UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Communication channel characteristics and behaviour of intrabuilding power distribution circuits Chan, Morgan Hing-Lap 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A7 C42_5.pdf [ 5.06MB ]
Metadata
JSON: 831-1.0096358.json
JSON-LD: 831-1.0096358-ld.json
RDF/XML (Pretty): 831-1.0096358-rdf.xml
RDF/JSON: 831-1.0096358-rdf.json
Turtle: 831-1.0096358-turtle.txt
N-Triples: 831-1.0096358-rdf-ntriples.txt
Original Record: 831-1.0096358-source.json
Full Text
831-1.0096358-fulltext.txt
Citation
831-1.0096358.ris

Full Text

COMMUNICATION CHANNEL CHARACTERISTICS AND BEHAVIOUR OF INTRABUILDING POWER DISTRIBUTION CIRCUITS by Morgan Hing-Lap CHAN B.A.Sc. University of Ottawa, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n THE FACULTY OF GRADUATE STUDIES (DEPARTMENT OF ELECTRICAL ENGINEERING) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1985 © Morgan Hing-Lap CHAN, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of «^"c«A-The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6 (3/81) ABSTRACT Intrabuilding power distribution circuits offer a number of unique advantages for local area networking. To enable the selection of proper error-control codes and protocols for reliable data communication services, error pattern statistics of intrabuilding power line channels are obtained. Also, error-causing disturbances are identified and their relationships to specific types of error patterns are determined. It is found that error occurrence is highly periodic, with periodicity being a function of the power line frequency which is nominally 60Hz in North America. Furthermore, results indicate that error pattern behaviour is relatively insensitive to communication carrier frequency and modulation schemes. Based on the measurement results, hybrid ARQ with bit-interleaving is suggested for reliable data transmission at high data rate (19,200bps). Burst error correcting codes can be used to reduce decoder cost and complexity with some sacrifices in performance. At lower data rates (1,200bps or below), effective error control can be accomplished more easily. Finally, the attenuation characteristics of a number of typical power line channels are presented. It is found that high frequency bypass can be used to improve signal transmission between different phases of the distribution transformer. i i TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i i i LIST OF ILLUSTRATIONS v i ACKNOWLEDGEMENTS x i i 1 . INTRODUCTION 1 1.1 Communication Through I n t r a b u i l d i n g Power D i s t r i b u t i o n C i r c u i t s 1 1.2 Review of Previous Work 3 1.2.1 Channel E r r o r S t a t i s t i c s 3 1.2.2 S i g n a l A t t e n u a t i o n C h a r a c t e r i s t i c s -- 4 1.3 Scope of the T h e s i s 5 2. CHANNEL ERROR STATISTICS FOR A FSK SYSTEM 7 2.1 Burst E r r o r D e f i n i t i o n • 7 2.2 The Measurement System 8 2.2.1 The Burst E r r o r Analyser 10 2.2.2 The T r a n s m i t t e r and Receiver 11 2.3 Measurement R e s u l t s : I n d u s t r i a l B u i l d i n g 11 2.3.1 Channel E r r o r S t a t i s t i c s : 19,200bps Data Rate, 120KHz C a r r i e r Frequency 12 2.3.2 Channel E r r o r S t a t i s t i c s : 4,800bps Data Rate, 120KHz C a r r i e r Frequency 30 2.3.3 Channel E r r o r S t a t i s t i c s : 1,200bps Data Rate, 120KHZ C a r r i e r Frequency 34 2.3.4 Channel E r r o r S t a t i s t i c s at Other C a r r i e r Frequencies 38 2.4 Measurement R e s u l t s : R e s i d e n t i a l Apartment Complex 48 3. CHANNEL ERROR STATISTICS FOR A PSK SYSTEM — 53 3.1 The PSK Receiver 53 i i i 3.2 Measurement R e s u l t s : 38,400bps Data Rate 54 3.3 Measurement R e s u l t s : 19 ,200bps Data Rate 54 3.4 Measurement R e s u l t s : 4,800bps and 1,200bps Data Rate --" 60 4. CHANNEL ERROR STATISTICS FOR SPREAD SPECTRUM SYSTEMS 63 4.1 Direct-Sequence SSMA Frequency-Shift-Keyed (DS/SSMA-FSK) System 64 4.1.1 The T r a n s m i t t e r and Receiver 64 4.1.2 Channel E r r o r S t a t i s t i c s : 4,800bps Data Rate, 19,200bps Code Rate, 120KHZ C a r r i e r Frequency 66 4.1.3 Channel E r r o r S t a t i s t i c s : 1,200bps Data Rate, 19,200bps Code Rate, 12OKHz C a r r i e r Frequency 66 4.1.4 Channel E r r o r S t a t i s t i c s at 60KHz C a r r i e r Frequency 71 4.2 Direct-Sequence SSMA Phase-Shift-Keyed (DS/SSMA-PSK) System 71 4.2.1 The T r a n s m i t t e r and Receiver 71 4.2.2 Channel E r r o r S t a t i s t i c s : 38,400bps Code Rate, 115KHz C a r r i e r Frequency 74 5. IMPLICATIONS OF THE TEST RESULTS 79 5.1 Forward E r r o r C o r r e c t i o n (FEC) Coding 79 5.2 Automatic Repeat Request (ARQ) S t r a t e g i e s -- 80 5.3 ARQ with FEC Coding 83 5.4 B i t I n t e r l e a v i n g 84 5.5 Summary 84 6. TRANSMISSION CHARACTERISTICS OF INTRABUILDING POWER LINE CHANNELS 87 6.1 The Measurement System 88 6.2 A t t e n u a t i o n C h a r a c t e r i s t i c s : I n d u s t r i a l B u i l d i n g 89 6.3 A t t e n u a t i o n C h a r a c t e r i s t i c s : R e s i d e n t i a l Environment 96 6.4 A t t e n u a t i o n C h a r a c t e r i s t i c s : H o s p i t a l 98 i v. 6.5 E f f e c t s of E x t e r n a l L o a d i n g on S i g n a l T r a n s m i s s i o n 102 6.6 Comments on T r a n s m i s s i o n D i s t a n c e 107 6.7 Summary 109 7 . CONCLUSION 1 1 1 7.1 C o n c l u d i n g Remarks 111 7.2 S u g g e s t i o n s f o r F u t u r e Work 115 REFERENCES 1 1 6 v LIST OF ILLUSTRATIONS Page F i g . 2.1 D e f i n i t i o n of burst e r r o r . •-- 8 F i g . 2.2 E r r o r s t a t i s t i c s measurement system. 9 F i g . 2.3 Burst e r r o r a n a l y s e r . 10 F i g . 2.4 Repr e s e n t a t i v e burst e r r o r d i s t r i b u t i o n . FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 13 F i g . 2.5 Repr e s e n t a t i v e e r r o r - f r e e l e n g t h d i s t r i b u t i o n . FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 15 F i g . 2.6 Power l i n e n o i s e . 16 F i g . 2.7 Observed e r r o r p a t t e r n . 16 F i g . 2.8 Channel e r r o r s t a t i s t i c s under very strong and s t a t i o n a r y impulse noise impairment. FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 18 F i g . 2.9 P e r i o d i c impulse n o i s e . 19 F i g . 2.10 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 120KHZ c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 21 F i g . 2.11 Re p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 22 F i g . 2.12 Channel e r r o r s t a t i s t i c s under severe p e r i o d i c s i g n a l f a d i n g at subpowerline frequency. FSK: 19,200bps data r a t e , 120KHZ c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 24 F i g . 2.13 Burst e r r o r d i s t r i b u t i o n : Comparison between normal channel and f a d i n g channel. 25 F i g . 2.14 P e r i o d i c s i g n a l f a d i n g at twice the power l i n e v o l t a g e frequency. 28 F i g . 2.15 Channel e r r o r s t a t i s t i c s under severe p e r i o d i c s i g n a l f a d i n g at twice the power l i n e v o l t a g e frequency. FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . - 29 v i F i g . 2.16 E f f e c t of t r a n s c e i v e r s e p a r a t i o n on channel e r r o r s t a t i s t i c s . BER = 2 x 10~3 . 31 F i g . 2.17 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 4,800bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 32 F i g . 2.18 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 4,800bps data r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 33 F i g . 2 . 19 Channel e r r o r s t a t i s t i c s under severe p e r i o d i c s i g n a l f a d i n g at subpowerline frequency. FSK: 4,800bps data r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 35 F i g . 2.20 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 1,200bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 36 F i g . 2.21 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 1,200bps data r a t e , 120KHZ c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 37 F i g . 2.22 Background power l i n e n o i s e . 39 F i g . 2.23 E f f e c t of v a r i a t i o n of c a r r i e r frequency on e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 105, 110, 120, 125 KHz c a r r i e r f r e q u e n c i e s . Across-phase t r a n s m i s s i o n . BER = 10"3 . 40 F i g . 2.24 E f f e c t of v a r i a t i o n of c a r r i e r frequency on e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 105, 110, 120, 125 KHz c a r r i e r f r e q u e n c i e s . In-phase t r a n s m i s s i o n . BER = 10 3 . 41 F i g . 2.25 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 60KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 42 F i g . 2.26 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 60KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 43 F i g . 2.27 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 4,800bps data r a t e , 60KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 44 F i g . 2.28 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 4,800bps data r a t e , 60KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 45 v i i F i g . 2.29 Re p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 1,200bps data r a t e , 60KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 46 F i g . 2.30 Re p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 1,200bps data r a t e , 60ZKHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 47 F i g . 2.31 Channel e r r o r s t a t i s t i c s under p e r i o d i c impulse s p i k e s p l u s a p e r i o d i c impulse b u r s t s impairment. FSK: 19,200bps data r a t e , 120KHZ c a r r i e r frequency. In-phase t r a n s m i s s i o n . 50 F i g . 2.32 Channel e r r o r s t a t i s t i c s under p e r i o d i c impulse b u r s t s impairment. FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. Off-phase t r a n s m i s s i o n . 51 F i g . 2.33 Channel e r r o r s t a t i s t i c s under a p e r i o d i c impulse b u r s t s impairment. FSK: 19,200bps data r a t e , 120KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 52 F i g . 3.1 PSK r e c e i v e r . 54 F i g . 3.2 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 38,400bps data r a t e , 115KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 55 F i g . 3.3 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 38,400bps data r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 56 F i g . 3.4 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 19,200bps data r a t e , 115KHz c a r r i e r frequency. In-phase t r a n s m i s s i o n . 57 F i g . 3.5 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 19,200bps data r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 58 F i g . 3.6 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 4,800bps data r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 61 F i g . 3.7 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . PSK: 1,200bps data r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 62 F i g . 4.1 (a) DS/SSMA-FSK t r a n s m i t t e r , (b) DS/SSMA-FSK r e c e i v e r . 65 v i i i F i g . 4.2 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 4,800bps data r a t e , 19,200bps code r a t e , 120KHZ c a r r i e r frequency. In-phase t r a n s m i s s i o n . 67 F i g . 4.3 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 4,800bps data r a t e , 19,200bps code r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 68 F i g . 4.4 Re p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 1,200bps data r a t e , 19,200bps code r a t e , 120KHZ c a r r i e r frequency. In-phase t r a n s m i s s i o n . 69 F i g . 4.5 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 1,200bps data r a t e , 19,200bps code r a t e , 120KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 70 F i g . 4.6 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 4,800bps data r a t e , 19,200bps code r a t e , 60KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 72 F i g . 4.7 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-FSK: 1,200bps data r a t e , 19,200bps code r a t e , 60KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 73 F i g . 4.8 (a) DS/SSMA-PSK t r a n s m i t t e r , (b) DS/SSMA-PSK r e c e i v e r . 74 F i g . 4.9 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-PSK: 19,200bps data r a t e , 38,400bps code r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 75 F i g . 4.10 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-PSK: 4,800bps data r a t e , 38,400bps code r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 76 F i g . 4.11 R e p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . DS/SSMA-PSK: 1,200bps data r a t e , 38,400bps code r a t e , 115KHz c a r r i e r frequency. Across-phase t r a n s m i s s i o n . 78 F i g . 5.1 Throughput e f f i c i e n c i e s : the i d e a l s e l e c t i v e - r e p e a t ARQ with i n f i n i t e b u f f e r and the go-back-N ARQ with code block l e n g t h n = 2024. — 81 ix F i g . 6.1 Experimental setup f o r a t t e n u a t i o n measurement. 88 F i g . 