CSP

1
Project IEEE 802.15 Working Group for Wireless
Personal Area Networks (WPANs) Submission Title
Requirements for a UWB Common Signaling
Protocol Date Submitted 23 February,
2004 Source Yasaman Bahreini, John Santhoff,
Kai Siwiak, Ismail Lakkis Company1
PulseLINK Company2 Wideband Access,
Address1 1969 Kellogg Ave. Carlsbad, CA 92008,
Address 2 12396 World Trade Ave. Suite 117, San
Diego, CA 92128, Voice 1(760) 607-0844, FAX
1 [phone], E-Mails 1ybahreini_at_ieee.org
, jsanthoff_at_pulselink.net, k.siwiak_at_ieee.org
Voice 2(858) 618-1930, FAX 2 (858)
618-1980, E-Mail 2ilakkis_at_widebandaccess.com R
e Ad hoc Meeting Submission Abstract At
least one task group has chosen a UWB PHY, and
another group in P802.15 is considering UWB PHY
that will operate in common spectrum. This
presentation focuses on potential coexistence
issues of multiple UWB PHY layers in a common
frequency band. Purpose Focus and attention
need to be directed to P802.15 and discussions
started on issues effecting coexistence of
multiple UWB PHY layers using common spectrum. It
is early enough in the standards draft process to
consider preemptive measures to ensure
coexistence. 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.
2
Coexistence of Multiple UWB Physical Layers ?

• Allowing Many Flavors of UWB Signaling to Coexist

3
Outline

• Coexistence picture today
• How do we address UWB Coexistence?
• A Common Signaling Protocol (CSP)
• Design goals for a CSP
• Proposed options for CSP
• Summary comparison of proposed CSP options

4
Coexistence Picture Today

• Currently only one UWB standard in draft
  P802.15.3a
• Today concern is UWB transmitter to Narrow-Band
  receiver.
• Coexistence, however, is the ability of one
  system to perform a task in a given shared
  environment where other systems that may or may
  not be using the same set of rules. IEEE
  802.15.2 definition of coexistence,
  http//grouper.ieee.org/groups/802/15/pub/2000/
  Sep00/99134r2P802-15_TG2-CoexistenceInteroperabili
  tyandOtherTerms.ppt
• In the works
• P802.15.4SG4a might consider a UWB solution
• P802.11 are looking for solutions which might
  involve UWB
• We need a pre-emptive action to ensure the
  orderly introduction of various UWB PHYs

5
How do we Solve the Problem?

• We need an Etiquette to manage peaceful
  coexistence of different UWB PHY layers
• Todays action set the path for UWB evolution for
  decades.
• A framework is needed that addresses guidelines
  on what spectrum is accessed when
• A common signaling protocol can act as such an
  arbitrator

6
Why a Common Signaling Protocol?

• We have a Once-in-a Lifetime opportunity to
  define an emerging wireless standard on a
  potentially global scale
• Lets use this opportunity to address
  interoperability and coexistence as a part of the
  standard instead of as an after thought
• Example of the recent past
• the unlicensed ISM bands have experienced
  explosive growth with multiple PHY layer
  interfaces defined
• In the ISM 2.4 GHz band, there are no less than
  five different PHY standards competing for
  coexistence in the same spectrum. (802.11b,
  802.15.3, 802.15.4, Bluetooth and Cordless
  Phones)
• UWB is gaining momentum
• It is likely that multiple UWB based PHY layers
  will emerge
• Already in 802.15.3a potential in 802.15.4SG4a

7
What is a Common Signaling Protocol?

• A UWB operational mode understandable to all UWB
  air interfaces
• Methodology for allowing multiple different UWB
  PHY layers to coexist in
• the same spectrum bands
• same coverage areas
• Uses cooperative management of allocated PHY
  resources of time and frequency

8
Design Philosophy

• Common mechanism that is pro-active rather than
  re-active
• Uses a pre-defined framework potentially allowing
  fair UWB PHY layer resource allocation
• Collaborative
• Collaborated TDMA /FDMA techniques to allow for
  alternate transmissions among different UWB
  standards
• Collaborated techniques for managing packet
  transmission based on channel monitoring
• Non-Collaborative
• Adaptive packet (time/frequency) selection and
  scheduling

