Project: IEEE P Working Group for Wireless Personal Area Networks N

Transcription

1 Project: IEEE P Working Group for Wireless Personal Area Networks N (WPANs) Title: [Merged Proposal of Chaotic UWB System for a] Date Submitted: [March 7, 2005] Source: [(1) Young-Hwan Kim, Jae-Hyon Kim, Chia-Chin Chong, Su Khiong Yong, Seong-Soo Lee, (2) Hyung Soo Lee, Cheol Hyo Lee, (3) Jeongsuk Lee, (4) Namhyong Kim, (5) Kyung Sup Kwak, (6) A. S. Dmitriev, A. I. Panas, S. O. Starkov, Yu. V. Andreyev, E. V. Efremova, L. V. Kuzmin, (7) Haksun Kim, (8) Jaesang Cha, (9) Dong Jo Park, Dan Keun Sung, Sung Yoon Jung, Chang Yong Jung, (10) Joon Yong Lee, (11) Dong In Kim, Serhat Erküçük] Company: [(1) Samsung Electronics Co., Ltd. (Samsung Advanced Institute of Technology (SAIT)), (2) Electronics and Telecommunications Research Institute (ETRI), (3) Samsung Electro-Mechanics Co., Ltd. (SEM), (4) Samsung Electronics (DM), (5) UWB-ITRC, Inha University, (6) Institute of Radio Engineering and Electronics (IRE), (7) Hanbat Univ., (8) Seokyeong Univ., (9) Korea Advanced Institute of Science and Technologies (KAIST), (10) Handong Global University (HGU), (11) Simon Fraser University] [(1) jae.kim@samsung.com, (2) clee7@etri.re.kr, (3) js0305.lee@samsung.com, (4) namhyong.kim@samsung.com, (5) kskwak@inha.ac.kr, (6) chaos@mail.cplire.ru, (7) hskim@hanbat.ac.kr, (8) chajs@skuniv.ac.kr, (9) syjung@kaist.ac.kr, (10) joonlee@handong.edu, (11) dikim@sfu.ca] Re: [Response to IEEE a Call for Proposals (04/380r2)] Abstract: [Proposal for the IEEE a PHY standard based on the chaotic UWB system technology.] Purpose: [Proposal for the IEEE a PHY standard.] Notice: This document has been prepared to assist the IEEE P It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P Slide 1

2 Samsung Electronics (SAIT)/IRE/Samsung Electro-Mechanics(SEM), Samsung Electronics (DM), Electronics and Telecommunications Research Institute(ETRI)/KAIST/HGU, Simon Fraser University(SFU)/Inha University, Chaotic UWB System Young-Hwan Kim, Jae-Hyon Kim, Chia-Chin Chong, Su Khiong Yong, Seong-Soo Lee, A. S. Dmitriev, A. I. Panas, S. O. Starkov, Yu. V. Andreyev, E. V. Efremova, L. V. Kuzmin, Jeongsuk Lee, Haksun Kim, Jaesang Cha, Namhyong Kim, Haksun Kim, Jaesang Cha, Hyung Soo Lee, Cheol Hyo Lee, Dong Jo Park, Dan Keun Sung, Sung Yoon Jung, Chang Yong Jung, Joon Yong Lee, Dong In Kim, Serhat Erküçük, Kyung Sup Kwak, Sarm Goo Cho Slide 2

3 CONTENTS 1. INTRODUCTION 2. CHAOTIC COMMUNICATION SYSTEM 3. GENERAL SOLUTION CRITERIA 3.1. Unit Manufacturing Cost/Complexity (UMC) 3.2. General Definitions 3.3. Signal Robustness 3.4. Technical Feasibility 3.5. Scalability 4. MAC PROTOCOL SUPPLEMENT 4.1 MAC Enhancements and Modifications 5. PHY LAYER CRITERIA 5.1. Channel models and payload data 5.2. Size and Form Factor 5.3. PHY-SAP Payload Bit Rate and Data Throughput 5.4. Simultaneously Operating Piconets 5.5. Signal Acquisition 5.6. System Performance 5.7. Ranging 5.8. Link Budget 5.9. Sensitivity Power Management Modes Power Consumption Antenna Practicality Compatible Modulation Scheme: DCSK Compatible Modulation Scheme: MC-PPM Slide 3

4 1. INTRODUCTION Features of Proposed System Low Hardware Complexity / Low Cost Chaotic signal can be generated directly into the desired microwave band (Simple RF circuit) Efficient Power Management Sleep / Wake-up capability can save the battery life time Robust in Multipath In case of OOK Modulation, BER performance against multipath is close to the AWGN (only few db difference) Flexible Pulse Length Chaotic radio pulse can be transmitted with different pulse time duration regardless of the spectral bandwidth Slide 4

5 2. CHAOTIC COMMUNICATION SYSTEM Chaotic Source Chaotic source generates oscillations directly in a specified microwave band. Information component is put into the chaotic carrier to form a stream of chaotic radio pulses. Information can be retrieved from the chaotic radio pulses without intermediate heterodyning. Chaotic Source Generator Circuit Experiment device Slide 5