6.2 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : i n d u s t r i a l b u i l d i n g . In-phase s i g n a l t r a n s m i s s i o n through 20-30 f e e t of power d i s t r i b u t i o n w i r i n g . 90 F i g . 6.3 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : i n d u s t r i a l b u i l d i n g . In-phase s i g n a l t r a n s m i s s i o n through an unknown path at day time. 92 F i g . 6.4 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : i n d u s t r i a l b u i l d i n g . Across-phase s i g n a l t r a n s m i s s i o n through unknown paths at day time. 93 F i g . 6.5 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : i n d u s t r i a l b u i l d i n g . In-phase s i g n a l t r a n s m i s s i o n through an unknown path at night time. 94 F i g . 6.6 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : i n d u s t r i a l b u i l d i n g . Across-phase s i g n a l t r a n s m i s s i o n through unknown paths at night time. 95 F i g . 6.7 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : r e s i d e n t i a l complex ( l o w - r i s e ) . In-phase and off-phase s i g n a l t r a n s m i s s i o n through unknown paths. 97 F i g . 6.8 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : r e s i d e n t i a l complex ( h i g h - r i s e ) . In-phase s i g n a l t r a n s m i s s i o n through a short path. 99 F i g . 6.9 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : r e s i d e n t i a l complex ( h i g h - r i s e ) . In-phase s i g n a l transmssion through unknown paths. 100 F i g . 6.10 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : h o s p i t a l . In-phase s i g n a l t r a n s m i s s i o n through an unknown path. 101 F i g . 6.11 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : h o s p i t a l . Across-phase s i g n a l t r a n s m i s s i o n through unknown paths. •• 103 F i g . 6.12 A t t e n u a t i o n v s . frequency c h a r a c t e r i s t i c s : s i n g l e f a m i l y home. In-phase and off-phase s i g n a l t r a n s m i s s i o n without s p e c i f i c l o a d i n g switched on. 105 x F i g . 6.13 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : In-phase s i g n a l t r a n s m i s s i o n w i t h water heater and TV r e c e i v e r switched on. 106 F i g . 6.14 A t t e n u a t i o n vs. frequency c h a r a c t e r i s t i c s : s i n g l e f a m i l y home. Off-phase s i g n a l t r a n s m i s s i o n with e l e c t r i c range and c l o t h e s dryer switched on. 107 x i ACKNOWLEDGEMENTS I would l i k e to express my s i n c e r e g r a t i t u d e to my s u p e r v i s o r Dr. R.W. Donaldson f o r h i s s u p e r v i s i o n , encouragement and support. I would l i k e to thank F. Chiu f o r d i s c u s s i o n s and making a v a i l a b l e the FSK modem. I would a l s o l i k e to thank E. Casas, J . Poon and C. Kwong f o r t h e i r suggestions. F i n a l l y , s p e c i a l thanks go to Miss Camy Kwong who c o n s c i e n t i o u s l y typed the manuscript and i t s r e v i s i o n s . F i n a n c i a l support from NSERC i s g r a t e f u l l y a p p r e c i a t e d . - 1 -1. INTRODUCTION 1.1 Communication Through I n t r a b u i l d i n q Power  D i s t r i b u t i o n C i r c u i t s The demand f o r l o c a l area networks f o r o f f i c e automation, s e c u r i t y monitoring, energy management, computer communications and i n t r a b u i l d i n g environmental c o n t r o l enjoys ever i n c r e a s i n g demand. I n s t a l l a t i o n of communication c a b l e s fo r such purposes may i n v o l v e s u b s t a n t i a l i n s t a l l a t i o n c o s t s , appearance compromises to b u i l d i n g i n t e r i o r s , inconvenience or l i m i t a t i o n s on equipment l o c a t i o n s . Radio or i n f r a - r e d l i n k s are p o t e n t i a l a l t e r n a t i v e s ; however, r a d i o i m p l i e s l i c e n s i n g and i n t e r f e r e n c e , while i n f r a - r e d r e q u i r e s 1 i n e - o f - s i g h t ' t r a n s m i s s i o n . Another a l t e r n a t i v e i s t o u t i l i z e a channel which i s a l r e a d y a v a i l a b l e , namely the i n t r a b u i l d i n g e l e c t r i c power d i s t r i b u t i o n network [1-2], T h i s network i s almost u n i v e r s a l i n coverage and i s e a s i l y i n t e r f a c e d using a standard w a l l plug and o u t l e t s o c k e t . Power l i n e channels do s u f f e r from frequency s e l e c t i v e a t t e n u a t i o n , r e l a t i v e l y high n o i s e l e v e l , and v a r y i n g l e v e l s of impedance, n o i s e and a t t e n u a t i o n [2-4]. A p propriate e r r o r c o n t r o l i s needed for o p e r a t i o n on any data channel to p r o v i d e high r e l i a b i l i t y . In p a r t i c u l a r , - 2 -forward e r r o r c o r r e c t i o n (FEC) cod i n g or e r r o r d e t e c t i o n and r e t r a n s m i s s i o n are the most commonly used methods to achieve such improvement. Both e r r o r c o r r e c t i o n and r e t r a n s m i s s i o n imply the use of e r r o r c o n t r o l codes; however, the s e l e c t i o n of an e f f i c i e n t and r e l i a b l e code r e q u i r e s d e t a i l e d knowledge of channel e r r o r s t a t i s t i c s . T h i s t h e s i s d e s c r i b e s e x t e n s i v e measurements and a n a l y s i s of channel e r r o r s t a t i s t i c s f o r a number of i n t r a b u i l d i n g power l i n e channels. S t a t i s t i c s o btained i n c l u d e burst l e n g t h as w e l l as i n t e r - b u r s t ( e r r o r - f r e e ) l e n g t h d i s t r i b u t i o n s . R e s u l t s were ob t a i n e d at v a r i o u s c a r r i e r f r e q u e n c i e s (from 60KHz-120KHz) and data r a t e s (from 1,200-38,400bps), with a t t e n t i o n focused on high speed t r a n s m i s s i o n (19,200bps or above). In a d d i t i o n , the e r r o r - c a u s i n g d i s t u r b a n c e s are i d e n t i f i e d and t h e i r e f f e c t s on e r r o r p a t t e r n behaviour are determined. Narrow-band (FSK, PSK) s i g n a l l i n g as w e l l as wide-band s i g n a l l i n g (DS/SSMA-FSK, DS/SSMA-PSK) schemes were used i n the t e s t s . The measurements were performed at v a r i o u s days and time p e r i o d s , r e f l e c t i n g d i f f e r e n t l e v e l s of e l e c t r i c a l a c t i v i t i e s on the channels. A t t e n u a t i o n , n o i s e and impedance are among the most fundamental c h a r a c t e r i s t i c s of any communication channel, i n c l u d i n g power l i n e c i r c u i t s . These b a s i c channel f a c t o r s - 3 -are always the s u b j e c t s of. e x t e n s i v e study as a good understanding of them i s d e s i r a b l e . Recent s t u d i e s [3-4] have r e v e a l e d the behaviour of nois e and impedance found on r e s i d e n t i a l power d i s t r i b u t i o n c i r c u i t ; however, d e t a i l e d measurements on the a t t e n u a t i o n aspect of power l i n e channel are c u r r e n t l y u n a v a i l a b l e . Thus, the a t t e n u a t i o n c h a r a c t e r i s t i c as a f u n c t i o n of frequency f o r v a r i o u s types of power l i n e channels were examined and compared. E f f e c t s of some common household a p p l i a n c e s on s i g n a l t r a n s m i s s i o n through power l i n e channels were a l s o c o n s i d e r e d . The r e s u l t s obtained can be used i n the e f f e c t i v e design of communication systems i n v o l v i n g the use of i n t r a b u i l d i n g power l i n e channels. 1.2 Review of Previous Work 1.2.1 Channel E r r o r S t a t i s t i c s E r r o r c o n t r o l codes with v a r i o u s e r r o r h a n d l i n g a b i l i t i e s are r e a d i l y a v a i l a b l e i n the l i t e r a t u r e s [ 5 ] . However, the choice of an a p p r o p r i a t e code i s h i g h l y dependent upon the form of e r r o r p a t t e r n s encountered, p a r t i c u l a r l y when the code i s used f o r e r r o r c o r r e c t i o n . When FEC coding i s employed, the use of a code with e x c e s s i v e redundancy may degrade system throughput, while the use of - 4 -too l i t t l e redundancy may i n c r e a s e the b i t e r r o r r a t e of the system. In order to choose a p p r o p r i a t e codes f o r e r r o r c o n t r o l , d e t a i l e d knowledge of channel e r r o r s t a t i s t i c s i s r e q u i r e d , i n c l u d i n g burst e r r o r l e n g t h as w e l l as i n t e r - b u r s t l e n g t h d i s t r i b u t i o n s . Some f i e l d measurements of e r r o r p a t t e r n s on t r a n s e q u a t o r i a l HF r a d i o l i n k s [ 6 ] , and on urban lan d mobile r a d i o channels at the 463MHz band [7] are a v a i l a b l e , and s i m i l a r r e s u l t s e x i s t f o r simulated VHF, UHF urban land mobile r a d i o channels at the 800MHz band [8-9], However, there have been no r e p o r t s of e r r o r p a t t e r n s t a t i s t i c s f o r i n t r a b u i l d i n g power l i n e channels. 1.2.2 S i g n a l A t t e n u a t i o n C h a r a c t e r i s t i c s S i g n a l a t t e n u a t i o n i s one of the fundamental c h a r a c t e r i s t i c s of any communication channel. With t h i s and other knowledge, such as channel noi s e and impedance l e v e l s , one can estimate the amount of t r a n s m i t t e r power needed to achieve a s p e c i f i c l e v e l of performance (average b i t e r r o r r a t e ) . Such knowledge a l s o enables e f f i c i e n t u t i l i z a t i o n of the frequency spectrum. R e c e n t l y , Vines and others [3-4] measured the noise and - 5 -i m p e d a n c e l e v e l s o f r e s i d e n t i a l p o w e r d i s t r i b u t i o n c i r c u i t s . A s u r v e y o f p o w e r l i n e n o i s e l e v e l s a t a number o f o t h e r l o c a t i o n s , r a n g i n g f r o m an u r b a n b u s i n e s s o f f i c e t o a r u r a l f a r m , was r e p o r t e d by S m i t h [ 1 0 ] . I m p e d a n c e m e a s u r e m e n t s on p o w e r d i s t r i b u t i o n s y s t e m s t h r o u g h o u t t h e U n i t e d S t a t e s a n d E u r o p e a n c o u n t r i e s w e r e r e p o r t e d i n [ 1 1 - 1 2 ] . H o w e v e r , v e r y l i t t l e p r e v i o u s w o r k h a s b e e n d o n e t o r e v e a l t h e t r a n s m i s s i o n c h a r a c t e r i s t i c s o f i n t r a b u i l d i n g p o w e r l i n e c h a n n e l s . O s c h n e r [ 2 ] h a d r e p o r t e d some r e s u l t s i n t h i s r e g a r d ; n e v e r t h e l e s s , e x t e n s i v e r e s u l t s o t h e r t h a n t h o s e p r e s e n t e d i n t h i s t h e s i s a r e n o t a v a i l a b l e . 1 .3 S c o p e o f t h e T h e s i s I n c h a p t e r 2 , t h e d e f i n i t i o n o f b u r s t e r r o r s i s p r e s e n t e d , f o l l o w e d by a d e s c r i p t i o n o f t h e o v e r a l l e r r o r s t a t i s t i c s m e a s u r e m e n t s y s t e m . R e s u l t s o f a FSK s y s t e m o p e r a t e d i n a n i n d u s t r i a l b u i l d i n g a n d an a p a r t m e n t c o m p l e x a r e t h e n g i v e n , f o r v a r i o u s d a t a r a t e s a n d c a r r i e r f r e q u e n c i e s . I n c h a p t e r 3 , s t a t i s t i c s o f e r r o r p a t t e r n s o b t a i n e d u s i n g a PSK s y s t e m a r e p r e s e n t e d . The r e s u l t s a r e c o m p a r e d w i t h t h o s e o b t a i n e d u s i n g t h e FSK s y s t e m . The g e n e r a l p r o b a b i l i t y o f e r r o r p e r f o r m a n c e o f s y s t e m s i n t h e p r e s e n c e - 6 -of impulse noise (which i s a dominant cause of e r r o r s i n power l i n e channels) i s d i s c u s s e d . In chapter 4 , a DS/SSMA-FSK and a DS/SSMA-PSK system are d e s c r i b e d . S t a t i s t i c s of e r r o r p a t t e r n s obtained using the two systems are presented. In chapter 5 , v a r i o u s e r r o r c o n t r o l schemes are c o n s i d e r e d , on the b a s i s of the measurement r e s u l t s and o b s e r v a t i o n s . Suggestions f o r the s e l e c t i o n of e r r o r - c o r r e c t i n g codes and design of e r r o r c o n t r o l p r o t o c o l s are g i v e n . The o v e r a l l design philosophy i s then d e s c r i b e d . In chapter 6 , the t r a n s m i s s i o n c h a r a c t e r i s t i c s of a number of t y p i c a l power l i n e channels are i n v e s t i g a t e d . The e f f e c t s of some common household a p p l i a n c e s on s i g n a l t r a n s m i s s i o n are presented. C o n c l u s i o n s and suggestions f o r f u t u r e work appear i n chapter 7. - 7 -2. CHANNEL ERROR STATISTICS FOR A FSK SYSTEM E r r o r s t a t i s t i c s measurements f o r i n t r a b u i l d i n g power l i n e channels were c a r r i e d out on a c t u a l power d i s t r i b u t i o n c i r c u i t s . An i n d u s t r i a l b u i l d i n g having a l a r g e v a r i e t y of loads, i n c l u d i n g i n d u s t r i a l machinery, power f a c t o r c o r r e c t i o n and motor s t a r t e r c a p a c i t o r s , computers and s p e c i a l i z e d equipment, was chosen. T e s t s were a l s o done at a r e s i d e n t i a l complex which has approximately 50 apartment u n i t s . R e s u l t s were obtained at v a r i o u s data r a t e s (1,200-19,200bps), c a r r i e r f r e q u e n c i e s (60-120KHZ) and time p e r i o d s , with the average b i t e r r o r r a t e (BER) as the parameter. S i t u a t i o n s where the t r a n s m i t t e r and r e c e i v e r were p l a c e d on the same phase of a s t e p down transformer, and on d i f f e r e n t phases (l e g s ) of a s t e p down transformer have been s t u d i e d . 