9
Design Goals

• Primary
• Address coexistence - interoperability among
  differing UWB PHYs
• Low cost so minimal PHY is not burdened
• Provide discovery beacon
• Provide coarse SYNCH and frequency acquisition,
  protocol selection
• Secondary
• Provide diversity, fine SYNCH and frequency
  acquisition, AGC, channel estimation
• Basis for a low rate long range PHY
• Enable simpler geo-positioning capabilities
• Provide mechanism for sleep mode, and wake up for
  power conservation

10
PHY CSP Preamble Requirements

• Introduce a mandatory/standard preamble before
  different UWB-based standards are ratified
• With the simplest modulation that is recognizable
  by all UWB systems, and in the frequency band
  that most likely would be used by all UWB devices
• Simple to detect and lock-to for both HDR or LDR
  systems
• Provides a type indicator to identify UWB
  protocols
• This CSP preamble would be used for AGC, Antenna
  Diversity, DC Offset Removal, Energy Detection,
  Signal Detection, Coarse Timing and Frequency
  Estimation. Different UWB standard-specific
  preamble fields (responsible for 2nd AGC, and
  Channel Estimation, and Fine Timing and Frequency
  Estimation) can be appended to this standard
  preamble.

11
CSP MAC Requirements

• Time-slot Allocation
• Scheduling of different PHY packets
• Provide some provisions for QoS
• ALL Min. Changes to MAC

12
Option 1

• CSP Preamble sent as in MB-OFDM proposal
• BPSK at Symbol Rate of 3.2 Msps
• Time-frequency interleaved over 3-bands in freq.
  range 3.17-4.75 GHz
• Packet Synch Sequence Hierarchical
• Every PNC is assigned to one set of Hierarchical
  sequences and a time-freq. combination
• A non-hopping LDR DEV sees a symbol rate of 1.066
  Msps.

13
Option 1 PNC to DEV Communications

• PNC - MB-OFDM DEV
• As currently specified
• PNC - DS-UWB DEV
• In order to operate need
• Rate conversion unit (polyphase filtering) to
  interpolate/decimate from 1.368 GSps to 528 MSps
  OR a frequency synthesizer capable of generating
  528 MHz.
• Loss of received RF energy at a portion of f1,
  and f3 bands due to mismatch of proposed front
  end filter requirements and band(s) of operation.
  Increased probability of false alarm.
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• PNC - LDR UWB DEV
• Can be supported via non-coherent simple receiver
  (delay multiply, integrate dump). Data rate
  1 Mbps or less via.
• Needs longer preamble for robust acquisition
  (extend the preamble for LDR mode)

14
Option 1 DEV to PNC Communications

• MB-OFDM DEV --- PNC
• As currently specified
• DS-UWB DEV --- PNC
• In order to operate need
• Rate Conversion
• Rate conversion unit (poly-phase) to interpolate
  528 MHz to 1.368 GHz OR a frequency synthesizer
  capable of generating 528 MHz.
• Need DAC to generate MB-OFDM preamble in the
  three bands of operation
• No Rate Conversion
• No need for rate conversion. Need to generate
  optimal Ternary Code set that resembles the
  closest to MB-OFDM Hierarchical SYNCH Sequences
  (will not be optimal)
• Loss of received RF energy at a portion of f1,
  and f3 bands due to mismatch of proposed front
  end filter requirements and band(s) of operation.
  Increased probability of false alarm.
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• LDR UWB DEV --- PNC
• Can not be supported. High-speed DAC is required.