6 2. CHAOTIC COMMUNICATION SYSTEM Spectral Properties of Chaotic Signal Spectral properties don t change even though the length or duration of the chaotic pulses are varied Amplitude Normalized Power Spectral Density Chaotic Signal Time, t [ns] Amplitude Time (s) x 10-6 Normalized Power Spectral Density Frequency [GHz] Frequency [GHz] Chaotic Pulse (OOK) Slide 6

7 2. CHAOTIC COMMUNICATION SYSTEM Modulation Schemes OOK (main modulation type) Advantages: Lower complexity ( TX and RX ) 3 db moreenergy efficiency than DCSK or PPM => battery saving Disadvantages: Requires non-zero detection threshold DCSK (compatible modulation type) PPM (compatible modulation type) Slide 7

8 3.1. Unit Manufacturing Cost/Complexity Complexity (OOK) RF part of the transceiver: Chaotic oscillator in GHz frequency band with 10 dbm output power amplifier (common complexity is equivalent to 4 power amplifiers) Switch-modulator LNA (amplification db) 2 Band Pass Filter with bandwidth 1 GHz (in band GHz) Envelope detector Antennas No mixers, no correlators, no RF VCO Baseband part of the transceiver: Reference oscillator 20 MHz Bandpass amplifiers Threshold detector or 4 bit A/D converter Frequency Synthesizer on MHz (for ranging) Digital part with ~ 10K gates Slide 8

9 3.4. Technical Feasibility Prototype 1 The communication test has successfully done using Chaotic pulses UWB DCC-OOK Test-bed Slide 9

10 3.4. Technical Feasibility Prototype 2 Battery Digital Block RF Receiver Generator Switch Antenna Slide 10

11 3.5. Scalability Chaotic Pulse Duration 10 0 BER with various β 20Mbps 10Mbps 5Mbps 10-1 T Bit duration BER 10-4 T Duty Cycle Eb/No = 10dB Eb/No = 12dB Eb/No = 14dB Eb/No = 16dB Eb/No = 18dB Eb/No = 20dB Eb/No = 22dB β(number of samples per one bit) T Repeated transmission Slide 11

12 5.1. Channel models and payload data Refer to the selection criteria document Industrial environment NLOS Indoor residential LOS Outdoor LOS Agricultural areas Body area networks Slide 12

13 5.2. Size and Form Factor Values PHY level (130 nm technology) RF part of transceiver => 0.3 mm 2 Analog part of transceiver PHY level baseband => 0.2 mm 2 Digital part of transceiver PHY level baseband => 0.3 mm 2 Common layout square for PHY-level => 1.0 mm 2 Antenna: 2.0 x 2.0 cm 2 Slide 13

14 5.3. PHY-SAP Payload Bit Rate / Throughput Payload Bit Rate PPDU (38 Bytes) Preamble SFD PHR PSDU Bytes 32 Bytes 1 0 bits T s = 100 ns : Pulse emission time T s T s T m = 400 ns : Pulse bin width or Bit period Duty cycle, D = 1/4 T s = 100 ns : Pulse emission time T m T m T m = 600 ns : Pulse bin width or Bit period Duty cycle, D = 1/6 Nominal PHY-SAP payload bit rate, X 0 = (1/400ns) (1000/1024) = 2.44Mbps Optional PHY-SAP payload bit rate, X i = (1/600ns) (1000/1024) = 1.63Mbps Slide 14

15 5.3. PHY-SAP Payload Bit Rate / Throughput Throughput Packet 1 Data Frame 1 (38 bytes) 32 bits ACK (11 bytes) 40 bits Data Frame 2 (38 bytes) t data-frame t ACK t ACK-frame LIFS Time for acknowledged transmission, t packet t packet = t data-frame + t ACK + t ACK-frame + LIFS = ( ns) + (32 400ns) + ( ns) + (40 400ns) = 121.6µs µs µs + 16µs = 185.6µs t packet = t data-frame + t ACK + t ACK-frame + LIFS = ( ns) + (32 600ns) + ( ns) + (40 600ns) = 182.4µs µs µs + 24µs = 278.4µs Nominal Data Throughput, T 0 = (32 8/185.6µs) (1000/1024) = 1.35Mbps Optional Data Throughput, T i = (32 8/278.4µs) (1000/1024) = 898kbps Slide 15

16 5.4. Simultaneously Operating Piconets Three Methods to Achieve SOP Frequency division multiplexing (FDM) Four independent frequency channels on 500 MHz guaranties simultaneously operating four piconets. Code division multiplexing (CDM) Deployed a class of unipolar codes (0,1) having ZCD/LCD property maintain orthogonality among piconets. Four set of codes can support four simultaneously operating piconets. Frequency-code division multiplexing (FCDM) Two independent frequency channels with 1 GHz bandwidth and within each frequency channel, a set of codes is used Ex: Only two codes are required to support four SOPs Slide 16

