LYU0401_term1

Document Sample
LYU0401_term1 Powered By Docstoc
					LYU0401 Location-Based
Multimedia Mobile Service

                        Clarence Fung
                              Tilen Ma
     Supervisor: Professor Michael Lyu
          Marker: Professor Alan Liew
                    Outline
   Introduction
   Objective
   Location-Based Service
   Current Localization Methods
   Experimental Study
   Wi-Fi Location System
   Future Work
   Conclusion
                 Introduction
   In this semester, we mainly focus on the
    problem of localization
   We have chosen the 1st floor of the Ho Sin-
    Hang Engineering Building to study the
    problem of localization
   Our goal is to locate a person when he/she is
    walking around on the floor
                   Objective
   To meet the need for Location-Based Service
   To find out if Wireless LAN provide enough
    information for localization in 2D space
   Study on different localization algorithms
   Develop an application in a mobile device
             Location-Based Service
   Localization is necessary for many higher level sensor network
    functions such as tracking, monitoring and geometric-based
    routing

   Three categories:
       Global location systems
       Wide-area location systems
       Indoor location systems

   Systems in indoor environment
       Infrared (IR)
       Ultrasound
       Radio signal
        Wireless LAN (WLAN)-Based
             Positioning System
Advantages over all other systems
 Economical
       WLAN network usually exists already as part of the
        communications infrastructure
   Covers a large area
       Work in a large building or even across many buildings.
   Stable system
       Video- or IR-based location systems are subject to
        restrictions, such as line-of-sight limitations
    Current Localization Methods
   Point-based approach
     goal is to return a single point for the mobile object
     E.g. Simple Distance Matching



   Area-based approach
     goal is to return the possible locations of the mobile
      object as an area rather than a single point
     E.g. Simple-Point Matching, Area-Based Probability
  Area-Based Probability (ABP)
Advantages:
 Presents the user an understanding of the system
  in a more natural and intuitive manner
 High accuracy

 More mathematical approach
             Steps in using ABP
Decide the                 Measure Signals                   Create a
  Areas                    at Different Areas               Training Set



                   Create a                          Measure Signals
                  Testing Set                       at Current Position


                 Find out the
                                                Calculate Probability
             Probability of Being
                                                      Density
              at Different Areas


                             Return the Area with
                              Highest Probability
Applying Area-based Approach
     Some Terms and Definitions
   n Access Points
       AP1, AP2, …, APn

   Training set T0
     Offline measured signal strengths at different
      locations an algorithm uses
     Consists of a set of fingerprints (Si) at m different
      areas Ai
     T0 = ( Ai, Si ), i = 1 … m
     Some Terms and Definitions

   Fingerprints Si
     Set of n signal strengths at Ai, one per each access
      point
     Si = (si1, …, sin), where sij is the expected average
      signal strength from APj
          Generating Training Set
   In one particular Ai, we read a series of signal
    strengths (sijk ) for a particular APj with a
    constant time between samples
       k = 1… oij ,where oij is the number of samples from
        APj at Ai
   We estimate sij by averaging the series, {sij1, sij2…,
    sijo }
           Generating Training Set
   We do the same for all n APs, so we have the
    fingerprints at Ai,
       Si = (si1, …, sin)
   We do the same for all m areas, so we have the
    training set
       T0 = ( Ai, Si ), i = 1… m
                        Collecting Signals
    At each area chosen, we measure the signal strength
     from the access points for 1 minute
        Position    1     2     3     4     5        6        7         8     9     10    11    12
AP MAC address                                  Signal Strength (dBm)
00:02:2d:28:be:9e   -70   -62   -58   -67   -73       -78     -83       -86   -84   -81   -78   -55
00:02:2d:28:be:5d   -67   -59   -60   -71   -76       -79     -81       -86   -81   -83   -79   -52
00:60:1d:1e:43:9b   -79   -87   -85   -84   -89       -80     -76       -77   -66   -63   -77   -90
00:0f:34:f3:60:40   -63   -69   -65   -74   -76       -72     -77       -84   -76   -74   -66   -79
00:02:2d:21:39:1f               -82   -78   -82       -59     -78       -73   -83   -85   -82
00:11:93:3d:6f:c0                     -90   -85       -86     -89       -88
00:11:20:93:65:c0                           -89       -89                                       -90
00:0f:34:bb:df:20                     -89   -90       -82     -88       -88
00:0c:ce:21:1b:9d                                     -87
00:0c:85:35:33:d2                           -88                         -88
00:11:20:93:63:90                                     -89                                       -88
00:0c:85:35:33:d4                                                             -87
00:04:76:a7:ab:a3                                                                               -90
                Data Processing
   We have chosen 7 out of 13 access points
     least contribution to localization
     shorten computation time

