence in block o mapping. A symbol distribution is a result o mapping. This distribution is shown in a constellation diagram and it is depended on chosen modulation. The pilot signals are necessary to include to the transer in case o continuous system detection. The estimation is important or determination o amplitude and phase o map s constellation each o subcarrier. The estimation o transer channel in the OFDM system requests a inserting o known symbols or a pilot structure to the OFDM signal. The useul data are transerred to the parallel stream in the S/P converter. The number o parallel streams matches to the number o carriers. These carriers will transer useul data. A protect interval is used in OFDM to prevent o ISI (inter symbol intererence). A cyclic prei (CP) is created by a ew o last samples o OFDM symbols. CP creates a protect 14

3 interval between adjacent transerred OFDM symbols in time area. This is a way how to keep orthogonally carries. IFFT block transers data rom requency to time area. V. THE SOURCES OF INTERFERENCE Besides simulation o transer characteristic, it is necessary to identiy possible sources o intererence because the power line has a signiicant attenuation o signal and various intererence and. Thereore the data transer has a high error rate without any checking algorithm. The undamental inluence on data transmission over power lines are mainly the negative characteristics o power networks. Can be summarized in a ew points: Mismatched impedance Attenuation on the communication channel Intererence () Intererence changing in time Fig. shows a simpliied block model o the PLC communication channel, in which the described characteristics and parameters are included. The parameters o intererence, ecept, are represented as a time variable linear ilter described by the requency parameters. Noise is depicted as additive random intererence process. N(t) (interering random procces) H() transmission channel unction Transmitter s(t) Asynchronous impulsive Colored background H() N(t) Transmission channel as linear ilter Sources o intererence Fig. PLC channel model r(t) Synchronous impulsive Narrow-band Receiver This model captures the whole range o parameters which are necessary or a model o the communication system with corresponding characteristics, although this model is schematically simpliied in the igure. Transmission unction and can be either estimated rom the measurement or derived rom the theoretical analysis. In this paper, we deal with our dierent types o [6]: Background : it is every time present in the network. It is caused by assembling o multiple sources o with low power. It can be described by a PSD (Power Spectral Density) that it declines with a growly requency. The background power density can be described with equation: 0 0 A ( ) A A e (1) where A is power density or,and A 0 is a dierences between A( ) and A(0). This model enables modelling background as a white process, which gets a spectral colouring by a ilter. Narrow-band : this primary originates rom the broadcasting stations that they transmit in a long, middle a short wave range. The amplitude can be changed in dependence on time and place. The narrow-band can be modelled as a sum o multiple sine with dierent amplitude: N n( t) A ( t) sin( t ) () i 1 i where N is a number o waves o dierencing requencies i, amplitude A i (t) and phase φ i. The amplitude A i (t) is a constant in simplest case but it can be established rom broadcast transmission. The phase φ i is randomly established rom interval [0;π]. Asynchronous impulsive : this type o is characterized by high and short spikes o voltage with length µs. These spikes can reach up to kv level. This is the cause o the switching equipments in the distribution network. These kinds o are considered as a part o background. Synchronous impulsive : is caused by thyristors in light dimmers and copiers. They are bursts o intererence spikes with repeating o period. Synchronous impulse can be modelled by a source o white with a spectral colouring together with a periodical switching o rectangular wrap (Fig. 3). White Periodic rectangular signal Spectral colouring ilter Fig. 3 Synchronous impulsive VI. POWER LINE MODEL A. Primary parameters o power line The line can be described by using R, L, C, G parameters. These parameters denote a resistance, inductance, capacity and leakage relative to the length o the line. Let's consider a single-phase signal distribution and management structure which is shown in Fig 4. The wire line includes phase, neutral and ground wires where each wire is i i Synchronous impulsive 15