17 5.4. Simultaneously Operating Piconets Combination of FDM and CDM (FCDM) 2 sub-bands and a set of codes for each subbands => at least 4 SOPs BPF A Set of Codes Chaotic Source Freq, GHz Freq, GHz Freq, GHz GHz bandwidth for each sub-band Subband fc, GHz fl, GHz fr, GHz Slide 17

18 5.4. Simultaneously Operating Piconets CDM Methods to Achieve SOP CDM for SOP can be achieved using Unipolar ZCD/LCD Code in chaotic-ook modulation ZCD(Zero Correlation Duration): Local time duration with zero autocorrelation function sidelobe & zero cross-correlation function LCD(Low Correlation Duration): Local time duration with low zero autocorrelation function sidelobe & low cross-correlation function * Local time duration function as an Interference rejection interval for SOP Characteritics of combined schemes Simple circuit with noncoherent envelope detector Novel Inter/Intra Piconet Interference immunity for an efficient SOP Slide 18

19 5.4. Simultaneously Operating Piconets Example of Unipolar ZCD Codes Type Type1 : Circular type sequence A code set is constructed by chip shift of a seed code An example of (8,4,0,0) with M=2 code a=[ ] code b=[ ] Type2 : Non-Circular type sequence An example of (5,2,0,0) with M=3 code a=[ ] code b=[ ] code c=[ ] Where (N,W,A,C) is N = sequence period, W = number of nonzero elements, A = ACF sidelobe in ZCD/LCD, C = CCF value in ZCD/LCD M = family size, Truncation of N/M = W Slide 19

20 5.4. Simultaneously Operating Piconets Transceiver Architecture of Chaotic-OOK Based ZCD/LCD-CDM Tx1(Desired user) Tx1 Tx t Unipolar DATA t Spreading t Unipolar Code1 Unipolar Code4 OOK Modulation Chaotic Source OOK Modulation Unipolar DATA Spreading t Chaotic Source t PA PA Radio Channel CDM Code1:Piconet1 Code2:piconet2 Code3:piconet3 Code4:piconet4 t Rx1 t 1 Matched Recovered DATA Envelope 0 Filter t Detection Detector BPF LNA Received signal Slide 20

21 5.4. Simultaneously Operating Piconets Baseband Chaotic-OOK- ZCD-CDM Slide 21

22 5.4. Simultaneously Operating Piconets Chaotic-OOK-ZCD-CDM Slide 22

23 10 0 March System Performance AWGN & Multipath AWGN & Multipath BER Vs. Eb/No 2 GHz Bandwidth 10 0 PER Vs. Eb/No 2 GHz Bandwidth BER 10-4 PER AWGN Residential LOS (CM1) Open outdoor LOS (CM5) Industrial NLOS (CM8) E /N, db b AWGN Residential LOS (CM1) Open outdoor LOS (CM5) Industrial NLOS (CM8) E /N b 0 Modulation: OOK, Bandwidth: 2GHz, Pulse width: Tm=400ns, Pulse emission time: Ts = 100ns, PSDU length: 32 bytes Slide 23

24 5.6. System Performance Values: Bit Rate and Distance Xo (Mbps) Channel M+L1+L2 (free space), db PL, db PL 0, db n Distance, m AWGN AWGN AWGN Slide 24

25 5.7. Ranging Ranging Algorithm MHz Pulse source MHz Pulse source N3 Overlap detector N2 delay N1 start both pulse sources & counter N3 no 1st delayed pulse? yes start counter N1 Digital Block Counter N1 counts delayed pulses Counter N2 counts overlaps between delayed pulses( MHz) and reference pulses( MHz) Counter N3 counts reference pulses no no 1st overlap match? yes stop N1 & N3, start N2 last overlap match? yes stop N2, calculate Tx Slide 25

26 5.7. Ranging Operation of Counters С 1 С 2 С 3 Ref. f1 T x N3 f0 N1 N2 t* * N1, N2, N3 pulse numbers T x = (N3+0.5 N2)/f 1 (N1+0.5 N2)/f 0 distance S = 0.5*c*(T x -τ 0 ) t 0 t 1 t 2 t 3 τ 0 retranslation time Operation time of counters C 1,C 2,C 3. Slide 26

27 5.7. Ranging Overlapping of Delayed & Reference Pulses Delayed pulse Reference pulse Pulse overlap Slide 27

28 5.7. Ranging Values: Range System supports ranges: Range from 0 to 30 m (typical) Range up to 100 m (max 10 kbps data rate) Slide 28

29 5.8. Link Budget Slide 29

30 5.10. Power Management Modes Sleep and Wake-up Scheme Wake Up Signal Wake Up Structure Wake Up Radio Power Detector Main Transceiver Slide 30

31 5.11. Power Consumption Power Calculation Transceiver Tx Rx CU P e is emitted power, η is efficiency, Control Unit Operation time T oper T oper = C b U b / P av η best is the best of all possible efficiencies, P in is instantaneous emission power, T e is time of emission for given transmission rate, Slide 31 Average power consumption P av P av = P Tx + P Rx + P CU P Tx = P e / η P Rx = P e / η best P e = P in T e = 1/2 D P in T bit R T bit is duration of one bit, R is transmission rate, C b is battery capacity, U b is battery voltage, D is duty cycle.

