logging in or signing up 15-05-0113-02-004a-merged-uwb-proposal-i srikrishna.hegde Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT lite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 55 Category: Entertainment License: All Rights Reserved Like it (0) Dislike it (0) Added: October 13, 2009 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript Slide 1: Feb 2005 Francois Chin (I2R), et. al. Slide 1 Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [Merged UWB proposal for IEEE 802.15.4a Alt-PHY] Date Submitted: [22 Feb 2005] Source: [Francois Chin, et.al.] Company: [Institute for Infocomm Research, Singapore] Address: [21 Heng Mui Keng Terrace, Singapore 119613] Voice: [65-68745687] FAX: [65-67744990] E-Mail: [chinfrancois@i2r.a-star.edu.sg] Re: [Response to the call for proposal of IEEE 802.15.4a, Doc Number: 15-04-0380-02-004a ] Abstract: [Merged Proposal to IEEE 802.15.4a Task Group] Purpose: [For presentation and consideration by the IEEE802.15.4a committee] Notice: This document has been prepared to assist the IEEE P802.15. 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 P802.15. This contribution is a technical merger between*: : Feb 2005 Francois Chin (I2R), et. al. Slide 2 This contribution is a technical merger between*: Institute for Infocomm Research [05/032] General Atomics [05/016] Thales & Cellonics [05/008] KERI & SSU & KWU [05/033] Create-Net & China UWB Forum [05/019] Staccato Communications [04/0704] Wisair [05/09] * For a complete list of authors, please see page 3. Authors : Feb 2005 Francois Chin (I2R), et. al. Slide 3 Authors Institute for Infocomm Research: Francois Chin, Xiaoming Peng, Sam Kwok, Zhongding Lei, Kannan, Yong-Huat Chew, Chin-Choy Chai, Rahim, Manjeet, T.T. Tjhung, Hongyi Fu, Tung-Chong Wong General Atomics: Naiel Askar, Susan Lin Thales & Cellonics: Serge Hethuin, Isabelle Bucaille, Arnaud Tonnerre, Fabrice Legrand, Joe Jurianto KERI & SSU & KWU: Kwan-Ho Kim, Sungsoo Choi, Youngjin Park, Hui-Myoung Oh, Yoan Shin, Won cheol Lee, and Ho-In Jeon Create-Net & China UWB Forum: Zheng Zhou, Frank Zheng, Honggang Zhang, Xiaofei Zhou, Iacopo Carreras, Sandro Pera, Imrich Chlamtac Staccato Communications: Roberto Aiello, Torbjorn Larsson Wisair: Gadi Shor, Sorin Goldenberg Slide 4: Feb 2005 Francois Chin (I2R), et. al. Slide 4 Multiband Ternary Orthogonal Keying (M-TOK) for IEEE 802.15.4a UWB based Alt-PHY Goals : Feb 2005 Francois Chin (I2R), et. al. Slide 5 Goals Good use of UWB unlicensed spectrum Good system design Path to low complexity CMOS design Path to low power consumption Scalable to future standards Graceful co-existence with other services Graceful co-existence with other UWB systems Support different classes of nodes, with different reliability requirements (and $), with single common transmit signaling Main Features : Feb 2005 Francois Chin (I2R), et. al. Slide 6 Main Features Proposal main features: Impulse-radio based (pulse-shape independent) Common preamble signaling for different classes of nodes / type of receivers (coherent / differential / noncoherent) Band Plan based on multiple 500 MHz bands Robustness against SOP interference Robustness against other in-band interference Scalability to trade-off complexity/performance Slide 7: Feb 2005 Francois Chin (I2R), et. al. Slide 7 Proposed System Parameters System Description : Feb 2005 Francois Chin (I2R), et. al. Slide 8 System Description Each piconet uses one set of code sequences for different classes of nodes / type of receivers (coherent / differential / non-coherent receivers) 16 Orthogonal Sequences of code length 32 to represent a 4-bit symbol PRF (chip rate): 24 MHz Low enough to avoid significant interchip interference (ICI) with all 802.15.4a multipath models High enough to ensure low pulse peak power FEC: optional (or low complexity type) Band Plan : Feb 2005 Francois Chin (I2R), et. al. Slide 9 Band Plan Multiple access : Feb 2005 Francois Chin (I2R), et. al. Slide 10 Multiple access Multiple access within piconet: TDMA+CSMA/CA same as 15.4 Multiple access across piconets: CDM + FDM Different Piconet uses different Base Sequence & different 500 MHz band Types of Receivers Supported : Feb 2005 Francois Chin (I2R), et. al. Slide 11 Types of Receivers Supported Coherent Detection: The phase of the received carrier waveform is known, and utilized for demodulation Differential Chip Detection: The carrier phase of the previous signaling interval is used as phase reference for demodulation Non-coherent Detection: The carrier phase information (e.g.pulse polarity) is unknown at the receiver Slide 12: Feb 2005 Francois Chin (I2R), et. al. Slide 12 Criteria of Code Sequence Design The sequence Set should have orthogonal (or near orthogonal) cross correlation properties to minimise symbol decision error for all the below receivers For coherent receiver For differential chip receiver For non-coherent symbol detection receiver Energy detection receiver Each sequence should have good auto-correlation properties Slide 13: Feb 2005 Francois Chin (I2R), et. al. Slide 13 To minimise impact of DC noise effect on energy collector based non-coherent receiver For OOK signaling, the transmitter transmits {+1,-1,0} ternary sequences Conventional receive unipolar code sequence – follows transmit sequence After the energy capture in the receiver, the noise has positive DC components in each chip; error occurs in thresholding, especially at lower SNR This will accumulate noise unevenly in symbol decision An ideal receive despreading chip sequence should then have bipolar chip values, preferrably with equal number of ‘+1 and ‘-1’ chips This, to certain extent, will nullify DC noise energy in symbol decision This, will also nullify energy components from other interfering piconets Criteria of Code Sequence Design Slide 14: Feb 2005 Francois Chin (I2R), et. al. Slide 14 Base Sequence Set 31-chip Ternary Sequence set are chosen Only one sequence and one fixed band (no hopping) will be used by all devices in a piconet Logical channels for support of multiple piconets 6 sequences = 6 logical channels (e.g. overlapping piconets) for each FDM Band The same base sequence will be used to construct the symbol-to-chip mapping table Slide 15: Feb 2005 Francois Chin (I2R), et. al. Slide 15 Base Sequence #1 Symbol-to-Chip Mapping: Gray coded 16-ary Ternary Orthogonal Keying To obtain 32-chip per symbol, cyclic shift the Base Sequence first, then append a ‘0’-chip in front Slide 16: Feb 2005 Francois Chin (I2R), et. al. Slide 16 Good Properties of the Mapping Sequence Cyclic nature, leads to simple implementation Zero DC for each sequence No need for carrier phase tracking (i.e. coherent receiver) Slide 17: Feb 2005 Francois Chin (I2R), et. al. Slide 17 Synchronisation Preamble Code sequences has good autocorrelation properties Preamble is constructed by repeating ‘0000’ symbols Long preamble is constructed by further symbol repetition Correlator output for synchronisation Frame Format : Feb 2005 Francois Chin (I2R), et. al. Slide 18 Frame Format PPDU Octets: PHY Layer Preamble 4? 1 Frame Length SFD 1 SHR PHR PSDU MPDU Data: 32 (n=23) Frame Cont. Seq. # Address Data Payload CRC Octets: 2 1 0/4/8 2 MAC Sublayer n MHR MSDU MFR For ACK: 5 (n=0) Transmission Mode : Feb 2005 Francois Chin (I2R), et. al. Slide 19 Transmission Mode Modulation & Coding (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 20 Modulation & Coding (Mode 1) Bit to symbol mapping: group every 4 bits into a symbol Symbol-to-chip mapping: Each 4-bit symbol is mapped to one of 16 32-chip sequence, according to 16-ary Ternary Orthogonal Keying Symbol Repetition: for data rate and range scalability Pulse Genarator: Transmit Ternary pulses at PRF = 24MHz Bit-to- Symbol Symbol Repetition Binary data From PPDU Pulse Generator {0,1,-1} Ternary Sequence Symbol- to-Chip Modulation & Coding (Mode 2) : Feb 2005 Francois Chin (I2R), et. al. Slide 21 Modulation & Coding (Mode 2) Bit to symbol mapping: group every 4 bits into a symbol Symbol-to-chip mapping: Each 4-bit symbol is mapped to one of 16 32-chip sequence, according to 16-ary Ternary Orthogonal Keying Symbol Repetition: for data rate and range scalability Ternary to Binary conversion: (-1/+1 → 1,0 → -1) Pulse Genarator: Transmit bipolar pulses at PRF = 24MHz Bit-to- Symbol Symbol Repetition Binary data From PPDU Ternary- Binary {0,1,-1} Ternary Sequence Symbol- to-Chip Pulse Generator {1,-1} Binary Sequence Auto Correlation Properties for Non-Coherent Symbol Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 22 Auto Correlation Properties for Non-Coherent Symbol Detection Receiver Cross Correlation Properties for Coherent Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 23 Cross Correlation Properties for Coherent Detection Receiver TxSeqSet * RxSeqSet' (Mode 2) = TxSeqSet * RxSeqSet' (Mode 1) = Differential Multipath Combining : Feb 2005 Francois Chin (I2R), et. al. Slide 24 Differential Multipath Combining Auto Correlation Properties for Differential Chip Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 25 Auto Correlation Properties for Differential Chip Detection Receiver Cross Correlation Properties for Differential Chip Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 