   For missing signal strengths, we input -92 dBm
    as entry
                              Training Set
Position            1     2     3     4     5     6     7     8     9     10    11    12

AP MAC address                              Signal Strength (dBm)

00:02:2d:28:be:9e   -70   -62   -58   -67   -73   -78   -83   -86   -84   -81   -78   -55

00:02:2d:28:be:5d   -67   -59   -60   -71   -76   -79   -81   -86   -81   -83   -79   -52

00:60:1d:1e:43:9b   -79   -87   -85   -84   -89   -80   -76   -77   -66   -63   -77   -90

00:0f:34:f3:60:40   -63   -69   -65   -74   -76   -72   -77   -84   -76   -74   -66   -79

00:02:2d:21:39:1f   -92   -92   -82   -78   -82   -59   -78   -73   -83   -85   -82   -92

00:11:93:3d:6f:c0   -92   -92   -92   -90   -85   -86   -89   -88   -92   -92   -92   -92

00:0f:34:bb:df:20   -92   -92   -92   -89   -90   -82   -88   -88   -92   -92   -92   -92
             Getting Testing Set
   The object to be localized collects a set of
    received signal strengths (RSS) when it is at
    certain location
   A testing set (St) is created similar to the
    fingerprints in the training set
   It is a set of average signal strengths from APs,
    St = (st1, …, stn)
                    RSS
AP MAC address        Signal Strength (dBm)

00:02:2d:28:be:9e              -71

00:02:2d:28:be:5d              -72

00:60:1d:1e:43:9b              -89

00:0f:34:f3:60:40              -49
              Testing Set
AP MAC address        Signal Strength (dBm)

00:02:2d:28:be:9e              -71

00:02:2d:28:be:5d              -72

00:60:1d:1e:43:9b              -89

00:0f:34:f3:60:40              -49

00:02:2d:21:39:1f              -92

00:11:93:3d:6f:c0              -92

00:0f:34:bb:df:20              -92
                 Applying ABP
   Goal: return the area with a highest probability

   Approach: compute the likelihood of the testing
    set (St) that matches the fingerprint for each area
    (Si)
                Applying ABP
Assumptions:
 Signal received from different APs are
  independent

   For each APj, j = 1…n, the sequence of RSS sijk,
    k = 1… oij, at each Ai in To is modeled as a
    Gaussian distribution
            Applying Bayes’ rule
   We compute the probability of being at different
    areas Ai, on given the testing set St
    P(Ai |St) = P(St |Ai)* P(Ai)/ P(St)    (1)
        P(St) is a constant
        Assume the object is equally likely to be at any location.
         P(Ai) is a constant
   P(Ai |St) = c*P(St |Ai)                                (2)
             Area Based Probability
   We compute P(St |Ai) for every area Ai ,i=1…m, using the
    Gaussian assumption

   Finding Probability Density
       the object must be at one of the 12 areas
       ΣP(Ai | St) =1 for all i

   Max{P(Ai |St) } = Max{c*P(St |Ai) }
                        = Max{P(St |Ai) }

   Return the area Ai with top probability
            Gaussian Distribution
   In our application, we
    can take μ as the
    expected average signal
    strengths for the access
    point to be calculated

   we take σ as 8.5
     Integral of Normal Function
   Find probability by
    integration
   Take interval as 1
           Error function erf(x)
   Express Integral of
    Normal Function in
    terms of erf
   Approximate value of
    erf by a series
   Choose iteration of 50
                   Experimental Study
   Area 5 is near the
    North-West stairway                              ABP Localization Graph at Area 5
                                           0.7
    on the 1st floor                                                                               Probability at A1

                                           0.6                                                     Probability at A2

                                                                                                   Probability at A3
   deep purple line is on                 0.5                                                     Probability at A4
    the top of other lines
                                                                                                   Probability at A5
                             Probability

                                           0.4
                                                                                                   Probability at A6

   Localization system                                                                            Probability at A7
                                           0.3
    returns the correct                                                                            Probability at A8
    result                                 0.2                                                     Probability at A9