2.1 Burst E r r o r D e f i n i t i o n A t y p i c a l e r r o r p a t t e r n i s shown i n F i g . 2.1. A "0" denotes a c o r r e c t b i t , an "e" denotes an e r r o r b i t . - 8 -e OOOO OOOQeeOeOOOeOOOO OOOOe error- burst error-< free >c-region-s< free > region region Fig. 2.1 Definition of Burst Error Following [8-9], a burst region begins and ends with an error bit. A burst region may contain some correct bits whose run lengths are less than a specified value. On the other hand, an inter-burst (error-free) region consists of consecutive correct bits whose run length is equal to or greater than the specified value. A burst region is always preceded and followed by an error-free region and vice versa. The specified value used in the error statistics measurements was 50 bits. 2.2 The Measurement System The error stat i s t i c s measurements were carried out with the measurement setup shown in Fig. 2.2. The data source is a pseudorandom binary sequence (PRBS) generator with period equal to 211-1. The coupling network is a passive high pass f i l t e r used to block the 60Hz power voltage. The recovered data from the receiver output is compared with the locally - 9 -generated r e f e r e n c e data to o b t a i n the e r r o r p a t t e r n s . Data Source T r a n s m i t t e r Coupling Network Power L i n e Channel B u f f e r I-L o c a l l y Generated Data Receiver Coupling Network Burst Time > Computer E r r o r I n t e r v a l Analyser Counter F i g . 2 . 2 E r r o r s t a t i s t i c s measurement system The e r r o r p a t t e r n s are b u f f e r e d i n order to e l i m i n a t e p o s s i b l e g l i t c h e s produced as a r e s u l t of the e x c l u s i v e - o r o p e r a t i o n . The b u f f e r e d e r r o r p a t t e r n s are then fed i n t o the burst e r r o r a n a l y s e r which produces p u l s e s whose widths correspond to the l e n g t h of e r r o r b u r s t s as d e f i n e d i n S e c t i o n 2 . 1 . At the same time, i n t e r - p u l s e width corresponds - 10 -to the length of the error-free region. The outputs of the burst error analyser are measured by a time-interval counter, and the results are logged into a personal computer, s t a t i s t i c a l l y processed and plotted. 2.2.1 The Burst Error Analyser The burst error analyser is adopted from [8-9] and shown in Fig. 2.3. Fig. 2.3 Burst error analyser [8-9] The analyser monitors the error patterns for the above error-free length is the value set in the presettable counter. Error bursts as defined in Section 2.1 are detected and pulse widths proportional to the length of the error bursts are produced. Detailed descriptions of the operation mentioned specified error-free lengths. This spec i f ied - 1 1 -of the burst e r r o r a n a l y s e r can be found i n [ 8 ] . Predetermined e r r o r p a t t e r n s were used to v e r i f y proper o p e r a t i o n of the o v e r a l l measurement system. 2.2.2 The T r a n s m i t t e r and Receiver The t r a n s m i t t e r i s a FSK modulator with frequency o f f - s e t +_f equal to 1/4 of the data r a t e ; t h i s corresponds to the MSK modulation c o n d i t i o n [26-27] f o r that p a r t i c u l a r data r a t e . The c e n t r e f r e q u e n c i e s used i n the t e s t s were 60KHz and 120KHz. The r e c e i v e r c o n s i s t e d of a coherent FSK demodultor preceeded by band pass f i l t e r f o r noise s u p p r e s s i o n . The band pass f i l t e r has a 3-dB down bandwidth of about 40KHz ce n t e r e d around the ce n t e r frequency. T e s t s of the modem i n the presence of a d d i t i v e white gaussian n o i s e confirms proper f u n c t i o n i n g of the modem hardware. 2.3 Measurement R e s u l t s : I n d u s t r i a l B u i l d i n g E x t e n s i v e measurements were made on an i n d u s t r i a l b u i l d i n g having a l a r g e v a r i e t y of loads, such as i n d u s t r i a l machinery, power f a c t o r c o r r e c t i o n and motor . s t a r t e r - 1 2 -capacitors, computers, and other loads. Thi's building represents a very hostile environment for power line communications. In large buildings in North America including our test building, three-phase power is sent directly from a three-phase distribution transformer to circuit panels inside the building from where branch circuits carry power to desired locations. Placement of the transmitter and receiver on the same phase of the step down transformer will hereinafter be referred to as "in phase" signal transmission, while placement of the transmitter and receiver on different phases of the step down transformer will hereinafter be referred to as "across phase" transmission. The transmitter power was adjusted to obtain -2 average bit error rates in the c r i t i c a l range between 10 and 10 . The results presented were selected from the large number of available experiments to show both typical and rare error-causing disturbances that affect power line communications. 2.3.1 Channel Error Statistics: 19,200bps Data Rate, 120KHz  Carrier Frequency The cumulative distribution of burst error length obtained during typical in phase signal transmission is shown in Fig. 2.4. It is observed that single bit errors occupy a - 13 -Fig. 2.4 Representative burst error distribution. FSK: 19,200bps data rate, 120KHz carrier' frequency. In-phase transmission. - 1 4 -l a r g e percentage of the e r r o r s and burst e r r o r s composed of only a few b i t s predominate i n the t o t a l e r r o r occurrence. The cumulative d i s t r i b u t i o n of e r r o r - f r e e l e n g t h i s shown i n F i g . 2.5. F i g . 2.5 shows that the percentage of long e r r o r - f r e e l engths i s l a r g e . I t i s a l s o observed that there are many d i s c r e t e jumps i n the curves, which means that there are high r e l a t i v e f r e q u e n c i e s of some s p e c i f i c e r r o r - f r e e l e n g t h s . T h i s phenomenon suggests that e r r o r occurrence i s mostly p e r i o d i c . In other words, e r r o r occurrence must be somehow synchronous or c o r r e l a t e d with the p e r i o d i c 60Hz power l i n e v o l t a g e . P e r i o d i c e r r o r e f f e c t s can be seen from F i g . 2.6 and 2.7. In F i g . 2.6, the upper t r a c e i s the 60Hz power l i n e v o l t a g e ; the lower t r a c e i s the power l i n e n o i s e , a f t e r the r e c e i v e r ' s band pass f i l t e r . The noise p a t t e r n , p a r t i c u l a r l y the h i g h - l e v e l impulse n o i s e s p i k e s , i s seen to be c o r r e l a t e d with the p e r i o d i c 60Hz power v o l t a g e . In f a c t , the n o i s e - l e v e l p a t t e r n tends to repeat i t s e l f every h a l f power c y c l e . T h i s o b s e r v a t i o n of power l i n e noise g e n e r a l l y agrees with those r e p o r t e d i n [ 3 ] . In F i g . 2.7, the upper t r a c e i s the r e c e i v e d s i g n a l p l u s n o i s e , a f t e r the r e c e i v e r ' s band pass f i l t e r ; the lower t r a c e i s the e r r o r p a t t e r n . As one can see from the f i g u r e , a l o t - 15 -100 10 E r r o r - F r e e L e n g t h D i s t 10 10 Error-free Length (bits) F i g . 2.5 R e p r e s e n t a t i v e e r r o r - f r e e l e n g t h d i s t r i b u t i o n . FSK: 19,200bps data r a t e , 120KHZ c a r r i e r frequency. In-phase t r a n s m i s s i o n . Fig. 2.6 Power line noise. - 17 -of e r r o r s occur d u r i n g the presence of impulse noise s p i k e s and these e r r o r s are c o r r e l a t e d with the p e r i o d i c 60Hz power l i n e v o l t a g e as are the impulse noise s p i k e s . The r e c e i v e d s i g n a l l e v e l shown in the p i c t u r e was w e l l above the average background noise l e v e l . T h i s e x t r a r e c e i v e d power i s needed i n order to combat impulse noise d i s t u r b a n c e s . Such a r e c e i v e d s i g n a l l e v e l would enable v i r t u a l l y e r r o r - f r e e data t r a n s m i s s i o n i f the h i g h l e v e l noise impulses were absent. C l e a r l y , power l i n e communications are very v u l n e r a b l e to impulse n o i s e . Impulse noise e f f e c t s can be understood more c l e a r l y with the r e s u l t s shown i n F i g . 2.8 which were obtained d u r i n g the p e r i o d when very s t r o n g and s t a t i o n a r y p e r i o d i c impulse n o i s e o c c u r r e d . The n o i s e p a t t e r n was s i m i l a r to that shown i n F i g . 2.9, i . e . one hugh nois e spike every h a l f power c y c l e . The time i n t e r v a l between the noise s p i k e s was 1/120 second which was e q u i v a l e n t to 160 i n f o r m a t i o n b i t s f o r a 19,200bps data ra t e system. Since v i r t u a l l y a l l e r r o r s found under t h i s channel c o n d i t i o n were caused by the hugh impulse n o i s e s p i k e s , the time i n t e r v a l s between the e r r o r b u r s t s should be around n x 1/120 second, where n= 1 , 2 , 3 , depending whether or not s u c c e s s i v e impulse s p i k e s caused e r r o r s . In other words, the e r r o r - f r e e l e n g t h should be 160 - 18 -Burst Length D i s t 1 W 3 — - ; T - - r i BER I i 1111 - — i r—r—r-rriT x ?5 IO"4 -inoy ~~~~S. z *. IO'3 • w I 5 0 ^ 2 . 3 * IO3 L. h . • • > "^-4-. 4 * IO3 • 25 • e o -e I i f l \ i | I 1 1 1 1 1 \ I i i i i i i 11 e t 8 1 0 ie le Buret Length (bits) 10 18 1 t 1 » 1 0 IfcO 3Z» W> W« 1 0 Error-free Length (bits) 10 Fig. 2 . 8 Channel error statistics under very strong and stationary impulse noise impairment. * FSK: 1 9 , 2 0 0 b p s data rate, 120KHz carrier frequency. In-phase transmission. - 19 -- 20 -b i t s or m u l t i p l e of 160 b i t s minus the burst l e n g t h . T h i s argument i s c l e a r l y confirmed by the s t a i r c a s e - l i k e d curves shown i n F i g . 2.8, where each step jumps at the p r e d i c t e d l o c a t i o n s . Power l i n e c i r c u i t s are the f i r s t and only communication channels found to g i v e e r r o r s with such high degree of p e r i o d i c i t y . T y p i c a l r e s u l t s on the d i s t r i b u t i o n s of b u r s t e r r o r l e n g t h and e r r o r - f r e e l e n g t h f o r a c r o s s phase s i g n a l t r a n s m i s s i o n are shown i n F i g . 2.10, 2.11. The r e s u l t s given here are f o r 2 d i f f e r e n t a c r o s s phase s i g n a l t r a n s m i s s i o n paths (phase A-phase B phase A-phase C). In both cases, the t r a n s m i t t e r remained f i x e d (at phase A), and the r e c e i v e r was moved to a v o i d b i a s of the r e s u l t s by the environment of a s i n g l e r e c e i v e r s i t e . F i g . 2.10 and 2.11 show that burst e r r o r s composed of only a few b i t s predominate the t o t a l e r r o r occurrence and that the percent of s i n g l e b i t e r r o r s i n c r e a s e s as BER decreases. The above r e s u l t s i n d i c a t e that, under t y p i c a l channel c o n d i t i o n s , no s i g n i f i c a n t d i f f e r e n c e f o r e r r o r p a t t e r n behaviour i s found between i n phase and a c r o s s phase s i g n a l t r a n s m i s s i o n , at the same BER v a l u e . T h i s c o n s i s t e n c y i s apparant because impulse n o i s e , the predominant cause of e r r o r s , tends to be independent of r e c e i v e r l o c a t i o n and Error—Free Length Dist. Error-fre* Length (bits) Fig. 2 . 1 0 Representative channel error s t a t i s t i c s . FSK: 1 9 , 2 0 0 b p s data rate, 120KHz carrier frequency. Across-phase transmission. - 22 -Burst Length Dist. 100 ^ 1 i i i i j|y i i r 11 i • 1 1 1 1 1 1 • -5 75 w Be R = \ ^ ^ - 5 . q x id 4 — X a * i d 4 • • | 50 - ^"^-5.3 * I O 3 -• > • *> • 25 e 3 — 0 i i i i 1111 \ i i i i i i ' 1 • • 1 1 1 1 11 0 1 s 10 10 10 10 Burst Length (bits) Error-Free Length Dist. Error—frse Lsngth (bits) F i g . 2.11 Re p r e s e n t a t i v e channel e r r o r s t a t i s t i c s . FSK: 19,200bps data r a t e , 120KHZ c a r r i e r frequency. Across-phase t r a n s m i s s i o n . - 23 -phase. However, the performance of a c r o s s phase s i g n a l t r a n s m i s s i o n can be degraded s i g n i f i c a n t l y by s i g n a l f a d i n g impairment. F i g . 2.12 shows the r e s u l t s obtained when the channel was impaired by very severe s i g n a l f a d i n g occurred p e r i o d i c a l l y at a frequency which i s a s u b m u l t i p l e of the 60Hz power frequency. Each fade d u r a t i o n l a s t e d about 2-3ms. During the deep fade p e r i o d , long e r r o r b u r s t s o c c u r r e d . The curves i n F i g . 