LYU0401 Location-Based Multimedia Mobile Service

1
LYU0401 Location-Based Multimedia Mobile Service

• Clarence Fung
• Tilen Ma
• Supervisor Professor Michael Lyu
• Marker Professor Alan Liew

2
Outline

• Objective
• Introduction to Localization
• Area-based Probability
• Center of Mass
• Multimedia Guidance System
• Conclusion

3
Objective

• To study different algorithms and techniques in
  localization
• To find out a suitable localization method for
  Location-Based Service
• To develop an application for rapid
  Location-Based Multimedia Service System
  Development

4
Main Steps in Localization
Decide the Areas
Measure Signals at Decided Areas
Create a Training Set
Measure Signals at Current Position
Create a Testing Set
Apply Suitable Algorithm
Return the most likely Area
5
Decide the Areas
6
Collecting Training Set Data (I)

• In one particular area Ai, we read a series of
  signal strengths (sijk ) for a particular APj
  with a constant time between samples
• We estimate sij by averaging the series, sij1,
  sij2, sijo

7
Collecting Training Set Data (II)

• We read signals of all n APs, so we have the
  fingerprints at Ai
• We read signals at all m areas

8
Collecting Testing Set Data

• collects a set of received signal strengths when
  it is at certain location
• similar to the fingerprints in the training set
• a set of average signal strengths from APs

9
Data Processing

• delete some access points that have least
  contribution to localization
• To shorten computation time
• input -92 dBm for missing signal strengths

10
Training Set Example
Position 1 2 3 4 5 6 7 8 9 10 11 12
AP MAC address Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm) Signal Strength (dBm)
00022d28be9e -70 -62 -58 -67 -73 -78 -83 -86 -84 -81 -78 -55
00022d28be5d -67 -59 -60 -71 -76 -79 -81 -86 -81 -83 -79 -52
00601d1e439b -79 -87 -85 -84 -89 -80 -76 -77 -66 -63 -77 -90
000f34f36040 -63 -69 -65 -74 -76 -72 -77 -84 -76 -74 -66 -79
00022d21391f -92 -92 -82 -78 -82 -59 -78 -73 -83 -85 -82 -92
0011933d6fc0 -92 -92 -92 -90 -85 -86 -89 -88 -92 -92 -92 -92
000f34bbdf20 -92 -92 -92 -89 -90 -82 -88 -88 -92 -92 -92 -92
11
Testing Set Example
AP MAC address Signal Strength (dBm)
00022d28be9e -71
00022d28be5d -72
00601d1e439b -89
000f34f36040 -49
00022d21391f -92
0011933d6fc0 -92
000f34bbdf20 -92
12
Area-Based Probability (ABP)

• Advantages
• Area-based approach
• More mathematical approach
• Higher accuracy
• Approach
• compute the likelihood of the testing set (St)
  that matches the fingerprint for each area (Si)

13
Assumptions in ABP

• Signal received from different APs are
  independent
• For each AP, the sequence of RSS sijk, is modeled
  as a Gaussian distribution

14
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) cP(St Ai) (2)

15
Gaussian Distribution

• We compute P(St Ai) for every area Ai ,i1m,
  using the Gaussian assumption
• Integral of Normal Function taking the interval
  as 1

16
Error function erf(x)

• Express Integral of Normal Function in terms of
  erf
• Approximate value of erf by a series

17
Use of probability

• Discrete Space Estimation
• Only fixed number of locations in the training
  set can be returned
• Eg. Return the area Ai with top probability
• Continuous Space Estimation
• Continuous range of locations can be returned
• Return locations may or may not be in the
  training set
• Eg. Finding Center of Mass

18
Accuracy of Discrete Space Estimation

• Default sample size of testing set 4
• 80 testing sets for each of the 12 locations

19
Continuous Space Estimation (CSE)

• Advantage
• Return locations may or may not be in the
  training set
• Higher accuracy
• Suitable for mobile application
• Two techniques
• Center of Mass
• Time-Averaging

20
Center of Mass

• Assume n locations
• Treat each location in the training set as an
  object
• Each object has a weight equals to its
  probability density
• Obtain Center of Mass of n objects using their
  weighted positions

21
Center of Mass

• Let p(i) be the probability of a location xi,
  i1,2 n
• Let Y be the set of locations in 2D space and
  Y(i) is the corresponding position of xi
• The Center of Mass is given by

22
Example
23
Experiment on Center of Mass

• 16 positions for the training set
• In every location, we use our system to perform
  positioning for 100 times
• Totally we get 1600 records

24
Experiment Result
25
(No Transcript)
26
Summary of statistics at all the positions
27
(No Transcript)
28
Analysis of Experiment Result

• If the tolerance of error is 2 meters and 3
  meters, the accuracy of our system is 85 and 94
  respectively
• A few estimated points with a large distance
  error (3m to 5m)
• Inaccuracy in testing set
• Frustration in signal strength

29
Application System

• In our project, we have implemented two
  development tools
• Wi-Fi Location System (WLS)
• To develop Location-Based System
• Multimedia Guidance System (MGS)
• To develop Location-Based System with multimedia
  services
• Developer can develop any Location-Based
  Multimedia System using our tools

30
Development Environment

• Platform
• Window CE
• Window XP, 2000
• Technology
• IEEE 802.11b
• Tools
• Embedded Visual C 4.0
• Visual Studio .NET 2003

