Docstoc

sensor-network-app

Document Sample
sensor-network-app Powered By Docstoc
					   Sensor Networks:
Applications and Services




                            tseng:1
                        Outline
   Sensor Networks
     applications and services
     sensor network and instant messaging
     “Where are you?”
        location management
        localization problem




                                             yctseng: 2
                  Definition
   Sensing technology:
     light, temperature, moisture, objects, ect
   Embedded computing:
     small, light, compact software and
      hardware
   Wireless communication:
     low/medium-rate antenna
     wireless ad hoc networking


                                               yctseng: 3
Generic System Architecture
  (感測網路系統基本架構 )




                              yctseng: 4
Sensor Hardware
(感測器硬體設計 )




                  yctseng: 5
            Example 1
   Digital Sun公司所發展的自動灑水系統
    「S.Sense Wireless Sensor」




                            yctseng: 6
               Example 2
 展覽會場的保全系統 -- Sensicast ART
 在博物館、圖書館、畫廊、藝術品展覽會場,
  防止有價值的藝術品或展覽品遭到竊盜、不經
  意的觸摸、任意搬動等情形。
 主要有兩個模組:
     OAS 物件警告系統(Object Alarm System):感測裝
      置安裝在藝術品底部或背面,藉由偵測燈光的亮度
      是否改變、測量是否遭受到振動等因素,來確保展
      覽品的安全。
     EMS 環境管理系統(Environment Management
      System):安裝在展覽會場的牆角、天花板等,偵
      測展覽環境的溫度、溼度是否超過安全值,以保護
      展覽品的品質。

                                      yctseng: 7
              Example 3
   Senera的橋樑安全監控系統
     (89.8.27)聯絡高雄與屏東之間的高屏大橋突然斷裂,造成多
      位民眾受傷
     Senera的感測系統,用於監視橋樑、高架橋、高速公路等道
      路環境。對於許多老舊的橋樑,橋墩長期受到溪水的沖刷,
      本感測器能夠放置在橋墩底部、用以感測橋墩結構;亦可放
      置在橋樑兩側或底部,蒐集橋樑的溫度、溼度、振動幅度、
      橋墩被侵蝕程度等,期望能減少斷橋所造成生命財產的損失。




                                 yctseng: 8
                    Example 4
   Habitat Monitoring on Great Duck Island
     遠端利用Mote形成的sensor network觀察動物的棲息地環境。
     Mote可以供環境光線、溫度…等變化的sensor node,藉由
      形成的Wireless Sensor Network可以遠端觀察環境,並且長
      時間蒐集環境的變化資料。




                                              yctseng: 9
                 Example 5
   工研院 SCAN project (2003):
     platform with interface to Berkeley Mote
     Bluetooth devices to form scatternet
     CO and other sensors


** Other Information:
     RFID sponsored by Walmart, UPS, etc.




                                             yctseng: 10
SCAN




       yctseng: 11
SCAN




       yctseng: 12
     Sensor Network + Instant Messaging




   A Master Thesis by Y. K. Liu,
    CSIE, NCTU



                                          tseng:13
                Smart Office

 Bluetooth-based Sensor Network
 Mike經理接獲即時訊息 “會議己ready”
 會議中… 助理要找Mike經理, 並設定即
  時訊息
 會議後… 助理收到即時訊息 “Mike經理
  己經回到office”

   (view flash and video)
                               yctseng: 14
Survey of Instant Messaging
Services and Sensor Network
   Instant Messaging Services
       Short Message Service (SMS) in GSM
       MSN Messenger / ICQ
   Wireless Sensor Networks
       Active Badge
       RFID



                                             15
Goals of our Event-Driven
Messaging Services
   Cross-network applications and services.
   Event driven messaging service
   Modular approach by dividing the
    system into several subsystems
    according to their functionalities.