32 5.11. Power Consumption Duty Cycle and Power Consumption Transmission Rate R, kbps Average Emitted Power P e, mw Average Power Consumption P av (η = 5%) Lifetime of the AAA battery, years µw µw mw % duty cycle 15 10% duty cycle % duty cycle P CU = 7.5 µw ; P in = 4 mw ; η best = 5%; U b = 1.5 V ; C b = 750 mah; D = 1/4 Example: R = 1 kbps; T bit = 400 ns; η = 5% P e = 1/2 D P in T bit R = 0.2 µw P av = P Tx + P Rx + P CU = P e /η + P e /η best + P CU = 15.5 µw Slide 32

33 Conclusion Chaotic communications meet the low power, low cost & low complexity requirements best suited for 15.4a applications. Proposed DCC-OOK compliant with FCC UWB PSD regulation. Feasibility and scalability are guaranteed with precision ranging and SOP capabilities. The implemented test bed demonstrated the feasibility of DCC technology. Slide 33

34 DCSK: Compatible Modulation Scheme for Direct Chaotic Communication Slide 34

35 DCSK Modulation DCSK Differential Chaos Shift Keying (DCSK) One of the modulation scheme as an alternative to OOK DCSK transmits a reference chaotic pulse and an information data pulse depending on whether information bit 1 (same ref. chaotic pulse) or 0 (inverted of the chaotic pulse) is being transmitted The information signal can be recovered in the receiver by a correlator with a constant decision threshold The Chaotic properties are maintained as same as OOK Data rate is as same as OOK SOP can be achieved by transmitting Chaotic pulses with different length Slide 35

36 DCSK Modulation Principle OOK Vs DCSK OOK DCSK 10-2 BER Transmitter Eb/No Receiver T/2 Integrator T/2 Chaotic Generator Delay T/2-1 Delay T/2 T Threshold Data Bit Stream Slide 36

37 DCSK Modulation System Simulation Results AWGN & Multipath BER Channel 1: Indoor residential LOS Channel 5: Outdoor residential LOS Channel 9: Agricultural area AWGN channel Eb/No Slide 37

38 5 0 DCSK Modulation SOP: LDMA Piconet 1 Piconet 2 Piconet 3 Piconet 4 Piconet Piconet Piconet Piconet All In DCSK SOP can be done using Chaotic Length Division Multiple Access (LDMA) LDMA works based on the exploitation of different chaotic length assigned to each piconets. LDMA is based on the spectral and correlation property of chaotic signal BER Piconet 1 User Detection 4 Users 8Mbps 5Mbps S/N Slide 38

39 5 Mbps March 2005 DCSK Modulation Scalability Bit = nsec Scalability can be achieved using Chaotic gain Varying bit duration Duty cycle Repeated transmission of information bearing chip. Chaotic Gain in DCSK Gain 5Mbps 4Mbps 2Mbps 250 nsec Mbps 0 BER nsec 5 1 Mbps S/N Slide 39

40 MCS-DCSK Modulation Combination of MCSK TH-IR with DCSK MCS-DCSK M-ary code shift keying (MCSK)/binary pulse position modulation (BPPM) for time hopping (TH) impulse radios (IR s) can be used in Chaotic Communications such as DCSK in order to increase the system performance Slide 40

41 MCS-DCSK Modulation DCSK TX Signal DCSK transmitting d=[d 1 d 2 ], d i ε ( 1,1) A Reference signal 50 ns Information signal Reference signal 50 ns Information signal 50 ns 50 ns 50 ns 0 T f 2T f d 1 (bit-1) d 2 (bit-2) transmitted transmitted where info. signal = sign( d i ) x ref. signal Slide 41

42 MCS-DCSK Modulation DCSK RX Signal DCSK receiver no AWGN, no MP fading Reference signal Information signal Reference signal Information signal Reference signal Information signal Reference signal Information signal delay T f / 2 integration over T f / 2 integration over T f / 2 detect d 1 detect d 2 Slide 42

43 MCS-DCSK Modulation DCSK TX and RX Signal DCSK: Transmitted and received signals (CM1, no AWGN) Slide 43

44 MCS-DCSK Modulation MCS-DCSK TX Signal MCS-DCSK transmitting d=[d1 d2], di ε ( 1,1) A Reference signal 100 ns Information signal 200 ns d 1 = ns 0 T f 2T f A Reference signal 100 ns Information 100 ns signal d 1 = 1 where info. signal = sign( d 2 ) x ref. signal Slide 44

45 MCS-DCSK Modulation MCS-DCSK RX Signal Reference signal Reference signal Information signal Information signal 0 2T f Reference signal Reference signal delay 3T f / 4 MCS-DCSK receiver no AWGN, no MP fading Decision Step 1: Step 2: integrate over T f / 2 integrate over T f / 2 dˆ ( 1 1) dˆ (1) 1 dˆ 1 ( 1) > dˆ 1 (1) d 1 = -1; dˆ 1 ( 1) < dˆ 1 (1) d 1 = 1 ( dˆ (-1) ); if d = 1 d sign ( dˆ (1) ) if d 1 = -1 d 2 = sign = 1 Slide 45