4 inserted into the insulating sleeve and all wires are insulated with own sheath U A1 e Ae, (7) G j C I ( A1 e A e ). (8) R j L The parameters which describe the section o lines are the characteristic impedance and the speciic transer actor:. Fig.4 The shape o power line For the purpose o modeling, we consider the three-phase line as the two conductors and one transmission conductor, which they are a conductive core and they are surrounded by the same dielectric material. Then it is possible to determine the primary parameters with the equations [8]: d r 0 a, (3) R a d 1 a L C r cosh 0 1 d cosh, (4) a r 1 0, (5) d a G Ctan, (6) where d is a distance between the centre o conductors, a is a radius o conductor, σ is a copper conductivity, ε r indicates a relative permittivity o insulation, ε 0 is a permittivity o vacuum, tan δ is a actor variance a μ r is a relative magnetic permittivity o the copper. B. Substitute power line model The signal propagation over a substitute model o the power line is shown in Fig. 5 [4] [6]. + - Source Z S US U 1 U X R d Substitute power line model L d C d G d U X-dU X Load U Z L R j L Z C, (9) G j C j ( R j L )( G j C ). (10) I we consider the line, which it is equivalent to a wave propagating rom the source to the drain, the transer the unction o the line with length l can be determined by: H U( U( l) 0) l ( ) l j ( ) l ( ) e e e. (11) R ( ) L ( ) a G ( ) C ( ) (1) 1, 30 MHz. or (13) It was ound that the L 'and C' dependence on the requency is insigniicant. Thereore the characteristic impedance and the speciic transer actor can be simpliied: L Z C, (14) C 1 R ( ) Z c 1 G ( ) Z c j L C. (15) We replace R ( ) with equation (17) to describe the real part o transmission actors as a unction dependent on requency. R 0 ( ) (16) r where µ 0 and κ represent the constant permeability and conductivity and r is the radius o conductor. =0 +d Fig. 5 Signal propagation over a power line =l It knows thatg ( ) ~. It is possible to modiy the ormulation o the real part o transer actor now: The voltage and current or the line are obtained rom the theory o the elementary section o line. 16

5 ) 1 Z ( Re{ } C 0 ZC. (17) r I we replace the parameters line (Z C, r and other) by the constants k 1,k a k 3, we can obtain the real and imaginary part o the transer actor: ( ) Re{ } k k, (18) 1 ( ) Im{ } k. (19) 3 The equation (18) was approimate by constants a 0, a 1 a k : k ( ) a0 a1. (0) The transer actor represents the loss on the line relate to the length o line in equation (18). Thereore the transer actor is a unction o the length l. It is possible to determine the attenuation o line as an amplitude o the transer unction o the channel deined in equation () by the appropriate choice o parameters a 0,a 1 and k. A C. Power line models k ) l ( a a ) l (, l) e ( e 0 1 (1) The power line model is required to simulate PLC communications. The power line model applies the methods used to model electricity distribution networks. The chain parameter matrices describing the relation between input and output voltage and current o two-port network can be applied or the modelling the transer unction o power line channel [9]. Modelling power line as cascaded elementary two-ports network enable working with elementary networks, two-ports, which can be described by cascade parameters. Cascade solution oers the possibility o choosing a certain degree o compleity and precision o the line being modelled. It is also possible to deine individual blocks as macromodels describing a data channel. An eample can be seen in the solution given in [10]. Fig. 6 shown a eample o cascade connected two-port networks. unction. The resultant cascade matri rom the source to the load can be ormed applying the chain rule: n A () A i i 1 The transer unction o appropriate two-port network is given by equation: U L H (3) U S The other alternative to modelling power line is a method, that the power line channel is assumed to be a multipath propagation environment. The parameters o the channel are acquired based on the topology o the distribution network and on the channel measurements. The transmitted signal arrives in the receiver via the N signaling path. On path i the arriving signal is delayed by the time τ i and attenuated by the comple attenuation actor C i. The impulse response o the channel h(t ) can be written as a sum o the delayed and attenuated Dirac pulses [3]: h N N j i ( t) Ci ( t i ) H( ) Ci e (4) i 1 i 1 From the transer unction o power line model has been calculated the coeicient o ilter and the power line channel has been modelled as a digital ilter. The s models have been modelled according to chapter B. The power line model with model is shown in Fig.7. The particular has been made by a Simulink block s help. In Digital ilter Power line channel Fig. 7 Power line model Out Background Narrow-band Impulsive Vs ~ Zs A 1 A A 3 A 4 Coupling interace, parallel raction Coupling interace, series raction Coupling interace, series raction Fig.6 Cascade model o power line Coupling interace, parallel raction Each part o the line is described by a separate cascade matri A 1 to A 4. Internal series impedance Z s o signal source V s and parallel impedance o load Z L can also be described by cascade parameters and included in the resultant transer ZL VII. COMPARISON OF THE DIFFERENT MODULATION USING IN OFDM SYSTEM The inal model has been made rom OFDM model and power lines model together with s models in Matlab/Simulink [1], [13]. The testing o various types o modulations and coding or data communication over power line have been accomplished on created model. Fig. 8 shows a comparison o the BPSK, QPSK, 16-QAM, 64-QAM and 56- QAM modulations in OFDM system rom the viewpoint o 17