2.12 i n d i c a t e t h at e r r o r - f r e e lengths of approximately 3,000 b i t s predominate a l l other l e n g t h s , r e f l e c t i n g the s i t u a t i o n that e r r o r occurrence concentrated d u r i n g the p e r i o d i c a l l y happened faded p e r i o d s , l e a v i n g the non-faded p e r i o d s e r r o r f r e e . The a c t u a l frequency of occurrence of the fades can be estimated as f o l l o w s : Average e r r o r - f r e e l e n g t h =3,000 b i t s Average burst e r r o r l e n g t h < 100 b i t s Data r a t e = 19,200bps Fading frequency = Data rate/(Average e r r o r - f r e e l e n g t h + Average burst e r r o r length) = 19,200/3,100 5 6 Hz. In F i g . 2.13, e r r o r d i s t r i b u t i o n s are compared for - 24 -Burst Length Dist. 100 4> • 25 e 3 o 8 l — i l i n n a.6 x id 3 very severe fadfajj 18 _l i—I I l I 11 J l I J J.i iiJ _i l—l l I i l J-J 18 18 Burst Length ( b i t s ) 18 188 75 i-x o c • 3 I 58 -> * 25 e 8 18 Error-Free Length Dist. — i r—i i i i i i i — T 1—I 1 1 I I I] Severe •fod'fn<}> very severe fadi -^ B E R = 2.6 * 10 2.1 * I0": I.I * 10 J 1—L 18 18 Error—free Length ( b i t s ) r - = = j J I 1.1 I 1 J 1 J 1 1 1 I I 1 I I I 10 Fig. 2.12 Channel error statistics under severe periodic signal fading at subpowerline frequency. FSK: 19,200bps data rate, 120KHz carrier frequency. Across-phase transmission. - 25 -Fig. 2.13 Burst error distribution: Comparison between normal channel and fading channel. - 26 -fa d i n g and normal channels, with the average BER =10 . The percentage of v a r i o u s b u r s t e r r o r lengths i s t a b u l a t e d i n Table 2.1. The r e s u l t s c o n f i r m that there are many more long burst e r r o r s present under s i g n a l f a d i n g than when fa d i n g i s absent. BURST ERROR LENGTH (BITS) NORMAL CHANNEL (PERCENT) FADING CHANNEL (PERCENT) 1 86.0674 39.1594 2 8.5724 1 1 .7745 3 2.8352 4.3029 4 .2668 2.1348 5 .0334 1.6678 6 .0000 1.6334 7 .0334 1.9346 8 .0000 1.9346 9 .0334 1.6334 1 0 .3669 1.1341 T o t a l Percent 98. 1988 67.3115 TABLE 2.1 Burst e r r o r l e n g t h percentage: Comparison between normal and f a d i n g channel. T h i s type of time c o r r e l a t e d s i g n a l f a d i n g occurs o c c a s i o n a l l y . The fade i n t e n s i t y d i f f e r s each time. However, very severe f a d i n g i s r e l a t i v e l y uncommon and has been observed only d u r i n g a c r o s s phase s i g n a l t r a n s m i s s i o n . - 27 -A second type of p e r i o d i c s i g n a l f a d i n g , whose occurrence i s a l s o synchronous with the 60Hz power v o l t a g e , i s much more frequent. F i g . 2.14 i l l u s t r a t e s a s i g n a l s u f f e r i n g from such an impariment when t r a n s m i t t e r and r e c e i v e r were p l a c e d on d i f f e r e n t phases of the d i s t r i b u t i o n t r a n s f o r m e r . The upper t r a c e i s the 60Hz power l i n e v o l t a g e ; the lower t r a c e i s the r e c e i v e d s i g n a l envelop of a 115KHz s i n g l e frequency tone. As observed, the frequency of occurrence of such f a d i n g i s twice the power v o l t a g e frequency. Deep fades are r a r e l y found i n t h i s type of s i g n a l f a d i n g , but when present can c r e a t e long e r r o r b u r s t s . The e f f e c t of deep fades i s shown i n F i g . 2.15 where the average bu r s t e r r o r l e n g t h p l u s the average e r r o r - f r e e l e n g t h i s approximately 160 b i t s , i . e . , h a l f the power c y c l e . Again, deep f a d i n g has only been observed d u r i n g a c r o s s phase s i g n a l t r a n s m i s s i o n . In g e n e r a l , t h i s second type of s i g n a l f a d i n g i t s e l f i s seldom s t r o n g enough to be the cause of e r r o r s . However, i n many cases, i t makes the system more v u l n e r a b l e to other types of impairments, p a r t i c u l a r l y those from r e l a t i v e l y weak noise impulses. In t h i s regard, i t was found to be d i f f i c u l t - 28 -F i g . 2.14 P e r i o d i c s i g n a l f a d i n g at twice the power l i n e v o l t a g e frequency. 100 75 ->> o c • | 50 L. > 4» • 25 E 3 O 0 100 75 -o c • I 50 L. u. > • 25 B 8 0 1 0 - 29 -Burst Length Dist. • T 1 1 T ' - M 11| 1 — i — i — i \ \^\ r\ ~ r — - T - T 1 T l II • - --I BE R = / 3.5 x \6Z • - -* • • • i i I I I I I 11 • i i i i i 111 i i i i i • i i 8 0 1 10 2 10 10 Burst Length ( b i t s ) Error—Free Length Dist. M i l l 1 1 — 1 1 1 1 BE R = 3.5 x \6Z .1. I I I I I JT 1 _i I I I l 111 _1 l i I I L L 10 10 E r r o r — f r e e Length ( b i t e ) 18 Fig. 2 . 1 5 Channel error sta t i s t i c s under severe periodic signal fading at twice the power line voltage frequency. FSK: 19,200bps data rate, 120KHz carrier frequency. Across-phase transmission. - 30 -to separate impulse noise and fading effects to determine the exact cause of errors in some of the tests. Effects on error distributions due to different transmitter-receiver separation distances is shown in Fig. " 2.16. Throughout the experiments, the receiver was fixed at the same outlet while the transmitter was placed at outlets located on different floors. The error rate was fixed at 2 x 10 during the tests. The curves show great similarity. A l l measurements mentioned so far were recorded on different days and various time periods. Typical transmitter output voltages used (before the coupling network) to obtain the BER range near 10 varied from 1Vrms to 4Vrms during across phase signal transmission. During in phase signal transmittion, the transmitter output voltage used was between 0.5Vrms to 1.5Vrms. In general, more attenuation is encountered during across phase signal transmission (see chapter 6). 2.3.2 Channel Error Statistics: 4,800bps Data Rate, 120KHz  Carrier Frequency The cumulative distributions of burst error length and error-free length for typical in phase and across phase signal transmission are shown in Fig. 2.17-2.18. - 31 -Burst Length Dist. 3 Buret Length ( b i t s ) Error-free Length (bite) Fig. 2.16 Effect of transceiver separation on channel error s t a t i s t i c s . BER =2x10" . - 32 -Fig. 2.17 Representative channel error s t a t i s t i c s . FSK: 4,800bps data rate, 120KHz carrier frequency. In-phase transmission. 108 E 3 U 25 - 33 -Burst Length Dist. ~i 1 — n n r i ] n 1 1 — r i i m i 0 j i—i i i 111 • i i i i n i l i i i i i 1111 10 10 10 B u r s t L e n g t h ( b i t . ) 10 100 75 >» o c • 3 J 50 b. E 3 25 0 10 Error—Free Length Dist. ~ i — i — i i i j i i i i t 10 10 Error-free L e n g t h ( b i t e ) 10 Fig. 2 .18 Representative channel error s t a t i s t i c s . FSK: 4,800bps data rate, 120KHz carrier frequency. Across-phase transmission. - 34 -Impulse noise continues to be the primary cause of e r r o r s ; however, i t s e f f e c t i s l e s s severe because i t i s more d i f f i c u l t now f o r an impulse noise spike to o b l i t e r a t e a data b i t whose d u r a t i o n exceeds that of the noise s p i k e . Thus, the system r e q u i r e s l e s s r e c e i v e d power f o r combating impulse n o i s e . During the course of the experiments, very severe s i g n a l f a d i n g s i m i l a r to that r e p o r t e d i n s e c t i o n 2.3.1 oc c u r r e d i n one of the a c r o s s phase s i g n a l t r a n s m i s s i o n paths. The e r r o r s t a t i s t i c s o b t a i n e d under such c o n d i t i o n i s shown i n F i g . 2.19. The a c t u a l frequency of oc c u r r e n c e of the f a d i n g can be estimated to be about 6Hz, by the method s i m i l a r to that i n s e c t i o n 2.3.1. T y p i c a l t r a n s m i t t e r v o l t a g e l e v e l used at t h i s data r a t e was between lOOmVrms to 300mVrms. 2.3.3 Channel E r r o r S t a t i s t i c s ; 1,200bps Data Rate, 120KHz  C a r r i e r Frequency The cumulative d i s t r i b u t i o n s of burst e r r o r l e n g t h and e r r o r - f r e e l e n g t h f o r t y p i c a l i n phase and a c r o s s phase s i g n a l t r a n s m i s s i o n are shown i n F i g . 2.20-2.21. Although impulse noi s e remains as the major cause of e r r o r s , i t s i n f l u e n c e i s very much reduced. As a r e s u l t , at t h i s data r a t e , the system operates with much lower r e c e i v e d power as 100 • 25 E 3 U 0 10 - 35 -Burst Length Dist. T 1—r i i M I B E R = •1.2. * io5 -4.0 * 10 I -I L I I I I I I 10 - J I 1.1 11)11 1 — i—I 1 I I I 2 3 10 10 Buret Length (bite) 100 75 >» o c e | 50 u. e > ••» - 25 e 3 U 0 Error-Free Length Dist ~\ 1 — r — r i 111] 1 1 — r - r r r r r p B E R = 4.0 * lO 3 1.2 * 10 J I — L 10 r m j . i . >-y 111 i i — i — u i i . j i u 3 3 4 10 10 10 Error-free Length (bite) Fig. 2.19 Channel error statistics under severe periodic signal fading at subpowerline frequency. FSK: 4,800bps data rate, 120KHz carrier frequency. Across-phase transmission. 180 S 75 o c • | 50 t u. • > «» • 25 0 10 - 36 -Burst Length Dist. " I 1 1 T T T T T 5. I x io 3.q x io3 10 10 Burst Length ( b i t s ) j i i I I I I I I i i — I — I i 1 J I i 1 1 — i i i l i-l 1 8 18 100 1 0 Error-Free Length Dist 1 0 1 0 E r r o r - f r s s Length ( b i t e ) Fig. 2.20 Representative channel error s t a t i s t i c s . FSK: 1,200bps data rate, 120KHz carrier frequency. In-phase transmission. - 37 -Erroi—Free Length Dist. Error—froo L o n g t h ( b t t « ) Fig. 2.21 Representative channel error s t a t i s t i c s . FSK: 1,200bps data rate, 120KHz carrier frequency. Across-phase transmission. - 38 -opposed to the high r e c e i v e d power needed by h i g h data r a t e system, f o r comparable BER performance. Hence, the system becomes more e a s i l y i n f l u e n c e d by channel a t t e n u a t i o n , f a d i n g and background noi s e d i s t u r b a n c e which i s non-Gaussian. In f a c t , the background noi s e e x h i b i t s a p e r i o d i c p a t t e r n at twice the power v o l t a g e frequency (see F i g . 2.22). As a r e s u l t , there can be s l i g h t l y more burst e r r o r s at low s i g n a l l e v e l . T y p i c a l t r a n s m i t t e r v o l t a g e l e v e l s used at t h i s data r a t e were around lOOmVrms. 2.3.4 Channel E r r o r S t a t i s t i c s at Other C a r r i e r Frequencies The e f f e c t of v a r i a t i o n i n c e n t r e frequency on the e r r o r d i s t r i b u t i o n s i s shown i n F i g . 2.23, 2.24 with data r a t e set at 19,200bps. The c e n t r e f r e q u e n c i e s chosen were at 105, 110, 120, 125KHz, and the average BER was f i x e d at 1 0 3 . The r e s u l t s o b t a i n e d at the v a r i o u s f r e q u e n c i e s are very s i m i l a r . E x t e n s i v e measurements were then performed to re c o r d the channel e r r o r s t a t i s t i c s when the t r a n s m i s s i o n system's c e n t r e frequency was f i x e d at 60KHz, f o r v a r i o u s data r a t e s . T y p i c a l r e s u l t s o b tained are shown i n F i g . 2.25-2.30. The e r r o r d i s t r i b u t i o n s do not show any s i g n i f i c a n t d i f f e r e n c e when compared with r e s u l t s o b t a i n e d with c e n t r e frequency at 120KHz, at comparable BER performance. T h i s i s because the - 39 -4* » F i g . 2 . 2 2 Background power l i n e n o i s e . (Note: the high l e v e l impulse s p i k e s are not shown i n the p i c t u r e ) 180 75 >» o c 0 f 50 i. - 40 -Burst Length Dist T 1—T 1 T T T I I I I I I I I T — i — r i i r i i i 4» - 25 e a o 0 -J I I I 1 I 1 I -i i _ i _ i . - j _ i a x .j 1—i i I I I i.J 10 10 10 Burst Length (bite) 10 180 10 Error-Free Length Dist. 1 0 1 0 Error—free Length (bits) Fig. 2.23 Effect of variation of carrier frequency'on error s t a t i s t i c s . FSK: 19,200bps data rate, 105, 110, 120, 125 KHz carrier frequencies. Across-phase transmission. BER = 10"3 . 100 E 3 U 25 0 10 - 41 -Burst Length Dist. _yf I I I 11 1 1—I I 1 1 1 ! J I I I I I 111 10 10 Buret Length (bite) J I l . l l l l l 10 100 10 Error-Free Length Dist 10 10 Error—free Length (bite) Fig. 2.24 Effect of variation of carrier frequency 'on error s t a t i s t i c s . FSK: 19,200bps data rate, 105, 110, 120, 125 KHz carrier frequencies. In-phase transmission. BER = 10"3 . - 42 -Erroi—Free Length Dist. Error—frea Length ( b i t s ) Fig. 2.25 Representative channel error s t a t i s t i c s . FSK: 19,200bps data rate, 60KHz carrier frequency. In-phase transmission. - 43 -180 75 >» o c • I 50 L. k. > E 3 10 Burst Length Dist. BER = • ^ ^ ^ ~~-2.5 * I03 •/ • • • i. i 10 10 B u r e t L e n g t h ( b t t a ) 10 100 10 Error-Free Length Dist 10 ie E r r o r — f r e e L e n g t h ( b i t e ) Fig. 2.26 Representative channel error s t a t i s t i c s . FSK: 19,200bps data rate, 60KHz carrier frequency. Across-phase transmission. - 44 -Error—Free Length Dist. E r r o r - f r e e L e n g t h ( b l t a ) Fig. 2.27 Representative channel error s t a t i s t i c s . FSK: 4,800bps data rate, 60KHz carrier frequency. In-phase transmission. - 45 -Error-Free Length Dist. E r r o r — f r e e L e n g t h ( b i t s ) Fig. 2.28 Representative channel error s t a t i s t i c s . FSK: 4,800bps data rate, 6OKH2 carrier frequency. Across-phase transmission. 