June 2005 IEEE-15-05-0334-00-004a

Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs)

Submission Title: Symetric Double Sided -Two Way Ranging

Date Submitted: []
Source: Rainer Hach
Company: Nanotron Technologies
Address: Alt-Moabit 61, 10555 Berlin, Germany
Voice: +49 30 399 954 207
E-Mail: r.hach@nanotron.com

Re: []

Abstract: Demonstrate the problem of crystal tolerances on Two Way Ranging and suggest a solution

Purpose: Material to be presented and discussed

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.

TG4a Slide 1 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

What’s the problem ?

• TWR as defined in the ranging

subcommittee final report (0581r07) is
vulnerable to timebase mismatch and
thus crystal tolerances.
• We propose that the UWB PHY should
support Symmetric Double Sided-Two
Way Ranging (SDS-TWR).

TG4a Slide 2 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

“Straightforward TWR Model” according to

04-0581r07
Unknown propagation delay tp
Unknown clock offset t0
Message 1
Message 2
Device A Device B
T2 AR  T2 BT  to  t p T1BR  T1 AT  to  t p
Offset between A and B is reflected here!
What about drift?
Two equations in two unknowns yield: Multiple measurements of tp and
tp  1
2  T2 AR  T1 AT   T2 BT  T1BR   to yield finer precision &
accuracy, and allow frequency
offset correction.
to  1
2
 T2 BT  T1BR   T2 AR  T1 AT  
* US Naval Observatory, Telstar Satellite, circa 1962
http://www.boulder.nist.gov/timefreq/time/twoway.htm
Unmatched detect-delays in the two devices may require one-time offset
calibration.

Figure 1 Two-Way Time Transfer Model

TG4a Slide 3 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

Let’s analyze the calculation of the

propagation delay
tp  1
2  T2 AR  T1AT   T2 BT  T1BR  

round trip time measured with clock A reply time measured with clock B

Usually the reply time will be significant larger

than the propagation delay.
Thus round trip time and reply time will be almost equal!

Example: tp=30ns, treply=1ms -> tround=1.000060 ms

TG4a Slide 4 Hach, Nanotron

 June 2005 IEEE-15-05-0334-00-004a

What happens if there is a drift (difference in

clock speed) between clock A and clock B?
• Assume +10 ppm for clock A
and -10 ppm for clock B:
t p  12  T2 AR  T1 AT   T2 BT  T1BR  
big number measured
with clock B
big number measured
with clock A wrong
+10 ppm -10 ppm number!
tp=0.5*(1.000060e-3*(1+10e-6)-1e-3*(1-10e-6)) = 40e-9

Problem: Difference between large numbers (e.g. 1ms) with

different accuracies (+10ns, -10 ns) has high inaccuracy!

TG4a Slide 5 Hach, Nanotron

 June 2005 IEEE-15-05-0334-00-004a

Effect of high inaccuracy after subtracting two large numbers measured with
different clocks can be avoided by using
Symmetric Double Sided-Two Way Ranging (SDS -TWR)

Device A Device B

tp unknown propagation delay

t roundA t replyB  t p reply time

tp

tp
t roundB  t replyA t replyA  t replyB
tp
troundA  treplyA  troundB  treplyB
tp 
4

TG4a Slide 6 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

Let’s analyze the calculation of the

propagation delay for SDS -TWR
two big numbers measured with the two big numbers measured with the
same timebase (clock A) same timebase (clock B)

troundA  treplyA  troundB  treplyB

tp 
4

Using the same example numbers as before yields:

tp=0.25*(1.000060e-3*(1+10e-6)-1e-3*(1+10e-6)+1.000060e-3*(1-10e-6)-1e-3*(1-10e-6))

-> tp= 30e-9

correct
number!

TG4a Slide 7 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

How many message are required for SDS-TWR? At least 3!

Device B Device A Device B

Device A

these two messages

can be combined
into one

TG4a Slide 8 Hach, Nanotron

June 2005 IEEE-15-05-0334-00-004a

Summary
• “Straightforward TWR” requires very low
crystal tolerances << 10 ppm or precise
phase tracking
• SDS-TWR is an enhancement on
“Straightforward TWR” which eliminates the
need of phase tracking
• SDS-TWR needs a minimum of only 3
messages
• SDS-TWR can work with standard tolerances
(up to 40 ppm or more)

TG4a Slide 9 Hach, Nanotron