26 Cross Correlation Properties for Differential Chip Detection Receiver DifferentialChip(TxSeqSet) * DifferentialChip(RxSeqSet)’ (Mode 1) = DifferentialChip(TxSeqSet) * DifferentialChip(RxSeqSet)’ (Mode 2) = Non-Coherent Receiver Architectures (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 27 Energy detection technique rather than coherent receiver, for low cost, low complexity Soft chip values gives best results Oversampling & sequence correlation is used to recovery chip timing recovery Synchronization fully re-acquired for each new packet received (=> no very accurate timebase needed) BPF ( )2 LPF / integrator ADC Sample Rate 1/Tc Soft Despread Non-Coherent Receiver Architectures (Mode 1) Auto Correlation Properties for Energy Detection Receiver (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 28 Auto Correlation Properties for Energy Detection Receiver (Mode 1) Cross Correlation Properties for Energy Detection Receiver (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 29 Cross Correlation Properties for Energy Detection Receiver (Mode 1) TxSeqSet * RxSeqSet ' = Slide 30: Feb 2005 Francois Chin (I2R), et. al. Slide 30 AWGN Performance Slide 31: Feb 2005 Francois Chin (I2R), et. al. Slide 31 AWGN Performance AWGN performance @ 1% PER Basic Data Rate Throughput (Low Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 32 Basic Data Rate Throughput (Low Rate Modes) Useful data rate calculation for 32 byte PSDU (Xo = 0.75 Mbps) Symbol Period = 1.33us Data frame time : 38 x 8 / 0.75= 405.3 µsec ACK frame time : 11 x 8 / 0.75 = 117.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 1355 µsec Useful Basic Data Rate = 189.0 kbps Basic Data Rate Throughput (High Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 33 Basic Data Rate Throughput (High Rate Modes) Useful data rate calculation for 32 byte PSDU (Xo = 3 Mbps) Symbol Period = 1.33us Data frame time : 38 x 8 / 3 = 101.3 µsec ACK frame time : 11 x 8 / 3 = 29.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 963 µsec Useful Basic Data Rate = 265.9 kbps Basic Data Rate Throughput (High Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 34 Basic Data Rate Throughput (High Rate Modes) Useful data rate calculation for 127 byte PSDU (Xo = 3 Mbps) Symbol Period = 1.33us Data frame time : 127 x 8 / 3 = 354.7 µsec ACK frame time : 11 x 8 / 3 = 29.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 1216 µsec Useful Basic Data Rate = 853.5 kbps Slide 35: Feb 2005 Francois Chin (I2R), et. al. Slide 35 Link Budget Ranging and Positioning : Feb 2005 Francois Chin (I2R), et. al. Slide 36 Ranging and Positioning Asynchronous Ranging Scheme : Feb 2005 Francois Chin (I2R), et. al. Slide 37 Asynchronous Ranging Scheme Synchronous ranging One way ranging Simple TOA/TDOA measurement Universal external clock Asynchronous ranging Two way ranging TOA/TDOA measurement by RTTs Half-duplex type of signal exchange TOF : Time Of Flight RTT : Round Trip Time SHR : Synchronization Header Synchronous Ranging Asynchronous Ranging But, High Complexity Features- Sequential two-way ranging is executed via relay transmissions- PAN coordinator manages the overall schedule for positioning- Inactive mode processing is required along the positioning- PAN coordinator may transfer all sorts of information such as observed - TDOAs to a processing unit (PU) for position calculationBenefits- It does not need pre-synchronization among the devices- Positioning in mobile environment is partly accomplished : Feb 2005 Francois Chin (I2R), et. al. Slide 38 Features- Sequential two-way ranging is executed via relay transmissions- PAN coordinator manages the overall schedule for positioning- Inactive mode processing is required along the positioning- PAN coordinator may transfer all sorts of information such as observed - TDOAs to a processing unit (PU) for position calculationBenefits- It does not need pre-synchronization among the devices- Positioning in mobile environment is partly accomplished PAN coordinator P_FFD1 P_FFD2 P_FFD3 RFD TOA 14 TOA 24 TOA 34 P_FFD : Positioning Full Function Device RFD : Reduced Function Device PU Proposed Positioning Scheme Process of Proposed Positioning Scheme : Feb 2005 Francois Chin (I2R), et. al. Slide 39 Process of Proposed Positioning Scheme TOA measurement More Details for obtaining TDOAs : Feb 2005 Francois Chin (I2R), et. al. Slide 40 More Details for obtaining TDOAs Distances among the positioning FFDs are calculated from RTT measurements and known time interval T Using observed RTT measurements and calculated distances, TOAs/TDOAs are updated T12 = (RTT12 – T)/2 T23 = (RTT23 – T)/2 T13 = (RTT13 – T12 – T23 – 2T) RTT34 = T34 + T + T34 RTT14 = T12 + T + T23 + T + T34 + T + T14 RTT24 = T23 + T + T34 + T + T24 TOA14 = (RTT14 - T12 - T23 - TOA34 - 3T) TOA34 = (RTT34 - T)/2 TOA24 = (RTT24 - T23 - TOA34 - 2T) TDOA12 = TOA14 – TOA24 TDOA23 = TOA24 – TOA34 Position Calculation using TDOAs : Feb 2005 Francois Chin (I2R), et. al. Slide 41 Position Calculation using TDOAs The range difference measurement defines a hyperboloid of constant range difference When multiple range difference measurements are obtained, producing multiple hyperboloids, the position location of the device is at the intersection among the hyperboloids Positioning Scenario Overview : Feb 2005 Francois Chin (I2R), et. al. Slide 42 Positioning Scenario Overview Cluster 1 Cluster 1 Case 1 Case 2 Using static reference nodes in relatively large scaled cluster : Power control is required Power consumption increases All devices in cluster must be in inactive data transmission mode Using static and dynamic nodes in overlapped small scaled sub-clusters : Sequential positioning is executed in each sub-cluster Low power consumption Associated sub-cluster in positioning mode should be in inactive data transmission mode Positioning Scenario for Star topology : Feb 2005 Francois Chin (I2R), et. al. Slide 43 Positioning Scenario for Star topology Star topology PAN coordinator activated mode Positioning all devices Re-alignment of positioning FFD’s list is not required Target device activated mode Positioning is requested from some device Re-alignment of positioning FFD’s list is required Positioning Scenario for Cluster-tree Topology : Feb 2005 Francois Chin (I2R), et. al. Slide 44 Positioning Scenario for Cluster-tree Topology Cluster-tree topology Analog Energy Window Bank : Feb 2005 Francois Chin (I2R), et. al. Slide 45 Analog Energy Window Bank Ranging Accuracy Improvement : Feb 2005 Francois Chin (I2R), et. al. Slide 46 Ranging Accuracy Improvement Technical requirement for positioning “It can be related to precise (tens of centimeters) localization in some cases, but is generally limited to about one meter ” Parameters for technical requirement Minimum required pulse duration : Minimum required clock speed for the correlator in the conventional coherent systems Fast ADC clock speed in the conventional coherent receiver is required for the digital signal processing High Cost ! Analog Energy Window Bank (1) : Feb 2005 Francois Chin (I2R), et. al. Slide 47 Analog Energy Window Bank (1) Digital signal processing with fast clock can be replaced by using analog energy window bank with low clock speed Why analog energy window bank? Conventional single energy window may support the energy detection for data demodulation in the operation mode However, this cannot guarantee the correct searching of the signal position in the timing mode (that also means the ambiguity of ranging accuracy) Analog energy window bank can sufficiently support timing and calibration as well as operation mode Widow Bank Size : ~4 nsec (smallest pulse duration) The number of energy windows in a bank : 11 Operation clock speed of each energy window : 24 MHz Number of the required energy windows depends on the power delay profile of the multipath channel (effective multipath components) Analog Energy Window Bank (2) : Feb 2005 Francois Chin (I2R), et. al. Slide 48 Analog Energy Window Bank (2) Modifying MAC : Feb 2005 Francois Chin (I2R), et. al. Slide 49 Modifying MAC Modifications of MAC Command Frame (1) : Feb 2005 Francois Chin (I2R), et. al. Slide 50 Modifications of MAC Command Frame (1) Features Frame control field frame type : positioning (new addition using a reserved bit) Command frame identifier field Positioning request/response (new addition) Positioning parameter information field Absolute coordinates of positioning FFDs POS range List of positioning FFDs and target devices Power control Pre-determined processing time (T) Modifications of MAC Command Frame (2) : Feb 2005 Francois Chin (I2R), et. al. Slide 51 Modifications of MAC Command Frame (2) Frame Control Command frame identifier Positioning parameter Summary : Feb 2005 Francois Chin (I2R), et. al. Slide 52 Summary The proposed system: Impulse-radio based system coupled with a Common ternary signaling allows operation among different classes of nodes / type of receivers, with varying cost / power / performance trade-off Has Band Plan based on multiple 500+MHz bands Is robust against SOP interference Is robust against other in-band interference You do not have the permission to view this presentation. In order to view it, please contact the author of the presentation.