15
Option 2

• CSP Preamble sent as in Option 1 AND
• DS-UWB Chip Rate is changed to 1.584 Gcps and
  hence all the relevant parameters adjusted to
  match this rate
• No change to DS-UWB filter requirements

16
Option 2 PNC to DEV Communications

• PNC - MB-OFDM DEV
• As currently specified
• PNC - DS-UWB DEV
• In order to operate need
• Loss of received RF energy at a portion of f1,
  and f3 bands due to mismatch of proposed front
  end filter requirements and band(s) of operation.
  Increased probability of false alarm.
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• PNC - LDR UWB DEV
• Can be supported via non-coherent simple receiver
  (delay multiply, integrate dump). Data rate
  1 Mbps or less via.
• Needs longer preamble for robust acquisition
  (extend the preamble for LDR mode)

17
Option 2 DEV to PNC Communications

• MB-OFDM DEV --- PNC
• As currently specified
• DS-UWB DEV --- PNC
• In order to operate need
• Need DAC to generate MB-OFDM preamble in bands
  f1, and f3 (due to rotation wrt f2)
• Loss of received RF energy at a portion of f1,
  and f3 bands due to mismatch of proposed front
  end filter requirements and band(s) of operation.
  Increased probability of false alarm.
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• LDR UWB DEV --- PNC
• Can not be supported. High-speed DAC is required.

18
Option 3

• CSP Preamble sent only in the middle band of
  MB-OFDM
• There is no FH, and hence no time-freq.
  combination per PNC
• Center band (f2 3.696 4.224 GHz) is used for
  all PNCs
• Hence different PNCs use different set of
  Hierarchical sequences
• Cyclic prefix, and guard interval can be removed
  hence a better TX/RX power efficiency and better
  probability of detection
• AND DS-UWB Chip Rate is changed to 1.584 Gcps and
  hence all the relevant parameters adjusted to
  match this rate
• No change to DS-UWB filter requirements

19
Option 3 PNC to DEV Communications

• PNC - MB-OFDM DEV
• As currently specified BUT frequency diversity is
  taken away
• Better probability of detection
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• PNC - DS-UWB DEV
• In order to operate need only to adjust front end
  gain settings when moving from the preamble to
  DATA mode
• PNC - LDR UWB DEV
• Can be supported via non-coherent simple receiver
  (delay multiply, integrate dump). Data rate
  1 Mbps or less via.
• Needs longer preamble for robust acquisition
  (extend the preamble for LDR mode)

20
Option 3 DEV to PNC Communications

• MB-OFDM DEV --- PNC
• As currently specified BUT frequency diversity is
  taken away
• Better probability of detection
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• DS-UWB DEV --- PNC
• In order to operate
• Need for adjusting front end gain settings when
  moving from the preamble to DATA mode.
• LDR UWB DEV --- PNC
• May be supported.

21
Option 4 Design Goals

• A LDR beacon that broadcasts around every 10ms
• The HDR modes are not bugged down since this is
  not so frequent
• HDR preambles would be categorized as Initial
  Mode Preamble CSP Preamble and Streaming Mode
  Preambles
• Power save mode (sleep mode)
• Additional hardware. Needs to be low cost.
• Very simple LDR TX/RX architecture

22
Option 4 Specifics

• Modulation as simple as DBPSK.
• Allows simple transceiver architecture. (coherent
  or non-coherent analog correlator receiver)
• Frequency band at least 984 MHz band centered
  around 4.0 GHz
• The base rate needs to be an integer sub-multiple
  of 24 Mbps
• Different PNCs use different codes

23
Option 4 PNC to DEV Communications

• PNC - MB-OFDM DEV
• Minimum Added circuitry
• Coexistence achieved
• PNC - DS-UWB DEV
• Min Added circuitry
• Coexistence achieved
• PNC - LDR UWB DEV
• Can be supported
• Coexistence achieved

24
Option 4 DEV to PNC Communications

• MB-OFDM DEV --- PNC
• Minimum Added circuitry
• Coexistence achieved
• DS-UWB DEV --- PNC
• Minimum Added circuitry
• Coexistence achieved
• LDR UWB DEV --- PNC
• Can be supported
• Coexistence achieved

25
Summary

• Coexistence among multiple UWB PHYs is feasible
• We proposed 4 different methods to achieve
  coexistence
• Option 3 and Option 4 seem to be the most
  attractive compromise points