questions about "Experiments on local positioning with Bluetooth"

1 view
Skip to first unread message

jun shen

unread,
Nov 10, 2007, 11:38:24 PM11/10/07
to ASU:CSE535 FALL 07 Mobile Computing
Hi Sushma,
Here are some of my questions about the paper.
Q1:in formula (4), what is the exact meaning of Zk? Is that the
observation value of state k?
Q2: in formula (7), why only using 1st order term of Taylor series?
if we use more terms, will that promise a more accurate value?
Q3: I dont understand how to obtain formula(9)? Can you help explain
that?

one general questions:
Q1: in most cases the scope of bluetooth(bt) is only around
10meters, the error range of 3.76meter seems a significant variation,
is the localization meaningful?

Sushma Myneni

unread,
Nov 14, 2007, 12:57:30 AM11/14/07
to asu_cse...@googlegroups.com
Hi,
Q1:in formula (4), what is the exact meaning of Zk? Is that the
observation value of state k?

--> Yes, Zk represents the measurement state of the true state xk


 Q2: in formula (7), why only using 1st order term of Taylor series?
if we use more terms, will that promise a more accurate value?

---> Using first order gives more accurate values for a straight lines, where as for curves the second order gives better results


 Q3: I dont understand how to obtain formula(9)? Can you help explain
that?
 ---> We have to linearize the h(x) matrix using Taylor series and remove the higher order terms ( power greater than 1)


 one general questions:
  Q1: in most cases the scope of bluetooth(bt) is only around
10meters, the error range of 3.76meter seems a significant variation,
is the localization meaningful?
 --->Yes, error in positioning is too high, and as I have mentioned in today's class we need to improve the distance estimation metric. That is use RX power levels rather than trying to estimate them from RSSI values.

Thankyou,
Sushma
Reply all
Reply to author
Forward
0 new messages