100 75 o c • I 50 > 4» m 1 3 25 0 1 0 - 46 -Burst Length Dist. I L U I 1 1 1 I I I I I I B E R = 4.8 x io 1.3 x l6 S 5. I x |6 5 ' i I I i I t I I I | l_ l_UX_J. l I 1—I 1 I I I U 1 S 3 1 0 1 0 1 0 B u r s t L e n g t h ( b i t s ) 1 0 0 1 0 Error-Free Length Dist 1 0 1 0 E r r o r - f r e e L e n g t h ( b i t e ) Fig. 2.29 Representative channel error s t a t i s t i c s . FSK: 1,200bps data rate, 60KHz carrier frequency. In-phase transmission. 10C 1 0 - 47 -Burst Length Dist. 1 0 1 0 Buret Length (bite) 1 0 0 >. o c e | 50 L. * 25 e 3 0 1 0 Error-Free Length Dist T 1 — r - r i I 111 T 1 — r — r - T T T 75 [. BER = 5.0 x id 3 J 1 i I I l l l l 1 0 1 0 Error—free Length (bite) 1 0 Fig. 2.30 Representative channel error s t a t i s t i c s . FSK: 1,200bps data rate, 60ZKHz carrier frequency. Across-phase transmission. - 48 -same types of error-causing disturbances were affecting the signal transmission and reception. However, the two cases required different transmitter power levels to achieve the same BER performance. This is simply because signals operating at different frequency bands will encounter different levels of attenuation and noise. At 19,200bps data rate, typical transmitter output voltage levels used varied from 1Vrms to 4.5Vrms while at 1,200bps data rate, typical voltage level used was less than 300mVrms. 2.4 Measurement Results: Residential Apartment Complex A residential apartment complex having approximately 50 individual units was chosen for measurement of error s t a t i s t i c s . In such a residential building in North America, split-single-phase power is normally supplied to circuit panels inside the building by two 110-volt supply lines and one neutral line from a centre-tapped single-phase distribution transformer. Therefore, placement of the transmitter and receiver on the same side of the neutral line w i l l herinafter be referred to as "in phase" signal transmission, while placement of the transmitter and receiver on different sides of the neutral line will hereinafter be referred to as "off phase" transmission. Measurements were - 49 -made on v a r i o u s days and time p e r i o d s to o b t a i n e r r o r s t a t i s t i c s at a data r a t e of 19,200bps and c a r r i e r frequency of 120KHZ. E r r o r s t a t i s t i c s f o r one r e p r e s e n t a t i v e measurement d u r i n g in phase s i g n a l t r a n s m i s s i o n appear i n F i g . 2.31. During the t e s t , p e r i o d i c impulse spikes were observed to cause most of the e r r o r s , while a p e r i o d i c impulse b u r s t s caused some a d d i t i o n a l longer e r r o r b u r s t s . S i m i l a r r e s u l t s were observed f o r o f f phase s i g n a l t r a n s m i s s i o n . Impulse burst noise alone was o f t e n the dominant d i s t u r b a n c e a f f e c t i n g r e c e p t i o n . R e s u l t s r e p r e s e n t a t i v e of the e r r o r behaviour e x h i b i t e d d u r i n g p e r i o d i c impulse burst d i s t u r b a n c e are shown i n F i g . 2.32, and r e s u l t s r e p r e s e n t a t i v e of the e r r o r behaviour dur i n g a p e r i o d i c impulse burst d i s t u r b a n c e are shown i n F i g . 2.33. In g e n e r a l , impulse b u r s t s c r e a t e d s l i g h t l y longer burst e r r o r s than d i d impulse s p i k e s . During the e n t i r e course of the experiments, no n o t i c e a b l e f a d i n g e f f e c t was observed; p e r i o d i c impulse s p i k e s , p e r i o d i c impulse b u r s t s and a p e r i o d i c impulse b u r s t s were the predominant causes of e r r o r s . Fig. 2.31 Channel error statistics under periodic impulse spikes plus aperiodic impulse bursts impairment. FSK: 19,200bps data rate, 120KHz carrier frequency. In-phase transmission. 180 • 25 E 3 0 1 0 - 51 -Burst Length Dist. 4.4 x 10* 1.2 x io 2.6 " Id3 6.o x |63 I I I 1 11 I I 111 1 J I 1 1 U - l 1 1—1 I M L 2 3 1 0 1 0 Burst Length (bite) 10 1 0 0 75 -o c e I 50 e > • 25 E 3 O 0 1 0 Error—Free Length Dist. 1 — i — i i i m i T"" 1 1 " i i i B E R _ 2.6 x l d 3 -1.2 * ld 3 4.4 * Id 4 -_1 I I I I I I 1 0 1 0 Error-free Length (bite) 1 0 Fig. 2.32 Channel error statistics under periodic impulse bursts impairment. FSK: 19,200bps data rate, 120KHz carrier frequency. Off-phase transmission. 103 Q 75 o c • | 50 > 4» * 25 E 3 O - 52 -Burst Length Dist. i 1—i i r i i i 1 1 1—I T T T T 2.0 * |63 6.2 x 10 0 I I I 1, 1 1 1 111 1 1 — I I I I I I I 1 1 1 I LX-Uj a l 2 3 10 10 10 10 Burst Length ( b i t s ) 100 75 U o c e 3 f 50 e > 6 3 O 25 -0 10 Erroi—Free Length Dist T 1 1 I I T T T T 1 1 r T I I T T BER = fe.£ x I O 3 J I I I I I I u 10 10 E r r o r — f r e e Length ( b i t e ) 10 Fig. 2.33 Channel error statistics under aperiodic' impulse bursts impairment. FSK: 19,200bps data rate, 120KHz carrier frequency. In-phase transmission. - 53 -3. CHANNEL ERROR STATISTICS FOR A PSK SYSTEM Measurement of error sta t i s t i c s for a PSK system operating on intrabuilding power line channels were made in the same industrial building described in chapter 2. Results were obtained for various data rates (1,200-38,400bps) at a carrier frequency of 115KHz during various days and time periods. The measuring system and procedure remained the same except that the transmitter was a PSK modulator and the receiver had the structure shown in Fig. 3.1. 3.1 The PSK Receiver General probability of error expressions for coherent PSK and di f f e r e n t i a l coherent DPSK Systems in the presence of both impulsive and Gaussian noise have been derived and analysed [13]. The PSK receiver model used in the above study is shown in Fig. 3.1. The band pass f i l t e r ' s 3-dB bandwidth was about 40KHz centered around the carrier frequency. Coherent demodulation was achieved by using a separate timing signal derived from the power line voltage to synchronize both the transmitter and receiver. Tests of the modem in the presence of additive white gaussian noise confirms proper functioning of the modem hardware. - 54 -PSK S i g n a l BPF I n t e g r a t e and Dump D e c i s i o n C i r c u i t Coherent C a r r i e r S i g n a l F i g . 3.1 PSK r e c e i v e r [13] 3.2 Measurement R e s u l t s : 38,400bps Data Rate The e r r o r d i s t r i b u t i o n f o r one r e p r e s e n t a t i v e i n phase s i g n a l t r a n s m i s s i o n i s shown in F i g . 3.2. The curves show that most of the e r r o r s (over 90%) are s i n g l e b i t e r r o r s f o r v a l u e s of BER equals to 10 or s m a l l e r . R e s u l t s f o r r e p r e s e n t a t i v e across phase s i g n a l t r a n s m i s s i o n i s shown in F i g . 3.3. Again, the m a j o r i t y of the e r r o r s are s i n g l e b i t e r r o r s . T y p i c a l t r a n s m i s s i o n v o l t a g e l e v e l s used in t h i s case v a r i e d between 0.5Vrms to 3.5Vrms. 3.3 Measurement R e s u l t s : 19,200bps Data Rate E r r o r s t a t i s t i c s f o r r e p r e s e n t a t i v e i n phase and across phase s i g n a l t r a n s m i s s i o n are shown i n F i g . 3.4, 3.5. No s i g n i f i c a n t d i f f e r e n c e i s shown between r e s u l t s obtained- at 19,200bps and 38,400bps data r a t e . P e r i o d i c impulse noise s p i k e s were again the predominant causes of e r r o r s . When - 55 -Error-Free Length D i s t . Error-free Longth (bite) F i g . 3.2 R e p r e s e n t a t i v e c h a n n e l e r r o r s t a t i s t i c s . PSK: 38,400bps d a t a r a t e , 115KHz c a r r i e r f r e q u e n c y . In-phase t r a n s m i s s i o n . - 56 -Erroi—Free Length Dist. 1 8 S 4 10 10 10 10 Error-free Length ( b i t * ) F i g . 3.3 R e p r e s e n t a t i v e c h a n n e l e r r o r s t a t i s t i c s . PSK: 38,400bps d a t a r a t e , 115KHz c a r r i e r f r e q u e n c y . A c r o s s - p h a s e t r a n s m i s s i o n . - 57 -180 75 -o c • I 50 t. lu • > 4» * 25 e 3 u Burst Length Dist. i i i i i n n n 1—r i v m i BER = 4.1 * 10 1.2 * i d 3 2.3 x |6 3 4 .6 < id 3 0 J l l l l l l L L J ~ l I 1 1 ) 1 1 J 1 I J I I I , 10 1 ! 10 10 Burst Length (bits) 10 100 10 Error-Free Length Dist. 10 10 Error—free Longth (bits) Fig. 3.4 Representative channel error s t a t i s t i c s . PSK: 19,200bps data rate, 115KHz carrier frequency. In-phase transmission. 108 75 o c S" 58 - 58 -Burst Length Dist. T 1—r~r l M i n 1 1 1 TTTTT1 B E R = 4.0 * 10* 1.2 x i d 5 2.8 x i d 3 4-.fe x i d 3 • 25 B 3 U 8 _j i i i j i J i J i i i i M u l -1 I L I M ij 18 1 2 18 18 Burst Lsngth ( b i t s ) 18 188 75 r-o c f 50 L u. - 25 L. e 3 u 0 10 Error-Free Length Dist. i 1 — r — r 10 10 Error-fres Length (bi t s ) 10 F i g . 3 . 5 R e p r e s e n t a t i v e c h a n n e l e r r o r s t a t i s t i c s . P S K : 1 9 , 2 0 0 b p s d a t a r a t e , 1 1 5 K H z c a r r i e r f r e q u e n c y . A c r o s s - p h a s e t r a n s m i s s i o n . - 59 -compared with the FSK system bearing the same average BER performance, the results consistently indicate that PSK gives a higher percentage of single bit errors and a lower percentage of double and triple bit errors. Thus, the probability of an impulse noise spike causing multiple errors appears smaller in PSK than in FSK, probably because of the 3dB advantage inherent in the PSK signalling scheme. However, more measurements are required to confirm this hypothesis. Otherwise, PSK and FSK give similar performance as both are vulnerable to the harmful effects of impulse noise. Impulse noise is commonly found in a number of channels, including operational multichannel satellite systems and ter r e s t r i a l microwave systems. The general performance of systems in the presence of impulsive noise alone, or combinations of impulsive, and Gaussian noise has been analysed by others [13-20], and general probability of error expressions for various modulation schemes due to these disturbances are available. In particular, the error probability performance of PSK and DPSK systems was compared in [13] and was found to exhibit l i t t l e difference. It was concluded [13] that impulse noise, at the instant when i t enters the receiver, causes an error with high probability - 60 -independent of the signalling scheme. Furthermore, for impulse noise limited channels, the error probability verses the received signal-to-noise ratio curve does not follow the exponential shape exhibited under Gaussian noise disturbance. Indeed, the curve begins to bottom out in the high signal-to-noise ratio region. The resulting slow drop of BER is very serious when the data bit duration is comparable to the impulse spike duration, and dictates that appropriate error control be used in high speed power line communications. (Details regarding error control appear in chapter 5). 3.4 Measurement Results: 4,800bps and 1,200bps Data Rate Representative results obtained at these data rates are shown in Fig. 3.6 and 3.7. The results are consistent with those obtained in chapter 2 using the FSK modem. 100 75 -o c • | 50 L. u. • » • 25 E 3 O 0 10 - 61 -Burst Length Dist T 1 1 I I I I I 4.1 * 10 J 1 I I I I I 11 10 -I I I 1 I l 1X1 J 1 l . U U L I 2 3 10 18 Buret Length (bite) 180 75 o c e 3 8* 50 -k. e > 4 * * 25 E 3 U 0 10 Error-Free Length Otst. T 1 1 I I I I I I BER = J— i J .1 III. 10 10 Error—free Length (bite) 10 Fig. 3.6 Representative channel error s t a t i s t i c s . PSK: 4,800bps data rate, 115KHz carrier frequency. Across-phase transmission. 180 25 E a o - 62 -Burst Length Dist 9.6 * IO4 3.0 * IO3 T 1 1 I T T T T 1 8 J I J I I I 11 J I I I I .1 J I—I J U I. 18 i a 18 18 Burst Length (bits) 18 188 75 -x o c • ? 