15-05-0113-02-004a-merged-uwb-proposal-i srikrishna.hegde Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT lite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 55 Category: Entertainment License: All Rights Reserved Like it (0) Dislike it (0) Added: October 13, 2009 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript Slide 1: Feb 2005 Francois Chin (I2R), et. al. Slide 1 Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [Merged UWB proposal for IEEE 802.15.4a Alt-PHY] Date Submitted: [22 Feb 2005] Source: [Francois Chin, et.al.] Company: [Institute for Infocomm Research, Singapore] Address: [21 Heng Mui Keng Terrace, Singapore 119613] Voice: [65-68745687] FAX: [65-67744990] E-Mail: [chinfrancois@i2r.a-star.edu.sg] Re: [Response to the call for proposal of IEEE 802.15.4a, Doc Number: 15-04-0380-02-004a ] Abstract: [Merged Proposal to IEEE 802.15.4a Task Group] Purpose: [For presentation and consideration by the IEEE802.15.4a committee] Notice: This document has been prepared to assist the IEEE P802.15. 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 P802.15. This contribution is a technical merger between*: : Feb 2005 Francois Chin (I2R), et. al. Slide 2 This contribution is a technical merger between*: Institute for Infocomm Research [05/032] General Atomics [05/016] Thales & Cellonics [05/008] KERI & SSU & KWU [05/033] Create-Net & China UWB Forum [05/019] Staccato Communications [04/0704] Wisair [05/09] * For a complete list of authors, please see page 3. Authors : Feb 2005 Francois Chin (I2R), et. al. Slide 3 Authors Institute for Infocomm Research: Francois Chin, Xiaoming Peng, Sam Kwok, Zhongding Lei, Kannan, Yong-Huat Chew, Chin-Choy Chai, Rahim, Manjeet, T.T. Tjhung, Hongyi Fu, Tung-Chong Wong General Atomics: Naiel Askar, Susan Lin Thales & Cellonics: Serge Hethuin, Isabelle Bucaille, Arnaud Tonnerre, Fabrice Legrand, Joe Jurianto KERI & SSU & KWU: Kwan-Ho Kim, Sungsoo Choi, Youngjin Park, Hui-Myoung Oh, Yoan Shin, Won cheol Lee, and Ho-In Jeon Create-Net & China UWB Forum: Zheng Zhou, Frank Zheng, Honggang Zhang, Xiaofei Zhou, Iacopo Carreras, Sandro Pera, Imrich Chlamtac Staccato Communications: Roberto Aiello, Torbjorn Larsson Wisair: Gadi Shor, Sorin Goldenberg Slide 4: Feb 2005 Francois Chin (I2R), et. al. Slide 4 Multiband Ternary Orthogonal Keying (M-TOK) for IEEE 802.15.4a UWB based Alt-PHY Goals : Feb 2005 Francois Chin (I2R), et. al. Slide 5 Goals Good use of UWB unlicensed spectrum Good system design Path to low complexity CMOS design Path to low power consumption Scalable to future standards Graceful co-existence with other services Graceful co-existence with other UWB systems Support different classes of nodes, with different reliability requirements (and $), with single common transmit signaling Main Features : Feb 2005 Francois Chin (I2R), et. al. Slide 6 Main Features Proposal main features: Impulse-radio based (pulse-shape independent) Common preamble signaling for different classes of nodes / type of receivers (coherent / differential / noncoherent) Band Plan based on multiple 500 MHz bands Robustness against SOP interference Robustness against other in-band interference Scalability to trade-off complexity/performance Slide 7: Feb 2005 Francois Chin (I2R), et. al. Slide 7 Proposed System Parameters System Description : Feb 2005 Francois Chin (I2R), et. al. Slide 8 System Description Each piconet uses one set of code sequences for different classes of nodes / type of receivers (coherent / differential / non-coherent receivers) 16 Orthogonal Sequences of code length 32 to represent a 4-bit symbol PRF (chip rate): 24 MHz Low enough to avoid significant interchip interference (ICI) with all 802.15.4a multipath models High enough to ensure low pulse peak power FEC: optional (or low complexity type) Band Plan : Feb 2005 Francois Chin (I2R), et. al. Slide 9 Band Plan Multiple access : Feb 2005 Francois Chin (I2R), et. al. Slide 10 Multiple access Multiple access within piconet: TDMA+CSMA/CA same as 15.4 Multiple access across piconets: CDM + FDM Different Piconet uses different Base Sequence & different 500 MHz band Types of Receivers Supported : Feb 2005 Francois Chin (I2R), et. al. Slide 11 Types of Receivers Supported Coherent Detection: The phase of the received carrier waveform is known, and utilized for demodulation Differential Chip Detection: The carrier phase of the previous signaling interval is used as phase reference for demodulation Non-coherent Detection: The carrier phase information (e.g.pulse polarity) is unknown at the receiver Slide 12: Feb 2005 Francois Chin (I2R), et. al. Slide 12 Criteria of Code Sequence Design The sequence Set should have orthogonal (or near orthogonal) cross correlation properties to minimise symbol decision error for all the below receivers For coherent receiver For differential chip receiver For non-coherent symbol detection receiver Energy detection receiver Each sequence should have good auto-correlation properties Slide 13: Feb 2005 Francois Chin (I2R), et. al. Slide 13 To minimise impact of DC noise effect on energy collector based non-coherent receiver For OOK signaling, the transmitter transmits {+1,-1,0} ternary sequences Conventional receive unipolar code sequence – follows transmit sequence After the energy capture in the receiver, the noise has positive DC components in each chip; error occurs in thresholding, especially at lower SNR This will accumulate noise unevenly in symbol decision An ideal receive despreading chip sequence should then have bipolar chip values, preferrably with equal number of ‘+1 and ‘-1’ chips This, to certain extent, will nullify DC noise energy in symbol decision This, will also nullify energy components from other interfering piconets Criteria of Code Sequence Design Slide 14: Feb 2005 Francois Chin (I2R), et. al. Slide 14 Base Sequence Set 31-chip Ternary Sequence set are chosen Only one sequence and one fixed band (no hopping) will be used by all devices in a piconet Logical channels for support of multiple piconets 6 sequences = 6 logical channels (e.g. overlapping piconets) for each FDM Band The same base sequence will be used to construct the symbol-to-chip mapping table Slide 15: Feb 2005 Francois Chin (I2R), et. al. Slide 15 Base Sequence #1 Symbol-to-Chip Mapping: Gray coded 16-ary Ternary Orthogonal Keying To obtain 32-chip per symbol, cyclic shift the Base Sequence first, then append a ‘0’-chip in front Slide 16: Feb 2005 Francois Chin (I2R), et. al. Slide 16 Good Properties of the Mapping Sequence Cyclic nature, leads to simple implementation Zero DC for each sequence No need for carrier phase tracking (i.e. coherent receiver) Slide 17: Feb 2005 Francois Chin (I2R), et. al. Slide 17 Synchronisation Preamble Code sequences has good autocorrelation properties Preamble is constructed by repeating ‘0000’ symbols Long preamble is constructed by further symbol repetition Correlator output for synchronisation Frame Format : Feb 2005 Francois Chin (I2R), et. al. Slide 18 Frame Format PPDU Octets: PHY Layer Preamble 4? 1 Frame Length SFD 1 SHR PHR PSDU MPDU Data: 32 (n=23) Frame Cont. Seq. # Address Data Payload CRC Octets: 2 1 0/4/8 2 MAC Sublayer n MHR MSDU MFR For ACK: 5 (n=0) Transmission Mode : Feb 2005 Francois Chin (I2R), et. al. Slide 19 Transmission Mode Modulation & Coding (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 20 Modulation & Coding (Mode 1) Bit to symbol mapping: group every 4 bits into a symbol Symbol-to-chip mapping: Each 4-bit symbol is mapped to one of 16 32-chip sequence, according to 16-ary Ternary Orthogonal Keying Symbol Repetition: for data rate and range scalability Pulse Genarator: Transmit Ternary pulses at PRF = 24MHz Bit-to- Symbol Symbol Repetition Binary data From PPDU Pulse Generator {0,1,-1} Ternary Sequence Symbol- to-Chip Modulation & Coding (Mode 2) : Feb 2005 Francois Chin (I2R), et. al. Slide 21 Modulation & Coding (Mode 2) Bit to symbol mapping: group every 4 bits into a symbol Symbol-to-chip mapping: Each 4-bit symbol is mapped to one of 16 32-chip sequence, according to 16-ary Ternary Orthogonal Keying Symbol Repetition: for data rate and range scalability Ternary to Binary conversion: (-1/+1 → 1,0 → -1) Pulse Genarator: Transmit bipolar pulses at PRF = 24MHz Bit-to- Symbol Symbol Repetition Binary data From PPDU Ternary- Binary {0,1,-1} Ternary Sequence Symbol- to-Chip Pulse Generator {1,-1} Binary Sequence Auto Correlation Properties for Non-Coherent Symbol Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 22 Auto Correlation Properties for Non-Coherent Symbol Detection Receiver Cross Correlation Properties for Coherent Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 23 Cross Correlation Properties for Coherent Detection Receiver TxSeqSet * RxSeqSet' (Mode 2) = TxSeqSet * RxSeqSet' (Mode 1) = Differential Multipath Combining : Feb 2005 Francois Chin (I2R), et. al. Slide 24 Differential Multipath Combining Auto Correlation Properties for Differential Chip Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 25 Auto Correlation Properties for Differential Chip Detection Receiver Cross Correlation Properties for Differential Chip Detection Receiver : Feb 