46 MCS-DCSK Modulation MCS-DCSK TX and RX Signal MCS-DCSK: Transmitted and received signals (CM1, no AWGN) Slide 46

47 MCS-DCSK Modulation MCS-DCSK Simulation Results 10 0 Channel: AWGN BER DCSK - 2.0GHz BW MCS-DCSK - 2.0GHz BW DCSK - 1.0GHz BW MCS-DCSK - 1.0GHz BW DCSK - 0.5GHz BW MCS-DCSK - 0.5GHz BW E b /N 0 (db) Slide 47

48 MCS-DCSK Modulation MCS-DCSK Simulation Results 10 0 Channel: Residential LOS 10-1 BER DCSK - 0.5GHz BW DCSK - 1.0GHz BW DCSK - 2.0GHz BW MCS-DCSK - 0.5GHz BW MCS-DCSK - 1.0GHz BW MCS-DCSK - 2.0GHz BW E b /N 0 (db) Slide 48

49 DCSK Modulation Complexity, Cost & Technical Feasibility Complexity and cost will be slightly higher compare to the OOK chaotic system proposed Conclusion Chaotic communication based on DCSK modulation is an alternative solution for TG4a. Most hardware from OOK is retained. SOP and ranging can be solved effectively using DCSK. Slide 49

50 MC-PPM : Compatible Modulation Scheme for Direct Chaotic Communication Slide 50

51 MC-PPM Modulation MC-PPM Multi-coded Pulse Position Modulation (MC-PPM) Power efficient scheme Inherent coding gain due to orthogonal multi-codes Support wide pulse shaping in same data rate condition Constant decision threshold in the receiver OOK is one special mode of MC-PPM Slide 51

52 MC-PPM Modulation Principle Principle operation (L=3, Ns=4) = = = = 1 Multi-coded symbol ( Code rate : L/Ns ) Ex. Code rate = 3/ Data block ( L bits ) Ex. L=3 Orthogonal code set ( Code Length : Ns ) Ex. Ns=4 Modulation MC-PPM Signal : Slide 52

53 MC-PPM Modulation Data Frame Structure 1 data block (L data) interval of PSDU : Preamble SFD PHR PSDU T = N ( T + T ), T = N T, T = ( L+ 1) T d r s g s s c c m L N N T T s r m c T s T T g d : # of bits per data block : Orthogonal code length : # of repetitions : Pulse bin width (duration) : Multi-coded chip duration : Multi-coded symbol duration : Guard time for processing delay : Total transmit time duration of a data block T c 1 T g 2 T g r T s Ns 12 L +1 L T m T d N s... N r N : # of Repetitions : Orthogonal Code length : Position number for MC-PPM Tg Slide 53

54 MC-PPM Modulation Transceiver Architecture Transmitter [ ] T b = bb 1 2Lb L Data T = 1 2L s d d d d N Data Encoder Orthogonal Multi-code [ c c L c ] 1, 2,, L d = C b Data Modulator MC-PPM Pulse Generator Channel rt () Receiver T = 1 2L s d d d d N rt () Energy Detector Data DeModulator MC-PPM Location Detector Data Decoder Orthogonal Multicode [ c c L c ] 1, 2,, L T b= C d Data [ ] T b= bb 1 2Lb L Slide 54

55 MC-PPM Modulation PHY-SAP Data Rates Flexible data rates can be supported according to several design parameter (Tm, L, Ns, Nr, Tg) Tp = 20ns Tm = 200ns Tp Tm L Ns Nr Tg Data Rate 20ns 200ns ns kbps 20ns 200ns ns 228 kbps 20ns 200ns ns 457 kbps 20ns 200ns ns 2.44 Mbps Slide 55

56 MC-PPM Modulation Data Throughput Data Throughput t long_frame t tx t ACK t ACK_frame LIFS Transmission time (ttx) & Data throughput (Rth) For L=3, Ns=8, Nr=1,Tg=0ns (457kbps) ttx = tlong_frame + tack + tack_frame + LIFS = u u u u = 913 u Rth = 32 8 / 913u kbps ( Nominal throughput based on 32 bytes payload ) For L=3, Ns=16, Nr=1,Tg=0ns (228kbps) ttx = tlong_frame + tack + tack_frame + LIFS = u u u u = u Rth = 32 8 / u kbps ( Nominal throughput based on 32 bytes payload ) Slide 56

57 MC-PPM Modulation Signal Acquisition Energy detection based acquisition Acquisition should be performed in order to make synchronization and demodulate data Synchronization : Non-coherent Slide 57

58 MC-PPM Modulation Performance MC-PPM Performance : AWGN BER & PER L=3, Ns=8, Nr=1 (457 kbps PHY-SAP data rate) BER PER (%) EbNo (db) EbNo (db) Slide 58