31
Wireless LAN Terminology

• Media Access Control address (MAC Address)
• 48 bits long
• unique hardware address
• e.g. 0050FC2AA9C9
• Service set identifier (SSID)
• 32 character
• Wireless LAN identifier
• Receive Signal Strength Indicator (RSSI)
• signal strength
• unit is in dBm

32
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)

33
Wi-Fi Location System

• Wi-Fi Signal Scanner
• To collect the signal strength received from
  access points
• Wi-Fi Data Processor
• To process collected data
• To filter noise data
• Wi-Fi Location Detector
• To detect the location in target place
• To show the detected position name and
  corresponding position at Map Picture

34
Multimedia Guidance System (MGS)

• WLS Limitations
• only provides Location-Based services
• does not support any multimedia services
• Discrete Space Estimations
• We create new development tools to support
  multimedia services

35
Multimedia Guidance System (MGS)

• Multimedia Guidance System is a development tool
  to support multimedia service
• It consists of three components
• Wi-Fi Signal Scanner (WSS)
• Multimedia Guidance Processor (MGP)
• Location-Based Multimedia Service System (LBMSS)

36
Multimedia Guidance System

• Deploying Procedure

Processing Data and generating LBSData
Collecting Signal strength data
Deploying System with clients and server
application
37
Wi-Fi Signal Scanner (WSS)

• To collect the signal strength received from
  access points
• The collected data received from WSS is then
  processed by MGP
• Same as WSS in WLS

38
Wi-Fi Signal Scanner
Number of received data
MAC Address
SSID
WEP
Signal Strength
Total Number of Signal Strength
Mean of Signal Strength
39
Multimedia Guidance Processor (MGP)

• To process and filter collected signal strength
  data
• To set client environment in LBMSS
• Server Address
• Availability of Service
• Position Information
• Video
• Picture
• Position in Map Picture

40
Multimedia Guidance Processor

• Data Processing Procedure

Open the MGP
Set the project name, server path and map picture
path in setting
Setting Lines
Add/Delete the target Position
Set the name, data, video, picture, position in
the map picture and point of interest
Filter the noise data
Generate location-based data
Select service for mobile client
41
Multimedia Guidance Processor
Position Information
Access Point Information
Position Setting
Service Setting
Information Section
Environment and Options Setting
42
Location-Based Multimedia Service System (LBMSS)

• To provide multimedia service for users
• It mainly consists of two parts
• Client
• Server
• Chat and Management Server
• Paint Server
• Media Server

43
Overall Architecture
44
Overall Architecture

• Techniques on Client/Server model in LBMSS
• Socket Programming in .NET
• Multithread Model (AsyncCallback, Non-Blocking
  Function)
• Encode object variable into byte stream
• Import Native Code (Visual C API) to Managed
  Code (.NET API)
• Control other program by CreateProcess() API

45
Paint Server

• To provide paint service
• It has two main functions
• Store Clients Artwork
• Print Artwork for Client

46
Chat and Management Server

• To provide chat service
• It shows the current position of all clients on
  the map

47
Media Server

• To provide media service
• It is made by existing server architecture (e.g.
  HTTP, FTP, Streaming Server)

48
Client

• To provide multimedia service for users
• Client program mainly has 4 services
• Guide
• Paint
• Chat
• Video

49
Client

• Guide Service
• Select Destination
• Guide Line shows on Map Tab
• Implement by Shortest Path Algorithm

50
Shortest Path in the Guide Service

• We define the nodes as the turning points or the
  ending of the corridors

51

• add edges connecting two nodes
• give a weight which is equal to the distance
  between the two nodes

52

• add the source node and the destination node
• add new edges
• Use of Dijkstras algorithm

53
Dijkstras algorithm

• 1. Set i0, S0 u0s, L(u0)0, and
  L(v)infinity for v ltgt u0. If V 1 then stop,
  otherwise go to step 2.
• 2. For each v in V\Si, replace L(v) by minL(v),
  L(ui)dvu. If L(v) is replaced, put a label
  (L(v), ui) on v.
• 3. Find a vertex v which minimizes L(v) v in
  V\Si, say ui1.
• 4. Let Si1 Si cup ui1.
• 5. Replace i by i1. If iV-1 then stop,
  otherwise go to step 2.

54
Client

• Paint Service
• Get Picture from Media Server
• Draw on Picture
• Send Artwork to Paint Server

55
Client

• Chat Service
• Message can be sent both direction
• i.e. Client to Server or Server to Client

56
Client

• Video Service
• Play video by clicking Play Button
• Video is played by Windows Media Player (WMP)
• WMP is controlled by CreateProcess() API in
  Embedded Visual C

57
Comparison between Tradition Method and MGS Method

• Tradition Method
• Studying the Technology (1-2 week)
• Software Design (2-3 week)
• Architecture Design (2-3 week)
• Algorithm Design (1-2 week)
• MGS Method
• Wi-Fi Signal Scanner (1/2 day)
• Multimedia Guidance Processor (2-3 days)
• Location-Based Multimedia Service System (4-5
  days)

58
Comparison between Tradition Method and MGS Method

• Using MGS method, we can develop Location-Based
  Multimedia Service System in a short time.
• This work can be done by non-professionals
• It simplifies the development steps

59
Conclusion

• We are successful in achieving the goal of
  localization
• We have done experiments on accuracy of algorithm
• We have implemented two location-based
  development tools
• Wi-Fi Location System
• Multimedia Guidance System

60

• QA

61

• The END