2005 Francois Chin (I2R), et. al. Slide 26 Cross Correlation Properties for Differential Chip Detection Receiver DifferentialChip(TxSeqSet) * DifferentialChip(RxSeqSet)’ (Mode 1) = DifferentialChip(TxSeqSet) * DifferentialChip(RxSeqSet)’ (Mode 2) = Non-Coherent Receiver Architectures (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 27 Energy detection technique rather than coherent receiver, for low cost, low complexity Soft chip values gives best results Oversampling & sequence correlation is used to recovery chip timing recovery Synchronization fully re-acquired for each new packet received (=> no very accurate timebase needed) BPF ( )2 LPF / integrator ADC Sample Rate 1/Tc Soft Despread Non-Coherent Receiver Architectures (Mode 1) Auto Correlation Properties for Energy Detection Receiver (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 28 Auto Correlation Properties for Energy Detection Receiver (Mode 1) Cross Correlation Properties for Energy Detection Receiver (Mode 1) : Feb 2005 Francois Chin (I2R), et. al. Slide 29 Cross Correlation Properties for Energy Detection Receiver (Mode 1) TxSeqSet * RxSeqSet ' = Slide 30: Feb 2005 Francois Chin (I2R), et. al. Slide 30 AWGN Performance Slide 31: Feb 2005 Francois Chin (I2R), et. al. Slide 31 AWGN Performance AWGN performance @ 1% PER Basic Data Rate Throughput (Low Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 32 Basic Data Rate Throughput (Low Rate Modes) Useful data rate calculation for 32 byte PSDU (Xo = 0.75 Mbps) Symbol Period = 1.33us Data frame time : 38 x 8 / 0.75= 405.3 µsec ACK frame time : 11 x 8 / 0.75 = 117.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 1355 µsec Useful Basic Data Rate = 189.0 kbps Basic Data Rate Throughput (High Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 33 Basic Data Rate Throughput (High Rate Modes) Useful data rate calculation for 32 byte PSDU (Xo = 3 Mbps) Symbol Period = 1.33us Data frame time : 38 x 8 / 3 = 101.3 µsec ACK frame time : 11 x 8 / 3 = 29.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 963 µsec Useful Basic Data Rate = 265.9 kbps Basic Data Rate Throughput (High Rate Modes) : Feb 2005 Francois Chin (I2R), et. al. Slide 34 Basic Data Rate Throughput (High Rate Modes) Useful data rate calculation for 127 byte PSDU (Xo = 3 Mbps) Symbol Period = 1.33us Data frame time : 127 x 8 / 3 = 354.7 µsec ACK frame time : 11 x 8 / 3 = 29.3 µsec tACK (considering 15.4 spec) : 192 µsec LIFS (considering 15.4 spec) : 640 µsec Tframe = 1216 µsec Useful Basic Data Rate = 853.5 kbps Slide 35: Feb 2005 Francois Chin (I2R), et. al. Slide 35 Link Budget Ranging and Positioning : Feb 2005 Francois Chin (I2R), et. al. Slide 36 Ranging and Positioning Asynchronous Ranging Scheme : Feb 2005 Francois Chin (I2R), et. al. Slide 37 Asynchronous Ranging Scheme Synchronous ranging One way ranging Simple TOA/TDOA measurement Universal external clock Asynchronous ranging Two way ranging TOA/TDOA measurement by RTTs Half-duplex type of signal exchange TOF : Time Of Flight RTT : Round Trip Time SHR : Synchronization Header Synchronous Ranging Asynchronous Ranging But, High Complexity Features- Sequential two-way ranging is executed via relay transmissions- PAN coordinator manages the overall schedule for positioning- Inactive mode processing is required along the positioning- PAN coordinator may transfer all sorts of information such as observed - TDOAs to a processing unit (PU) for position calculationBenefits- It does not need pre-synchronization among the devices- Positioning in mobile environment is partly accomplished : Feb 2005 Francois Chin (I2R), et. al. Slide 38 Features- Sequential two-way ranging is executed via relay transmissions- PAN coordinator manages the overall schedule for positioning- Inactive mode processing is required along the positioning- PAN coordinator may transfer all sorts of information such as observed - TDOAs to a processing unit (PU) for position calculationBenefits- It does not need pre-synchronization among the devices- Positioning in mobile environment is partly accomplished PAN coordinator P_FFD1 P_FFD2 P_FFD3 RFD TOA 14 TOA 24 TOA 34 P_FFD : Positioning Full Function Device RFD : Reduced Function Device PU Proposed Positioning Scheme Process of Proposed Positioning Scheme : Feb 2005 Francois Chin (I2R), et. al. Slide 39 Process of Proposed Positioning Scheme TOA measurement More Details for obtaining TDOAs : Feb 2005 Francois Chin (I2R), et. al. Slide 40 More Details for obtaining TDOAs Distances among the positioning FFDs are calculated from RTT measurements and known time interval T Using observed RTT