58 • 25 E 8 8 18 Error—Free Length Dist. T 1—I T BER = 3.0 x q.e> x io 10 J I I L 18 18 Error-frss Lsngth (bits) 18 Fig. 3.7 Representative channel error s t a t i s t i c s . PSK: 1,200bps data rate, 115KHz carrier frequency. Across-phase transmission. - 63 -4. CHANNEL ERROR STATISTICS FOR SPREAD SPECTRUM SYSTEMS Spread spectrum signalling provides a means of information transmission in which the information signal uses a bandwidth larger than the minimum necessary to transmit the information [ 2 1 ] . The spectrum spreading process is achieved by using a pseudorandom code which is independent of the data. At the receiver, the same code sequence is used to despread the signal back to its original bandwidth. There are various ways that the code can be used for spectrum spreading. Two common techniques are direct-sequence (DS) and frequency-hopping (FH). In a DS system, the information signal is multiplied with the pseudonoise code to generate a transmitted signal whose bandwidth is then determined by the code rate. The factor by which the bandwidth is spread or the "processing gain" is equal to Rc/Rd where Rc and Rd denote the code rate and data rate, respectively. In a FH system, the code is used to shift the transmission frequency in a pseudorandom manner. If there are N frequency slots available, and one frequency slot is used every T seconds, then the total bandwidth of the system is approximately equal to N/T [ 2 1 ] . Advantages of spread spectrum (SSMA) signalling include - 64 -resistance to narrow band impartments, multiple user random access capability, selective addressing capability and immunity to casual eavesdropping. Primary disadvantages include code acquisition delay, as well as increased hardware complexity and implementation costs. Measurements of channel error statistics were made using a DS/SSMA-FSK system and a DS/SSMA-PSK system. The tests were conducted during various days and time periods in the same industrial building described in chapter 2. Results were obtained at various data rates, code rates and carrier frequencies. The measuring system and procedure remained the same as before except that DS/SSMA-FSK and DS/SSMA-PSK modems were used. 4.1 Direct-Sequence SSMA Frequency-Shift-Keyed (DS/SSMA-FSK)  System 4.1.1 The Transmitter and Receiver The transmitter and receiver block diagrams for the DS/SSMA-FSK system are shown in Fig. 4.1. - 65 -Data F>0 Code DS/SSMA-FSK Signal — > FSK Receiver FSK Modulator -> DS/SSMA-FSK Signal (a) r-> Integrator and Dump Decision Ci rcuit Code (b) Fig. 4.1 (a) DS/SSMA-FSK transmitter, (b) DS/SSMA-FSK receiver The transmitter consists of a pseudorandom code sequence generator and the FSK modulator used in chapter 2. The data message is f i r s t multiplied by the code sequence and the resulting data modulated code sequence is then used as the input signal to the FSK modulator. The received signal is f i r s t demodulated by the FSK demodulator to recover the data modulated code sequence which is then correlated with the locally generated code sequence. A majority decision using an integrate and dump f i l t e r is then employed to extract the data. Code synchronization is - 66 -done by using standard "search and lock" process employing sliding correlator and tracking [24,28]. 4.1.2 Channel Error Statistics: 4,800bps Data Rate, 19,200bps Code Rate, 120KHz Carrier Frequency Representative error statistics for 4,800bps data rate transmission are presented in Fig. 4.2 and 4.3. The 4,800bps system represents a 3 out of 4 majority decision at the receiver. For the same value of BER, the measured error statistics are similar to the FSK results in chapter 2. Impulsive noise spikes caused many single bit errors. Typical transmitter voltage levels used in this case varied between .4Vrms to 1.5Vrms. 4.1.3 Channel Error Statistics: 1,200bps Data Rate, 19,200bps Code Rate, 120KHz Carrier Frequency Representative error statistics for 1,200bps data rate system are presented in Fig. 4.4, 4.5. The 1,200bps system represents a 9 out of 16 majority decision at the receiver. Again the measured results do not indicate any significant difference when compared with FSK results in - 67 -Error-Free Length Dist. E r r o r - f r e e L e n g t h ( b i t * ) Fig. 4.2 Representative channel error st a t i s t i c s . DS/SSMA-FSK: 4,800bps data rate, 19,200bps code rate, 120KHz carrier frequency. In-phase transmission. - 68 -Fig. 4.3 Representative channel error st a t i s t i c s . DS/SSMA-FSK: 4,800bps data rate, 19,200bps code rate, 120KHz carrier frequency. Across-phase transmission. - 69 -Fig. 4.4 Representative channel error stat i s t i c s . DS/SSMA-FSK: 1,200bps data rate, 19,200bps code rate, 120KHz carrier frequency. In-phase transmission. 100 75 o c • 3 «»• 50 e L. 8 > - 25 e 3 u 0 100 75 >. o c 0 P- 50 e k. a > - 25 e 3 0 10 - 70 -Burst Length Dist. — r - i i i i i n-j- —• i I k s i r I I i 111 -• BER = • • »^/^l.4 * 5 .0 x io3 .6* id5 -• - --• -- -• J i i t i 1111 i 1 1 i i i i 11 i i i i i 11 a e 0 l 10 2 10 10 Buret Length (bite) Erroi—Free Length Dist. -i 1 — r r r i n i 1 1 — n m BER = i i i 11 j i t 1 I L I I I I I I J- I I I Ll 10 10 Error-free Length (bite) 10 Fig. 4.5 Representative channel error st a t i s t i c s . DS/SSMA-FSK: 1,200bps data rate, 19,200bps code rate, 120KHz carrier frequency. Across-phase transmission. - 71 -chapter 2. Typical transmitter voltage level used was less than 300mVrms. 4.1.4 Channel Error Statistics at 60KHz Carrier Frequency Extensive measurements were also made to record error st a t i s t i c s using 60KHz as the center frequency at 4,800bps and 1,200bps data rate. The results obtained again (Fig. 4.6-4.7) indicate that error pattern behaviour is insensitive to change in operating frequency. 4.2 Direct-Sequence SSMA Phase-Shift-Keyed (DS/SSMA-PSK)  System 4.2.1 The Transmitter and Receiver The transmitter and receiver block diagrams for the DS/SSMA-PSK system appear in Fig. 4.8. The transmitter consists of a pseudorandom code sequence generator and the PSK modulator used in Chapter 3. The data message is f i r s t multiplied by the code sequence and resulting data modulated code sequence is then used as the PSK modulator input signal. - 72 -Error-Free Length Dist. E r r o r - f r e e L e n g t h ( b i t e ) Fig. 4.6 Representative channel error st a t i s t i c s . DS/SSMA-FSK: 4,800bps data rate, 19,200bps code rate, 60KHz carrier frequency. Across-phase transmission. - 73 -Burst Length Dist. Buret Length (bite) Error-Free Length Dist. Error-free Length (bite) Fig. 4.7 Representative channel error statistics." DS/SSMA-FSK: 1,200bps data rate, 19,200bps code rate, 60KHz carrier ,frequency. Across-phase transmission. - 74 -Data PSK Modulator -> DS/SSMA-PSK Signal Code (a) DS/SSMA-PSK Signal BPF Integrate — > Dec i sion and Dump Circuit Code Carrier (b) Fig. 4.8 (a) DS/SSMA-PSK transmitter, (b) DS/SSMA-PSK receiver The received signal is correlated with the locally generated code sequence and carrier signal, followed by an integrate and dump f i l t e r to extract the data. A separate timing signal derived from the power line voltage is used to synchronize the operation of the transmitter and receiver. 4.2.2 Channel Error Statistics: 38,400bps Code Rate,  115KHz Carrier Frequency Representative error statistics for 19,200bps and 4,800bps data transmission are presented in Fig. 4.9, 4.10 - 75 -Error—Free Length Dist. Error-froo Length ( b i t e ) Fig. 4.9 Representative channel error st a t i s t i c s . DS/SSMA-PSK: 19,200bps data rate, 38,400bps code rate, 115KHz carrier frequency. Across-phase transmission. - 76 -75 >. o c • I 50 > • 25 c 3 0 10 Burst Length Dist. • i—i—i—1111 i 1 1 > i i i i i i—i—*_ 1 2 10 10 B u r s t L e n g t h ( b i t s ) 10 100 >» o c s 3 O* S 75 -50 • » • 25 £ 3 0 10 Error-Free Length Dist BER = 2.4 * lo 3 10 10 E r r o r - f r e e L e n g t h ( b i t s ) 10 Fig. 4.10 Representative channel error stat i s t i c s . DS/SSMA-PSK: 4,800bps data rate, 38,400bps code rate, 115KHz carrier frequency. Across-phase transmission. - 77 -for which impulse noise was the major cause of errors. Representative error stat i s t i c s for 1,200bps data rate are presented in Fig. 4.11. The results do not indicate any significant difference when compared with the PSK results in chapter 3. tee 75 _ o c J 58 U. > 6 8 25 a 18 - 78 -Burst Length Dist. 1 1-1 I I IT T 1—I I I I I I 8.5 * 10* 5.0 x |6S J I I I II.. 1 1 I I I I I 11 1 I I I I I I 11 1 2 3 18 18 Burst Length (bite) 18 188 18 Error-Free Length Dist T 1—r 18 18 Error—free Length (bite) Fig. 4.11 Representative channel error st a t i s t i c s . DS/SSMA-PSK: 1,200bps data rate, 38,400bps code rate, 115KHz carrier frequency. Across-phase transmission. - 79 -5. IMPLICATIONS OF THE TEST RESULTS" 5.1 Forward Error Correction (FEC) Coding Under typical channel conditions the measured error s t a t i s t i c s indicate that short error bursts comprised a large - 3 percentage of the total errors when BER = 10 . If FEC coding is employed for error control, a code capable of correcting several random errors or a short error burst would combat a high percentage of a l l errors encountered. Although burst error correcting codes are less powerful than random error correcting codes, their use may be favoured due to simpler and less expensive decoding (A low-cost modem is desirable). As well, burst error correcting codes have high code rates. As an example, a class of cyclic burst error correcting codes called Fire code [5] can be decoded relatively easy using shift-register circuits. A (105, 94) Fire code which is capable of correcting any single burst of four or less errors has a code rate equal to 94/105 = 0.9. By properly choosing a suitable code, FEC can significantly improve system performance. However, there w i l l remain some uncorrectable error patterns for handling by other techniques when very high r e l i a b i l i t y is required. - 80 -5.2 Automatic-Repeat-Request (ARQ) Strategies In many data communications systems, error detection and retransmission schemes are used for error control because these methods are easy to implement and can provide high system r e l i a b i l i t y . Such schemes are particularly effective when the average channel bit error rate is small. However, the throughput of an automatic-repeat-request (ARQ) error-control system deteriorates quickly as the channel error rate increases. Fig. 5.1 shows the throughput efficiencies of the ideal selective-repeat ARQ (which does not depend on the round-trip delay) and the go-back-N ARQ schemes for a satellite channel with a data rate of 1.5Mbps and a round-trip delay of 700ms. One can see that even for - 4 the ideal selective-repeat ARQ, a channel error rate of 10 is needed to provide acceptable throughput performance. For high speed data transmission in an impulsive noise environment, system performance can eventually reach an asymtote even though the received SNR continues to increase. Thus, a continual increasing of the transmitter power to reduce BER may not be viable. For example, up to a few watts of transmitter power was needed by the FSK system just to _ 3 obtain a 10 BER performance when i t operated in the - 81 -Fig. 5.1 Throughput efficiencies: the ideal selective-repeat ARQ with infinite buffer and the go-back-N ARQ with code block length n = 2024 [23]. - 82 -industrial building with a data rate of 19,200bps. As a result of the high BER encountered, error-control schemes employing conventional ARQ strategies are not suitable for high speed data transmission, due to the frequent retransmissions of codewords detected in error. This is particularly true if the maximum permissible transmitter power is limited in order to reduce costs or to meet regulatory limitations, such that the system cannot guarantee a small enough channel error rate to maintain a high throughput with ARQ. At lower data rates (1,200bps or below), impulse noise effects are small, and-the system performance follows more closely with the theoretically predicted AWGN result. In addition, a lower data rate means lower transmitted and received signal power for the same BER performance. For example, when, the data rate is reduced from 19,200bps to 1,200bps, a reduction of 12dB of received signal power is achieved. At low data rates i t becomes practical to use increased transmitter power levels to reduce BER so that simple error control such as conventional ARQ becomes feasible. - 83 -5.3 ARQ with FEC Coding ARQ and FEC schemes can be combined to form a class of error-control strategies commonly referred to as hybrid ARQ [23]. In a straight forward hybrid ARQ system, an error pattern within the correcting power of the designed code will be corrected as with ordinary FEC. However, if the detected error pattern is outside the correcting power of the designed code, the received codeword is rejected and a retransmission is requested, as under ordinary ARQ. This whole process is repeated until a codeword is either interpreted as error-free or successfully corrected. As a result, for a channel with a f a i r l y high BER a hybrid ARQ scheme can increase system throughput above that for an ARQ system alone while increasing system r e l i a b i l i t y above that of a FEC system alone. From the above investigation, a proper combination of ARQ and FEC seems to be the most suitable error-control strategy for power line channels. As majority of the error patterns involve short burst errors which can be corrected relatively easily and e f f i c i e n t l y . The remaining uncorrectable long error bursts would be handled by retransmission of the errored block. - 84 -5.4 Bit Interleaving As noted earlier, power line channels do occasionally suffer from severe signal fading. If the fading is deep enough, a large number of data bits will be in error resulting in very long error bursts. To counteract this infrequent but existing situation, bit interleaving would be very useful. Bit interleaving combined with hybrid ARQ can provide a very reliable power line data communications system. However, i t should be noted that with bit interleaving, a message is essentially received only when a l l interleaved codewords have been received [22]. From the results of the error statistics measurements in chapter 4, bit interleaving should be able to improve the system performance of the 3 out of 4 majority decision DS/SSMA-FSK system. Because the errors occurred in small bursts, a small degree of interleaving would be adequate to effectively disperse the error bits to achieve better performance. Actual implementation of a short interleaver 31 bits long confirmed the above ideas [24]. 