59 MC-PPM Modulation Performance MC-PPM Performance : 4a Channel Models BER & PER L=3, Ns=8, Nr= CM8 CM1 CM CM8 CM1 CM BER PER EbNo (db) EbNo (db) Slide 59

60 MC-PPM Modulation SOPs Time Division Configuration of SOPs Self configuration of SOPs is possible Piconet #1 Active Inactive Piconet #2 Piconet #3 Slide 60

61 MC-PPM Modulation SOPs Self Configuration of SOP Passive Scan Repeat scanning one channel Usage Starting a new piconet (FFD) Device higher layer MLME-SCAN.request ScanDuration Device MAC Beacon Beacon Coordinator MAC Coordinator MAC Association (FFD or RFD) ScanDuration Beacon Beacon MLME-SCAN.confirm Slide 61

62 MC-PPM Modulation Link Budget & Sensitivity Link Budget & Sensitivity based on MC-PPM Parameter (mandatory) Value at d=30m (mandatory) Value at d=10m peak payload bit rate (457kb/s) [ L=3,Ns=8,Nr=1] (457kb/s) [ L=3,Ns=8,Nr=1] Average Tx power (dbm) (dbm) Tx antenna gain 0 (dbi) 0 (dbi) geometric center frequency of waveform 3.90 (GHz) 3.90 (GHz) Path loss at 1 meter 44.5dB 44.5dB Path loss at d m db at d =30m 20 db at d =10m Rx antenna gain 0 (dbi) 0 (dbi) Rx power (dbm) (dbm) Average noise power per bit (dbm) (dbm) Rx Noise Figure 7 (db) 7 (db) Average noise power per bit (dBm) (dBm) Minimum Eb/N0 (S) [Ep/N0] 20 (db) 20 (db) Implementation Loss (I) 5 (db) 5 (db) Link Margin 2.85(dB) 12.39(dB) Proposed Min. Rx Sensitivity Level -85.4(dBm) -85.4(dBm) Slide 62

63 MC-PPM Modulation Ranging Scheme TOA/TWR -> Measurement of Roundtrip time T round trip Node 1 t 0 Packet 1 T propagation2 Packet 2 t 3 Node 2 T propagation1 t 1 Packet 1 Packet 2 T processing time t 2 Slide 63

64 MC-PPM Modulation Ranging Performance a channel (cm4) Single user No narrowband interference Pulse width = 20ns Integration time = 2ns Pulse repetition period = 200ns Length of search region = 40ns Threshold level was determined relative to noise floor A separate envelope detector for range estimation was employed Slide 64

65 Backup Slides Slide 65

66 Tolerance of Components Capacitor, C1 and inductance, L 20% tolerance. C2 and resistors, RE and R1 5% tolerance. C BFP620 X3 L L10 C L=L C17 C=C1 C C20 C=100 pf R RL1 R=RL Vampin Vampout Vout DT v a_hp_mga-66100_ Amp1 V_DC SRC1 Vdc=VC C C16 C=C2 V_DC SRC2 Vdc=VE R RE1 R=RE DA_LCBandpassDT1_colp_collector_amp_f lt DA_LCBandpassDT1 Fs1=2 GHz Fp1=3.1 GHz Fp2=5.1 GHz Fs2=6 GHz Ap=3 db As=40 db N=4 ResponseTy pe=elliptic Rg=50 Ohm Rl=50 Ohm R R12 R=50 Ohm E Slide 66

67 Summary of Features Information carrier Band division Chaotic radio pulses 3 bands within FCC Mask ( , and GHz) Channel bandwidth Pulse duration 2.0 GHz band or 4 channels with 500 MHz in each in the 2 GHz band 400 ns Individual bit rate 1 Kbps 10 Kbps 100 Kbps Transmit power -30 dbm -20 dbm -20 dbm 2.5 year 2.5 year 2.5 year Battery life 100% duty cycle 10% duty cycle 0.1% duty cycle Aggregated bit rate Up to 5 Mbps Slide 67

68 Tiny Chaotic Transmitter Transmitter consists of: - chaos generator - modulator - antenna Frequency band GHz Radiating power mw Slide 68

69 DCSK Modulation SOP System Block Data Data Data Bit Frame Generator Chaos Signal Generator Template Data Chaos Receiver Slide 69

70 DCSK Modulation SOP Transmission Frame1 T1 D11 D1n T2 D21 D2n Frame2 T1 D11 D1n T2 D21 D2n Template Bit Piconet1 Piconet2 Bit Frame Slide 70

71 DCSK Modulation SOP Detail Data Integrator 1 bit Duration Template Z Z Z Receiver Details Slide 71 Z

72 DCSK Modulation SOP Signal Processing User Multi_path Channel User User Slide 72

73 DCSK Modulation Ranging Block Diagram Z -1 Serial-to-Parallel Envelop Detection & Signal Point Detection Delay Circuit Slide 73

74 DCSK Modulation Ranging Coordinator Source Time Counter Device 1. Offset by Comparison between (Source Time Counter - Target Time Counter) & (Source Time Counter - Source2 Time Counter) 2. Distance from (Source Time Counter -Source2 Time Counter) Source Time Counter + Target Time Counter - Offset + Offset Adjusting Time Counter By Offset Confirm Counter Justification 0 Completion Slide 74