measurements and calculated distances, TOAs/TDOAs are updated T12 = (RTT12 – T)/2 T23 = (RTT23 – T)/2 T13 = (RTT13 – T12 – T23 – 2T) RTT34 = T34 + T + T34 RTT14 = T12 + T + T23 + T + T34 + T + T14 RTT24 = T23 + T + T34 + T + T24 TOA14 = (RTT14 - T12 - T23 - TOA34 - 3T) TOA34 = (RTT34 - T)/2 TOA24 = (RTT24 - T23 - TOA34 - 2T) TDOA12 = TOA14 – TOA24 TDOA23 = TOA24 – TOA34 Position Calculation using TDOAs : Feb 2005 Francois Chin (I2R), et. al. Slide 41 Position Calculation using TDOAs The range difference measurement defines a hyperboloid of constant range difference When multiple range difference measurements are obtained, producing multiple hyperboloids, the position location of the device is at the intersection among the hyperboloids Positioning Scenario Overview : Feb 2005 Francois Chin (I2R), et. al. Slide 42 Positioning Scenario Overview Cluster 1 Cluster 1 Case 1 Case 2 Using static reference nodes in relatively large scaled cluster : Power control is required Power consumption increases All devices in cluster must be in inactive data transmission mode Using static and dynamic nodes in overlapped small scaled sub-clusters : Sequential positioning is executed in each sub-cluster Low power consumption Associated sub-cluster in positioning mode should be in inactive data transmission mode Positioning Scenario for Star topology : Feb 2005 Francois Chin (I2R), et. al. Slide 43 Positioning Scenario for Star topology Star topology PAN coordinator activated mode Positioning all devices Re-alignment of positioning FFD’s list is not required Target device activated mode Positioning is requested from some device Re-alignment of positioning FFD’s list is required Positioning Scenario for Cluster-tree Topology : Feb 2005 Francois Chin (I2R), et. al. Slide 44 Positioning Scenario for Cluster-tree Topology Cluster-tree topology Analog Energy Window Bank : Feb 2005 Francois Chin (I2R), et. al. Slide 45 Analog Energy Window Bank Ranging Accuracy Improvement : Feb 2005 Francois Chin (I2R), et. al. Slide 46 Ranging Accuracy Improvement Technical requirement for positioning “It can be related to precise (tens of centimeters) localization in some cases, but is generally limited to about one meter ” Parameters for technical requirement Minimum required pulse duration : Minimum required clock speed for the correlator in the conventional coherent systems Fast ADC clock speed in the conventional coherent receiver is required for the digital signal processing High Cost ! Analog Energy Window Bank (1) : Feb 2005 Francois Chin (I2R), et. al. Slide 47 Analog Energy Window Bank (1) Digital signal processing with fast clock can be replaced by using analog energy window bank with low clock speed Why analog energy window bank? Conventional single energy window may support the energy detection for data demodulation in the operation mode However, this cannot guarantee the correct searching of the signal position in the timing mode (that also means the ambiguity of ranging accuracy) Analog energy window bank can sufficiently support timing and calibration as well as operation mode Widow Bank Size : ~4 nsec (smallest pulse duration) The number of energy windows in a bank : 11 Operation clock speed of each energy window : 24 MHz Number of the required energy windows depends on the power delay profile of the multipath channel (effective multipath components) Analog Energy Window Bank (2) : Feb 2005 Francois Chin (I2R), et. al. Slide 48 Analog Energy Window Bank (2) Modifying MAC : Feb 2005 Francois Chin (I2R), et. al. Slide 49 Modifying MAC Modifications of MAC Command Frame (1) : Feb 2005 Francois Chin (I2R), et. al. Slide 50 Modifications of MAC Command Frame (1) Features Frame control field frame type : positioning (new addition using a reserved bit) Command frame identifier field Positioning request/response (new addition) Positioning parameter information field Absolute coordinates of positioning FFDs POS range List of positioning FFDs and target devices Power control Pre-determined processing time (T) Modifications of MAC Command Frame (2) : Feb 2005 Francois Chin (I2R), et. al. Slide 51 Modifications of MAC Command Frame (2) Frame Control Command frame identifier Positioning parameter Summary : Feb 2005 Francois Chin (I2R), et. al. Slide 52 Summary The proposed system: Impulse-radio based system coupled with a Common ternary signaling allows operation among different classes of nodes / type of receivers, with varying cost / power / performance trade-off Has Band Plan based on multiple 500+MHz bands Is robust against SOP interference Is robust against other in-band interference