5.5 Summary Based on the measurement results and observations, error - 85 -c o r r e c t i o n combined with r e t r a n s m i s s i o n i s suggested f o r e r r o r c o n t r o l in high speed data communications systems o p e r a t i n g on power l i n e channels. The e r r o r c o r r e c t i n g c a p a b i l i t y of the code would be designed f o r raw channel BER - 3 = 1 0 . However, the system should be designed to operate with a b e t t e r BER performance because of the margin needed to compensate f o r the medium to long term channel BER f l u c t u a t i o n s . For p r o t e c t i o n a g a i n s t ( r e l a t i v e l y uncommon) severe s i g n a l fades, b i t i n t e r l e a v i n g c o u l d be used to p r o v i d e a d d i t i o n a l p r o t e c t i o n without i n t r o d u c i n g more - 3 redundancy. The 10 BER value i s chosen because b e t t e r BER performance i s more d i f f i c u l t to guarantee, while worse BER performance i s not s u f f i c i e n t f o r m a i n t a i n i n g high q u a l i t y s e r v i c e f o r many a p p l i c a t i o n s . - 5 At lower data r a t e s , BER performance b e t t e r than 10 can be guaranteed. With t h i s BER value simple ARQ e r r o r c o n t r o l i s adequate f o r h i g h r e l i a b i l i t y , while more complex schemes such as ARQ with FEC can provide e x t r a high r e l i a b i l i t y . A good p r a c t i c a l example i s given by the Nonwire powerline module [25] which supports 1,200bps data r a t e t r a n s m i s s i o n f o r i n phase t r a n s m i s s i o n (For communications among d i f f e r e n t phases, a s p e c i a l i z e d modem having 3 separate power l i n e c o u p l e r s f o r sending out s i g n a l - 86 -simultaneously on a l l three phases i s a p p a r e n t l y used). Nonwire's s p e c i f i c a t i o n s t a t e s : " E r r o r Rate: B e t t e r than 10 to the minus 5 s i n g l e b i t e r r o r r a t e " without any e r r o r h a n d l i n g . However, with e r r o r h a n d l i n g , the s p e c i f i c a t i o n says: "One undetected data e r r o r per 2 ye a r s continuous o p e r a t i o n , guaranteed by b u i l t - i n - s o f t w a r e e r r o r d e t e c t i o n and c o r r e c t i o n l o g i c . " The e r r o r c o n t r o l used by the Nonwire product i s s i n g l e e r r o r c o r r e c t i n g Hamming code plus a checksum f o r e r r o r d e t e c t i o n and r e t r a n s m i s s i o n . The Nonwire modem operates with a very high 8 v o l t peak-to-peak output s i g n a l l e v e l at 1,200bps, and power l i n e modems having higher data r a t e are c u r r e n t l y u n a v a i l a b l e . Regardless of the e r r o r - c o n t r o l scheme used, the choice of a proper e r r o r c o n t r o l code i s c r i t i c a l , e s p e c i a l l y when very good raw BER performance cannot be guaranteed. A proper code can provide good throughput as w e l l as r e l i a b l e performance. The r e s u l t s o b tained from the measured channel e r r o r s t a t i s t i c s enable i n t e l l i g e n t c h o i c e s to be made f o r e r r o r c o n t r o l codes and p r o t o c o l s . The f i n a l d e c i s i o n remains on the hands of the users, f o r s p e c i f i c a p p l i c a t i o n s . - 87 -6. TRANSMISSION. CHARACTERISTICS OF INTRABUILDING POWER LINE CHANNELS Unlike many other communication channels which have a relatively flat signal transmission spectrum, the transmission characteristics of intrabuilding power line channels can be highly frequency dependent. A small change in the operating frequency band can result in a large change in received signal strength. A good understanding of the signal spectrum is needed for the efficient system design, to enable least transmitter power for a given BER performance. Attenuation verses frequency measurements were made on five buildings, including industrial and residential buildings. Most measurements were performed during the time periods of heavy ele c t r i c a l loading, although some measurements were also made during other time periods of a day for comparison purposes. The effects of some typical consumer loads on signal transmission were investigated. Finally, some comments are made regarding transmission distance within a building's power distribution network. - 88 -6.1 The Measurement System A l l measurements were performed using the experimental setup shown in Fig. 6.1. Signal Generator Power Ampli f ier Coupling Network TRUE RMS Voltmeter Load Coupling Network Power1ine Channel Fig. 6.1 Experimental setup for attenuation measurement. A single frequency sinusoidal tone is generated and then amplified by the power amplifier and coupled onto the power line through the coupling network. The coupling network is a passive high pass f i l t e r having a 3dB cutoff frequency between 20-30KHz. At the receiver, a 1Oohm resistor is used as the load. For a given transmitter voltage, the attenuation is given by - 89 -2 2 V rec - V noise Attenuation (dB) = 10 log 10 where Vtr is the transmitted voltage measured at the power amplifer output Vrec is the received voltage measured across the lOohm load, with transmitter on Vnoise is the received voltage measured across the 1Oohm load, with transmitter off. A l l voltage readings were taken using a true RMS voltmeter. Measurements were made at 16.discrete frequencies in the 20KHz to 240KHz range. 6.2 Attenuation Characteristics; Industrial Building The four storey high industrial building used in the error statistics measurements was used again for studying the signal transmission characteristics of intrabuilding power line channels. Fig. 6.2 shows the attenuation verses frequency characteristics obtained during in phase signal transmission, where the transmitter and receiver were separated by 20-30 feet of the power distribution network inside a room with no specific loadings. Fig. 6.2 shows l i t t l e variation in signal strength over the frequency band. Although this result is too ideal for any r e a l - l i f e situation, i t does provide a comparison for other measurements. - 90 -CHRNNEL RTTENURTION vs FREQUENCY -I8f— c 0-38f •» 0 3 C •f» •» oc " B •—ft tc A A 4« alt te A A A t«e lie i Z B IU i l Frequency (KHz) 148 138 188 178 188 1S8 288 212 228 238 248 Fig. 6.2 Attenuation vs. frequency characteristics: industrial building. In-phase signal transmission through 20-30 feet of power distribution wiring. - 91 -Fig. 6.3 shows the result obtained during in phase signal transmission, where the transmitter and receiver were separated by an unknown length of transmission path. One can see that the channel remains relatively flat below 140KHz, but then starts to r o l l off very quickly. Fig. 6.4 shows the results obtained during two across phase signal transmissions. The two curves were obtained with the receiver fixed at one phase, and the transmitter placed at the other two phases different from that of the receiver. Both curves indicate that channel r o l l off starts at approximately 30KHz. The above measurements were performed during the day between 9 a.m. to 5 p.m.. Comparative studies were done during night time periods (after 7 p.m.). Fig. 6.5 shows the result obtained during in phase signal transmission. In general, the figure is similar to Fig. 6.3 in that rapid channel ro l l - o f f begins at around the same frequency range (140-160KHz). The results for across phase signal transmission are shown in Fig. 6.6. Both curves show continuous increase in attenuation as frequency increases. In addition, the curves indicate serious narrow band drop-outs at around 80KHz and 160KHz. From these results obtained, one sees that the average - 92 -Fig. 6.3 Attenuation vs. frequency characteristics: industrial building. In-phase signal transmission through an unknown path at day t ime. - 93 -Fig. 6.4 Attenuation vs. frequency characteristics: industrial building. Across-phase signal transmission through unknown paths at day t ime. - 94 -Fig. 6.5 Attenuation vs. frequency characteristics: industrial building. In-phase signal transmission through an unknown path at night time. - 95 -CHANNEL RTTENURTION vs FREQUENCY j Frequency (KHz) Fig. 6.6 Attenuation vs. frequency characteristics: industrial building. Across-phase signal transmission through unknown paths at night time. - 96 -level of attenuation for in phase signal transmission is less than that for across phase signal transmission, especially at high frequencies. Transformers cause strong signal attenuation from one transformer phase to another. As well across phase signal transmission also seems to suffer more easily from narrow band drop-out. Narrow band drop-outs are usually caused by reactive loads, standing waves, reflections and multipath propagation effects on the line [2,4]. 6.3 Attenuation Characteristics: Residential Environment Two apartment complexes, located in two separate residential areas, were used to measure attenuation verses frequency characteristics of intrabuilding power line channels. . One building has four floors with approximately 60 individual units; the other is a 12-storey building with more than one hundred individual units. Split-single-phase power is delivered to each building. Fig. 6.7 shows the attenuation characteristic measured in the low-rise during in phase and off phase signal transmission. Both curves indicate a better transmission medium than in the industrial building. The results presented were recorded from 6-8 p.m.. - 97 -Fig. 6.7 Attenuation vs. frequency characteristics:. residential complex (low-rise). In-phase and off-phase signal transmission through unknown paths. - 98 -Measurement results obtained in the high-rise building appear in Fig. 6.8 and 6.9. Fig. 6.8 shows the result obtained during in phase signal transmission where the transmitter and receiver were separated by a short transmission path. The measured frequency spectrum is rather f l a t . Fig. 6.9 shows the result obtained for in phase signal transmission and off phase signal transmission, where the transmitter and receiver were separated by an unknown length of transmission path. These curves indicate that signal attenuation increases rapidly as frequency increases. Again, a l l results presented were recorded during the 6-8 p.m. time period. 6.4 Attenuation Characteristics: Hospital A hospital located in downtown Vancouver, B.C. was used to measure the attenuation verses frequency characteristics. Three phase power is supplied directly to the building. Fig. 6.10 shows the result obtained during in phase signal transmission. The exact signal path between the transceiver was unknown. The curve indicates the beginning of increasing channel r o l l off at around 140KHz, consistent with what was observed under comparable environment as illustrated in Fig. 6.3. - 99 -Fig. 6.8 Attenuation vs. frequency characteristics: residential complex (high-rise). In-phase signal transmission' through a short path. - 100 -Fig. 6.9 Attenuation vs. frequency characteristics: residential complex (high-rise). In-phase signal transmssion through unknown paths. - 101 -CHRNNEL RTTENURTION vs FREQUENCY m lit 1*0 iii 140 158 ii Frequency (KHz) Fig. 6.10 Attenuation vs. frequency characteristics: hospital. In-phase signal transmission through an unknown path. - 102 -Figure 6.11 shows the ' attenuation characteristics obtained for across phase signal transmission. The receiver was fixed at one phase, the transmitter was placed at the other 2 phases different from the one that the receiver was on. One signal path shows greater attenuation than the other, especially at frequencies above lOOKHz. 