6 comparing bit error (BER) with the normalized signal to ratio (E b /N 0 ). I we set the desired value BER = 10 -, i.e. less than one aulty bit per hundred o the total value, that the desired value o normalized signal to ratio (E b /N 0 ) are shown in Tab. 1. TABLE I DESIRED VALUE OF NORMALIZED SIGNAL TO NOISE RATIO FOR BIT ERROR BER=10 - FOR PARTICULAR MODULATION Modulation BPSK QPSK QAM QAM QAM E b /N 0 [db] 4,7 7,1 10,3 14,8 19, Fig. 9 shows the bit error rate (BER) dependence on the normalized signal to ratio (E b /N 0 ) with using o various types o channels coding or 64QAM modulation. The best ability o correction achieves the convolution coder rom used types o channel coding. Fig. 10 Time or inormation transmission or dierent OFDM modulation Fig. 11 shows how much time is needed or the transmission o a given amount o data or particular coding. From this graph it ollows that the stronger the coding is, the more time is needed or the transmission o inormation. Fig. 10 shows how much time a particular modulation needs or the transmission o a ied amount o data. We have chosen symbols as the ied amount o data. Ater the transmission o these data, the simulation stops and the time o data transmission date is shown. The QPSK modulation was the least susceptible to intererence, but the transmission speed is the lowest whereas the 64-QAM modulation was the most susceptible to intererence, but the transmission speed is the highest. Fig. 11 Time or inormation transmission or 64-QAM 18

7 Fig. 1 Eect o SNR on constellation diagram o OFDM modulation with QPSK VIII. SIMULATION OF THE INTERFERENCE EFFECT BY QPSK AND 64-QAM USING There is one o the possible ways to view the eect o intererence. It is through constellation diagram. Thereore the eect o intererence to the symbols is simulated in constellation diagram or dierent modulations. The other way to show possibilities OFDM is bit rate error showing as we can see in [10]. A. OFDM with QPSK The basic symbol layout in constellation diagram and its behavior or the dierent levels o ratio (SNR) is shown in Fig. 1 or the OFDM with QPSK modulation. It can be seen that the more the distance o the signal/ ratio (SNR) is reduced the more the symbols are dispersed in the constellation diagram. It can be seen that the modulation is relatively resistant to intererence rom Fig. 1. It is due to using a smaller number o states which QPSK has. B. OFDM with 64-QAM The basic symbol layout in constellation diagram and its behavior or the dierent levels o ratio (SNR) is shown in Fig. 13 or the OFDM with 64-QAM. It can be seen that the more the distance o the signal to ratio (SNR) is reduced the more the symbols are dispersed in the constellation diagram. It can be seen that the modulation 64-QAM is very 19