75 DCSK Modulation Ranging Device (-2 Offset) Coordinator Initial : st Pass : PNC recalculates Device Arrival time : (16/2) 2. Compare value from 1 and Device : Transferred as Offset 4. 8 Kept for Distance between PNC and Device Slide 75

76 DCSK Modulation Location Awareness Special Mode Timing Counter Fine Synchronization PNC disseminates special frame to inform Device of Location special mode Device acknowledges with its own timing count PNC compares its own count with Device s count, and extract an offset between them PNC sends negative offset in order for Device to compensate its timer Device informs PNC of all being set Slide 76

77 DCSK Modulation Location Awareness Special Mode Template Frame Data Frame X Y Data Template Envelop Detection Delay Circuit by 1~3 ns Slide 77

78 DCSK Modulation Ranging Fine Precision TOA Estimation Suggest Special mode different from Normal mode, which needs faster clock In special mode, Estimate how far Signal detached from fixed time slot with finer clock This obtained value returned with Response command to Request command from MAC Slide 78

79 DCSK Modulation Ranging Delay Circuit 100 MHz Phase 0 Phase 90 Phase 180 Phase ns Slide 79

80 DCSK Modulation Ranging Simulation (BNR 16dB) Maximum Index of Moving Average by duty cycle Duration will be converted to distance. real distance : meter 2.5 ns precision distance : meter Error : meter Slide 80 real distance : meter 2.5 ns precision distance : meter Error : meter

81 M-ary Code Shift Keying/Binary PPM (MCSK/BPPM) Based Impulse Radio Slide 81

82 Motivation MCSK/BPPM increases the location/ranging capability of existing Time Hopping (TH) Impulse Radios (IRs) H/W complexity is not increased Same signal space with respect to TH-BPPM MCSK can be applied to other TH-IRs; eg. MCSK/BPSK Slide 82

83 TG4a Requirements a PHY scalable information rates high precision ranging/ location low power consumption MCSK/BPPM compared to TH- BPPM Better BER performance at the same/higher information rates and lower transmit power Improved ranging/location precision capability Lower transmit power at the same/higher information rates and better BER performance low complexity and cost No new circuit is needed / simple transceiver structure *MCSK/BPPM: M-ary Code Shift Keying/Binary Pulse Position Modulation **TH-BPPM: Time Hopping Binary Pulse Position Modulation Slide 83

84 MCSK/BPPM TH PPM user #1 MCSK: M-ary Code Shift Keying BPPM: Binary Pulse Position Modulation TX d 1) ( = [ ] 1 user specific TH code MCSK/BPPM user #1 TH-BPPM only for multiple access T b 2T b 3T b t d 1) ( = M=4 [ ] choose a code TH-BPPM T b 2T b 3T b t M user specific TH codes 1 TH code for multiple access and data modulation c3 T b T f : : Bit time Frame time Slide 84

85 PHY TX Structure (1/2) M user specific TH codes Example: d= [ ] MCSK M = s 4, N p = 8, N = 4 c 0 7 c1 = 3 TH= c 2 5 c TH codes are periodic with Np - each pulse should be repeated Ns times - Np/Ns=k is an integer TX C T b 2Tb 0 Tf 2T f 3T f 4T f 5T f 6T f 7T f 8T f t d = BPPM [ ] T b T f : : Bit time Frame time Slide 85

86 PHY TX Structure (2/2) M user specific TH codes TX - TH codes are periodic with Np - each pulse should be repeated Ns times - Np/Ns=k is an integer Information rate vs. BER performance for fixed Ns and varying Np and M Scenario N p / Ns = 1 M = 4 0 Time domain illustration 2 bits (MCSK) 1 bit (BPPM) T b Np/Ns same Info. rate BER performance N p / Ns = 1 3 bits (MCSK) M increasing M N p = 8 / Ns = bit (BPPM) T b 3 bits (MCSK) M same M = bit (BPPM) Tb 1 bit (BPPM) 2Tb Np/Ns increasing T : Bit time T : b f Frame time Slide 86

87 TH Code Assignment (1/2) Each user has M user specific TH codes N N M u p TX sample-long sequence? Generation of TH codes Case 1: random assignment l Nh ; l 0 T f m-sequence: For Tf = 100ns, Tc = 1ns: 100 slots for multiple access NO! 2 = 6, N = 64 [ ] h N p M = = 4 4 c2 c0 c0 c3 c3 c2 [ ] c1 user #1 c1 user #2 N u N p N u ( N + M 1) p Slide 87

88 TH Code Assignment (2/2) Generation of TH codes Case 2: no overlapping user #1 user #2 TX [ ] N p M = = 4 4 user #1 [ ] [ ] [ ] [ ] [ ] no collisions allowed within user codes... user #2 user #k [ ] ( N + M ) n Nu N p Nu p 1 + n: number of overlaps Slide 88