6.5 Effects of External Loading On Signal Transmission Observations made during the course of previous measurements indicated that the channel transmission characteristics depended on what external loads switched onto the line. This observation motivated a study to determine the effect of some common household appliances on signal transmission. The measurements were conducted in a single family house under controlled conditions. The house is a typical single family residence, where in North, America general purpose branch circuits carry 110-volt power to switched lights and plug-in recepticles, and special circuits carry 220-volt power to heavy appliances including clothes dryers, electric ranges and water heater. The attenuation verses frequency characteristic was f i r s t determined without specific loads being energized. Thereafter, individual - 103 -Fig. 6.11 Attenuation vs. frequency characteristics: hospital. Across-phase signal transmission through unknown paths. - 104 -consumer loads were placed on the same phase as the receiver then energized. Fig. 6.12 shows the results obtained for in phase and off phase signal transmission without any specific loading. Then two discrete frequency tones at 50 and 100KHz were used as test signals to identify which consumer load would influence signal transmission. A radio, cassette recorder, vaccum cleaner, razor, sewing machine, fan and flourescent lamp were found to have no effect on the level of signal strength received during in phase signal transmission test. An electric range and a clothes dryer were found to slightly attenuate the received signal level, while a 3KW water heater and TV receiver were found to attenuate the received signal level more significantly. The effects of the water heater and TV receiver on signal transmission are presented in Fig. 6.13. For the television receiver, i t was found that faster r o l l off of high frequencies occurred. For the water heater, a constant drop of received signal level was observed over the entire frequency band measured. The tests were repeated for off phase signal transmission. Results similar to those above were obtained - 105 -CHRNNEL RTTENURTION vs FREQUENCY -MF-"D c 0-38J m 3 c 4> A A A A *« A A A A A MB ik lie - i U L J l 188 118 188 198 148 198 168 17B 118 188 288 218 22* 232 240 Frequency (KHz) Fig. 6.12 Attenuation vs. frequency characteristics:. single family home. In-phase and off-phase signal transmission without specific loading switched on. - 106 -Fig. 6.13 Attenuation vs. frequency characteristics: In-phase signal transmission with water heater and TV receiver switched on. - 107 -with two exceptions given by the electric range and clothes dryer. These two appliances improved rather than degraded signal transmission so that signal reception by the receiver was enhanced as shown in Fig. 6.14. These appliances connect across two phases (240 volts) of the supply c i r c u i t , and provide a direct path for signal transmission from one phase to another without passing through the lossy transformer secondary. This result definitely suggests the usefulness of installing high frequency bypass capacitors across the different phases (legs) of the distribution transformer to enhance signal transmission between different phases. Similarly, high frequency bypass capacitors installed to connect transformer's primary and secondary would improve signal transmission through distribution transformers for power line carrier applications; without bypass capacitors primary-to-secondary transmission is feasible only below 1OKHz. 6.6 Comments on Transmission Distance A common misconception is that two el e c t r i c a l outlets physically close to each other are separated by a short (good) transmission path. However, this is not always true. If the exact wiring plan of a building is not available, one - 108 -CHANNEL RTTENUATION vs FREQUENCY cn C 0-3BE-•» 0 3 C XL Electric range V A A te 4* A A 7t A A life ill i i i lit 14» lit is* i * * tie ii* iii in & lit lfe iii is. iu 14* ist if Frequency (KHz) Fig. 6.14 Attenuation vs. frequency characteristics:-single family home. Off-phase signal transmission with electric range and clothes dryer switched on. - 109 -is generally unable to determine the transmission distance between two electrical outlets. A transmitter and receiver operating on the same floor may be separated by a longer electrical path than when they are located at different floors. 6.7 Summary The series of test measurements provide improved understanding of signal transmission through power line channels. The measurement data indicates some portions of the frequency spectrum possess better transmission characteristics than others. However, i t should be noted that the choice of a good frequency band is a compromise based on both the transmission characteristics and noise levels, and not either characteristic alone. The transmission characteristics are strongly affected by the loading on the transformer's secondary, and the measurements performed are not intended to cover a l l possible situations. The results do illustrate typical behaviour exhibited by the transmission medium. When combined with the noise characteristics such data is useful as a f i r s t step for choosing signal transmission frequencies. - 1 1 0 -In addition, the measurement results suggest that high frequency bypass networks can provide a simple way to reduce the heavy transmission losses between phases. - 1 1 1 -7. CONCLUSION 7.1 Concluding Remarks The primary purpose of this thesis is to present for the f i r s t time error patterns statistics of intrabuilding power line communication channels. These sta t i s t i c s are based on actual measurements on a number of real channels. The results w i l l be very useful for choosing efficient and reliable error control codes and. protocols which are essential for reliable communications, and for predicting performance. The emphasis is on high speed (19,200bps) data transmission where channel bit error rate performance is such that error control coding is essential. As well, high speed communications is desirable for an increasing number of applications. When a large number of users have to share the channel, a high speed channel increases the throughput of various channel multiple access schemes including CSMA and CSMA-CD which are commonly used in packet switched data communications. In these schemes the packet collision probability decreases (assumming a fixed number of bits/packet) as the data rate increases, provided channel errors are not too severe. These multiple-access schemes are - 1 1 2 -very suitable for short delay channels such as the power line circuits being investigated in this thesis. The results of the error pattern s t a t i s t i c a l analysis are summarized below: (1) For normally conditioned channels, impulsive noise is the limiting disturbance in high speed transmission, when the data bit duration approaches the duration of the impulsive disturbance. Depending on i t s strength and width, an impulse spike or several impulse spikes close together can cause from no error to several errors. (2) As BER is decreased, the trend is to shorter error bursts, and conversely. (3) Errors from impulse noise impairment are mostly periodic, since most impulse spikes themselves occur periodically according to the periodic 60Hz power line voltage. (4) The error pattern behaviour of power line channels is found to be relatively insentitive to modulation schemes and operating frequency. This result occurs because impulse noise is the primary cause of - 1 1 3 -errors. (5) Fading is also a cause of errors in power line channels. A deep fade can cause error bursts longer than 100 bits. A weak fade allows weaker impulses which are not normally strong enough to cause an error, to become the cause of an error. Power line fading is periodic with frequency at twice the power voltage frequency or its submultiples. (6) Based on the measurement results, ARQ with FEC is suggested for error control at high data rates. Burst error correcting codes can be used to reduce decoder cost and complexity, with some sacrifices in performance. Finally, bit-interleaving is proposed for additional protection against momentary decrease in system performance. (7) At lower data rates (1,200bps or below), impulse noise effects are relatively small. It becomes feasible at these data rates to use more transmitter power to guarantee a better BER performance, and error control becomes relatively easy. (8) Reliable data"communication on intrabuilding electric power lines appears viable, given proper - 1 1 4 -error control design, even for high speed transmi ssion. The attenuation characteristics measurements performed on various power line channels lead to the following conclusions: (1) Average attenuation is generally large, typically in excess of 20dB (2) Attenuation is frequency selective (3) Usable bandwidth is limited which also implies that frequency division multiplexing is often impract ical than across phase signal transmission (5) Attenuation can be very dependent on electrical load profile which is time varing (6) High frequency bypass networks can significantly improve signal transmission involving different phases. (4) Average attenuation for in phase In summary, this thesis enables a much improved understanding of the behaviour and use of intrabuilding power distribution circuits for communications. - 1 1 5 -7.2 Suggestions for Future Work Based on the measurement results, models could be developed to characterize the power line channel. From these models, one should be able to derive the error statistics having good agreement with experimental results. However, i t wi l l probably be extremely d i f f i c u l t to construct a model which is detailed enough to have general validity. Other useful work includes tests with actual FEC/ARQ error control strategies at high data rates, tests with simple error control schemes at lower data rates, and examination of ways to reduce losses for across phase transmission. - 1 1 6 -REFERENCES [I] P.K. Van Der Gracht and R.W. Donaldson, "Communication using pseudonoise modulation on electric power distribution circuits", IEEE Trans. Communications, Sept. 1985. [2] H. Oschner, "Data transmission on low voltage power distribution lines uisng spread spectrum techniques", Proc. Can. Commun. and Pwr. Conf., Montreal, Quebec, Oct. 15-17, 1980, pp. 236-239. [3] R.M. Vines, H.J. Trussell, L.J. Gale, and J.B. O'Neil, Jr., "Noise on residential power distribution ci r c u i t s " , IEEE Trans. Electromag. Compat., vol. EMC-26, pp. 161-168, Nov. 1984. [4] R.M. Vines, H.J. Trussell, K.C. Shuey and J.B. O'Neil, Jr., "Impedance of the residential power-distribution c i r c u i t " , IEEE Trans. Electromag. Compat., vol. EMC-27, pp. 6-12, Feb. 1985. [5] W.W. Peterson and E.J. Weldon, Error-Correcting Codes, Cambridge, MA: M.I.T. Press, 1972. [6] K. Brayer, "Error patterns measured on transequatorial HF communication links", IEEE Trans. Communications, Apr. 1968. [7] R.C. French, "Mobile radio data transmission in the urban environment", Conf. Rec. of the International Conference on Communications, Philadelphia, PA, June 14-19, 1976, pp. 27.15-27.20. [8] H. Omori and K. Otani, "Burst error characteristics of d i g i t a l land mobile radio", Int. Conf. Commun., Conf. Rec, vol. 1, 24-2, June 1980. [9] K. Otani, K. Daikoku and H. Omori, "Burst error performance encountered in digital land mobile radio channel", IEEE Tran. Vehicular Tech., Nov. 1981. [10] A.A. Smith, "Power Line Noise Survey", IEEE Trans. Electromag. Compat., vol. EMC-14, pp. 31-32, Feb. 1972. [II] J.R. Nicholson and J.A. Malack, "RF impedance of power lines and line impedance stabilization networks in conducted interference measurements", IEEE Trans. Electromag. Compat., May 1973. J.A. Malack and J.R. N i c h o l s o n , "RF impedance of Un i t e d S t a t e s and European power l i n e s " , IEEE Trans. Electromag. Compat., Feb. 1976. S. O s h i t a and K. Feher, "Performance of coherent PSK and DPSK systems i n an impulsive and Gaussian noi s e environment", I n t . Conf. Commun., Conf. Rec., v o l . 1, 56.4.1, 1981. P.A. B e l l o and R. E s p o s i t o , "A new method f o r c a l c u l a t i n g p r o b a b i l i t i e s of e r r o r s due to impulsive n o i s e " , IEEE Tran. Communication Tech., June 1969. P.A. B e l l o and R. E s p o s i t o , " E r r o r p r o b a b i l i t i e s due to impulsive noise i n l i n e a r and hard l i m i t e d DPSK systems", IEEE Trans. Communication Tech., Feb. 1971. A.B. Bodonyi, " E f f e c t s of impulsive noise on d i g i t a l data t r a n s m i s s i o n " , IRE Trans. Communications Systems, Dec. 1961. L.R. H a l s t e d , "On b i n a r y data t r a n s m i s s i o n e r r o r r a t e s due to combinations of Gaussians and impulsive n o i s e " , IEEE Trans. Communications Systems, Sept. 1963. J.S. Engel, " D i g i t a l t r a n s m i s s i o n i n the presence of impulsive n o i s e " , B.S.T.J., Oct. 1965. R.E. Ziemer, "Character e r r o r p r o b a b i l i t i e s f o r M-ary s i g n a l i n g i n impulsive noise environment", IEEE Trans. Communication Tech., Feb. 1967. A.D. Spanlding and D. Middleton, "Optimum r e c e p t i o n in an impulsive i n t e r f e r e n c e environment", IEEE Trans. Communication, Sept. 1977. R.L. Pick h o l t z , D.L. S c h i l l i n g and L.B. M i l s t e i n , "Theory of spread-spectrum communications--A t u t o r i a l " , IEEE Trans. Communication, Part I, May 1982. C. Leung and A. Lam, "Forward e r r o r c o r r e c t i o n f o r an ARQ scheme", IEEE Trans. Communication, Oct. 1981. S. L i n and D.J. C o s t e l l o , J r . , "Automatic-Repeat-Request e r r o r - c o n t r o l schemes", IEEE Communications Magazine, Dec. 1984. F. Chiu, I n t e r n a l communications. Nonwire M u l t i d r o p Network User's Guide, Rev. I, May 1,1985. - 1 1 8 -[26] K. F e h e r , D i g i t a l Communications: S a t e l l i t e / E a r t h S t a t i o n E n g i n e e r i n g , P r e n t i c e - H a l l I n c . , 1983, p. 191. [27] J.G. P r o a k i s , D i g i t a l Communications, M c G r a w - H i l l I n c . , 1983, p. 125. [28] R.C. D i x o n , Spread Spectrum Systems, John W i l e y and Sons, 1976, pp. 217-260. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096358/manifest

Comment

Related Items