8 Fig. 13 Eect o SNR on constellation diagram o OFDM modulation with 64-QAM susceptible to intererence rom Fig. 13. It is due to using a large number o states which 64-QAM has. IX. CONCLUSION The progress in PLC technology has come about in the last decade. Remote data acquisition is now necessary because it is given by legal conditions. PLC technology is seem as an alternative data channel. The article deals with design o the PLC communication system model. The model is composed o the OFDM communication model, the model o power lines and model. The comple PLC communication model can be used or comparison o the perormance o dierent modulation and coding schemes and or uture standardization. From the results o simulations it ollows that an inappropriate choice o modulation and error correction coding can signiicantly aect the resulting signal. The results o simulations based on the model will be compared with measurements in the uture work. REFERENCES [1] P. Mlynek, J. Misurec, M. Koutny, The communication unit or remote data acquisition via the Internet, In Proceedings o the 7th WSEAS International Conerence on Circuits, systems, electronics, control and signal processing (CSES'08). Puerto de La Cruz, Spain: WSEAS Press, 008. s ISBN: [] R. Achmad, U. Agung, F. Tetsuo, K. Hidehiro, Y. T. Kyaw, U. Yoshiyori, U. Yuki, U. Yosuke, Wireless lan and power line communication platorm or e-learning multimedia system in underdeveloped area in lombok island, in 6th WSEAS Int. Con. on Electronics, Hardware, Wireless and Optical Communications, pages 3 8, 007. [3] C. J. Koinakis, J. K. Sakellaris, Integration o building envelope and services via control technologies, in Proceedings o the 13th WSEAS international Conerence on Systems, Rodos, Greece, July - 4, 009). [4] M. Orgoň, PLC/BPL and Net Generation Networks, in POWER- COM - Conerence Communication over MV and LV power lines, Praha, 007 [5] H. Hrasnica, A. Haidine, R. Lehnert, Broadband Powerline Communications Network Design, Willey, c s. ISBN

9 [6] C. Ahn, H. Harada, S. Takahashiz, Unitary matri requency modulated OFDM or power line communications over impulsive channels, in Proceedings o the 5th WSEAS international Conerence on Electronics, Hardware, Wireless and Optical Communications, Madrid, Spain, February 15-17, 006. Pp [7] M. Koutny, O. Krajsa, P. Mlynek, Modelling o PLC communication or supply networks, in Proceedings o the 13th WSEAS International Conerence on Communication. Rhodos: WSEAS Press, 009. s ISBN: [8] M. Babic, M. Hagenau, K. Dostert, J. Bausch, Theoretical postulation o PLC channel mode, Open PLC European Research Alliance (OPERA). 005 [9] I. Papaleonidopoulos, C. Karagiannopoulos, N. Theodorou, C. Anagnostopoulos,I. Anagnostopoulos, Modelling o indoor low voltage power-line cables in the high requency range, International Symposium on Power Line Communications and Its Applications (ISPLC). [10] T. Esmailian, F. Kschischang, G. Gulak, In-building power lines as high-speed communication channels: channel characterization and a test channel ensemble, International Journal o Communication Systems [11] J. Ahola, Applicability o power-line communicatins to data transer o on-line condition monitoring o electrical drives, Thesis or the degree o Doctor o Science (Technology). Lappeenranta University o Technology, Lappeenranta 003, ISBN , ISSN [1] The MathWorks [online] [cit ]. Online: < [13] M. I. Anis, M. Waqas, M.U. Zammil, Analysis and System Level Simulation o BER on dierent Modulation Techniques using OFDM, 1th WSEAS International Conerence on Communications, Heraklion, Greece, July 3-5, 008, pp Petr Mlýnek (Ing. (MSc). in 008 (FEKT VUT Brno) and was born on December 8, His master s thesis deal with testing security o the communication unit LAN o remote data acquisition against attacks rom the Internet. Currently, he is a student at Ph.D. degree. His current research interests communication over the power line channel. Martin Koutný received MSc at the Department o Telecommunications at the Faculty o Electrical Engineering and Computer Science at Brno University o Technology in 007. He is currently a PhD student at the Faculty o Electrical Engineering and Computer Science at Brno. Jiří Mišurec Ing. (MSc). in 1985, (BUT), Ph.D. in 1991 (BUT). His dissertation dealt with accuracy enhancement o voltage-to-requency converters. In this period he co-authored 7 patent-author certiicates. Ater completing his post-graduate studies he joined the Dept. o Telecommunications as an Assistant Proessor. From 1997 to 004 he was an employee at the irm "E.ON a.s. Brno", but he has participated in the research and teaching at FEEC BUT. Now he is again ully employed in FEEC BUT. His research interest is ocused on the area o analog technique, converters, especially on converters working both in voltage and current mode. Now he is interested in generalization o sensistivity analysis o transer unctions. This should be used or comparison o newly developed applications. In the latest he also cooperates with a number o companies on implementation o undamental research results into practise. 1