                                                                                                   Probability at A10
                                           0.1                                                     Probability at A11

                                                                                                   Probability at A12
                                            0
                                                 1   5   9   13   17 21 25     29   33   37   41
                                                                   Sample No
    Accuracy of Localization System
   Default sample                                    Accuracy at Different Locations
    size of testing
    set = 4                        90   85                    85              87.5             87.5
                                             80
                                   80             75 74            75                                   75.94
                                                                         71               70          69
 80 testing sets                  70

                                   60
    for each of the                                                               52.5
    12 locations                   50
                      Percentage
                                   40

                                   30

                                   20

                                   10

                                    0
                                        1    2    3      4    5    6    7     8      9   10    11   12   Overall

                                                                       Area
Accuracy of Localization System
                      System Overall Accuracy

            100
             90
             80
             70
             60
 Percentage 50
              40 76   87.6       93      96   98    99   99.5
              30
              20
              10
               0
                 1
                      5
                                9        13
                                              17
                                                   21
                             Sample No                   25
Other Factors affecting Accuracy
   Property of signals
       The strength of signals fluctuates
   Hardware failure
       access points fails to give out signals or give out
        signals at unusual strength
   Change in environment
     addition access points on the floor
     opening the doors

   Orientation in collecting signal
    Wi-Fi Location System (WLS)
   Development Tool for Location-Based System
   Simplify development steps
   Increase the efficiency and productivity
   It divides into 3 components
     Wi-Fi Signal Scanner (WSS)
     Wi-Fi Data Processor (WDP)

     Wi-Fi Location Detector (WLD)
        Wireless LAN Terminology
   Media Access Control address (MAC Address)
       48 bits long
       unique hardware address
       e.g. 00:50:FC:2A:A9:C9
   Service set identifier (SSID)
       32 character
       Wireless LAN identifier
   Receive Signal Strength Indicator (RSSI)
       signal strength
       unit is in dBm
                       Overview
   Platform:
     Window CE
     Window XP, 2000

   Technology:
       IEEE 802.11b
   Tools
     Embedded Visual C++ 4.0
     Visual Studio .NET 2003
    Tradition Development Procedure
                 (TDP)
   The followings in the Tradition Development
    Procedure
                        1-2 week
         Studying the              Software
          technology                Design

                                          2-3 week
                        1-2 week
           Final                   Algorithm
          System                    design
Wi-Fi Location System Development
         Procedure (WLP)
           Collecting Data
                                    Using Wi-Fi
                        Several
                                  Signal Scanner
                        hours
           Processing Data

                        1 day      Using Wi-Fi
                                  Data Processor
           Deploying and
            Test System
                                   Using Wi-Fi
                        Several
                                    Location
                        days
                                    Detector
            Final System
Comparison between TDP and WLP
   Using WLP, we can develop Location-Based
    System in a short time.
   This work can be done by non-professionals
   It simplifies Development Steps
            Wi-Fi Signal Scanner
   To collect the signal
    strength received from
    access points
             Collected Data

 Mean of
  Total
Number of
  Strength
 Received
Received
   Signal
  Signal
  Signal
              Wi-Fi Data Processor
    To process collected data

  Access
Setting and
  Position
   Point
Information
   Region
   Region
  Region
            Wi-Fi Data Processor
   Two main steps in WDP
     Filter out useless data
     Set parameters at each position
         Data
         Name

         Point at Map Picture
        Wi-Fi Location Detector
   Three functions in WLD
     To detect the location in the target place
     To show the detected position name and
      corresponding position at the Map Picture
     To show calculated probability

   Three modes in WLD
     Data Mode
     Map Mode
     Probability Mode
                   Data Mode
   To show the sample data
           Map Mode

Position




  Name
               Probability Mode
   To show calculated probability at each position
                  Conclusion
   We are success in applying Area-Based
    Probability
   We have done experiments on accuracy of
    algorithm
   We have implemented Location-Based
    Development Tool—Wi-Fi Location System
   Based on our knowledge and developed tools in
    localization, we are able to further develop a
    location-based service
                Future Work
   Ho Sin-Hang Engineering Building Tour Guide
    Service
   Multimedia Application with video streaming
   Improvement in Localization Algorithm
   Increase the Accuracy in Localization
   Research on 3D localization algorithm in an
    building
Q&A
DEMO
THE END

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:10/11/2011
language:English
pages:50