89 General Modulation Format TX Fixed signal space Increased information rate Extra information Random selection of TH codes Improved spectrum R s log M = Np / Ns R Slide 89

90 Receiver Structure - MLSE RX M M template signals 2M Np/Ns hardware structure computation complexity 1 correlator ( p N s ) 2 N / M Slide 90

91 Information Rate (1/3) TH-BPPM A 0 f T 2 T f R s log M = N p / N s Ns = 2, M=1 Info. rate Rs R R MCSK/BPPM Constant Energy/Bit Constraint Ns = 2, Np=2, M=2 A 0 f T 2 T f 2R MCSK/BPPM Constant Power Constraint Ns = 2, Np=2, M=2 A 0 f T 2 T f 2R MCSK/BPPM (same info. rate) Constant Power Constraint Ns = 2, Np=2, M=2 A 0 T f 2T f can be adjusted to achieve higher information rate at lower transmit power and still maintain better BER performance at the same time A = R log2 M 1+ N / N M T log f = N p / Ns p s A T f Slide 91

92 N N p s =1 March 2005 Information Rate (2/3) A 0 A =2A T f T 2T f 3T f f 0 T f = 4T f A MCSK/BPPM Constant Power Constraint for Ns = 1, M=8 Scalable info. rates 4R R R s log M = N p / N s BER performance (wrt TH-BPPM) R - increased SNR - reduced collusions - no processing gain - not much improvement N N p s = 2 0 A =1.58A Tf 2T f 0 T 2T f f f 3T f 4T f A T 2 T = 5 f T f 2.5R R - increased SNR - reduced collusions - processing gain - improved BER - TX power can be lowered - info rate can be increased N N p s = 3 0 A =1.41A 0 T Tf = 2T f 3T f f 4T f Tf 2T f 3T f 5T f 6T f Slide 92 2R R A = log2 M 1+ N / N M T log f = N p / Ns p s A T f

93 Information Rate (3/3) Constant Power Constraint Improved performance at the same information rate for M= Power Const., Same Info. Rate, M=8 TH BPPM Np/Ns=1 Np/Ns=2 Np/Ns=3 Np/Ns= BER SNR (db) Slide 93

94 Location Accuracy MCSK/BPPM Constant Power Constraint Step 0 Procedure Result Comment Initial conditions for TH-BPPM R0 (information rate); BER0 (performance) TX0 (power) Step 1 Step 2 Step 3 Increase M Increase Np/Ns Increase T f Step 4 Increase A R > R ; BER1 > BER0 ; RTX> R= TX > R ; 1 BER TX R 1 TX R BER 4 BER TX 2 = TX > R > BER 2 > TX > R > BER ; > R 3 ; 0 ; BER2 may or may not be less than BER0 BER3 may or may not be less than BER0 = R3 > R0; Increased frame time with longer observation period, 3 > BER4 & BER0 > BER4; higher information rate, 0 > TX 4 > TX better BER performance 3 and lower transmit power Accurate Ranging/Location Slide 94

95 Conclusion MCSK/BPPM provides: increased information rate lower transmit power better BER performance improved spectral characteristics Simultaneously! MCSK/BPPM is capable of: information rate scalability location/ranging accuracy IEEE a PHY Slide 95

96 Back-up Slides Slide 96

97 MCSK/BPPM Constant Power Constraint BER Power Const. BER for Np/Ns = 1 (TH code set 1) M=2 M=4 M=8 TH BPPM SNR (db) Power Const. BER for Np/Ns = 2 (TH code set 1) M=2 M=4 M=8 TH BPPM Power Const. BER for Np/Ns = 4 (TH code set 1) M=2 M=4 M=8 TH BPPM BER 10 2 BER SNR (db) SNR (db) Slide 97

98 MCSK/BPPM Constant Energy/Bit Constraint BER Energy Const. BER for Np/Ns = 1 (TH code set 1) M=2 M=4 M=8 TH BPPM SNR (db) Energy Const. BER for Np/Ns = 2 (TH code set 1) M=2 M=4 M=8 TH BPPM Energy Const. BER for Np/Ns = 4 (TH code set 1) M=2 M=4 M=8 TH BPPM BER 10 2 BER SNR (db) SNR (db) Slide 98

99 Effects of TH Code Design on the Performance MCSK/BPPM Constant Power Constraint Np/Ns=1, TH code comparison M=4, rand. M=8, rand. M=4, det. M=8, det. conv Np/Ns=2, TH code comparison M=4, rand. M=4, det. conv. BER 10 2 BER SNR (db) SNR (db) Slide 99

100 TH Code Spectrum of: a) TH-BPPM, Np=10 b) ideal MCSK/BPPM, Np code spectrum c) realistic MCSK/BPPM frequency (Hz) x 10 8 Fig. a. TH-BPPM ideal spectrum M=8, N p =10 M=256, N p = M=8, N p =256 code spectrum code spectrum frequency (Hz) x 10 8 Fig. b. ideal MCSK/BPPM frequency (Hz) x 10 8 Fig. c. realistic MCSK/BPPM Slide 100