                                          16
System Architecture
 Overview




                      17
    System Architecture (1/2)
   Location Server
       Maintain the user-location mapping in a
        database.
   Action Server
       Carry out action with reliability
   Event Server
       Trigger an action

                                                  18
        System Architecture (2/2)
   SMPP Client and WAP Web Server
       Plays the role as a bridge between telephone network
        and Internet.
   Sensor
      Detects the user location
      Connects with mobile terminals through Bluetooth

   Client
      Bluetooth-enabled laptop/palmtop or a Bluetooth-

       enabled WAP handset.

                                                               19
Configuration Example 1
   Mike configures an event:
       (Alice Enter Sensor X) AND (Bob
        Enter Sensor X) Do Unicast (Mike)

       When the event becomes true, the
        system will send Mike a message.


                                           20
Message Flows (through NB)
                  (Message for
                                            Alice > X
                  Mike, index)
                                           & Bob > X
                                          Send to Mike
                                         index of Action



                                 3

                        4
                            7        2
                    6
      Alice > X
      Bob > X                        1
                    5


                                                     21
When the Event Happening
                     (Message for               Alice > X
                     Mike, index)              & Bob > X
                                             index of Action


                                     4
                       5
                           6
                               3
                                   7 9
                       2
       Bob > X

                                              Alice and Bob
                                         8
                 1                            have arrived!


                                                         22
Configuration Example 2
   Cathy Submits a request “On
    ( Manager LEAVE Sensor X +
    3 )” Do Unicast Cathy “ through
    WAP.
   When the event becomes true, the
    system will send Cathy a message


                                       23
Message Flows (through WAP)
                         (Message for
                         Cathy, index)
                                              (Mike < X)+3
                                              Send of Action
                                             Index to Cathy
                     3
      2
                                         5
                               4

                                   6
  1
          Mike < X




                                                         24
            When the Event Happening
                                                 The device of
                                                  Cathy is a
                             (Message for          handset
                             Cathy, index)
                                                       (Mike < X)+3
                                                      Index of Action

                               6
                      7                          3
                                ACK
              8                 4 5
                                             2

                  Mike < X
                               1                     3 min later!!
        9
 Mike is
available


                                                                  25
Definition:
Events and Actions (1/2)
   Events are expressed as the following
    format
         On < EvntVal > Do < Action >
   Action
       Unicast, Geocast, Multicast, or Broadcast
       Text or file



                                                    26
     Events and Actions (2/2)
   EvntVal
       Time Event
            Absolute time :    @04/16/04 9:00
            Relative time :    (Bob > Office) + 5
            Time interval :    04/16/04 9:00 ~ 04/16/04 11:00
            Periodical time:   04/16/04 9:00 ^ 10
       Location Event
            “Enter”, “Leave” and “At”
       Compound Event
            AND, OR, NOT, (),
                                                                 27
     EBNF Grammar of Event
     Expression
   < EvntVal > = < SubEvntVal > * < EvntVal > |
                  < SubEvntVal > + < EvntVal > |
                  < SubEvntVal >
   < SubEvntVal > = < SnglEvntVal > | (< EvntVal >) |
                     ! < SubEvntVal >
   < SnglEvntVal > = < LocEvnt > | < TimeEvnt >
   < LocEvnt > = < ID > < Rel > < Sensor X >
   < Rel > = > | < | @
   < TimeEvnt > = < min/hr/dat/mon/yr > |
                  < TimeOfEvnt(LocEvnt) + min> |
           <min/hr/dat/mon/yr~min/hr/dat/mon/yr > |
                  < min/hr/dat/mon/yr ^ period >
                                                         28
Sensing and Detecting
Capability
   A sensing field A with a N sensors. Sensors
    are not synchronized in time.
   The average Latency L : the user can be
    detected by any sensor after it appears.
   Ai be the area of A that is covered by
    exactly i sensors. (1 ≤ i ≤ n)
   Li be the latency such that a user is detected
    by any sensor after it appears in Ai .
   Therefore, L  A1  L  A2  L  A3  L ...  An  L
                          1       2        3            n
                     A        A       A           A 29
Sensing and Detecting
Example

                     A1
               A3
                    A2




    A1     A2     A3         An
 L   L1   L2   L3 ...   Ln
    A      A      A          A
                                     30
   Calculation of Li
     T -t T -t
L1         
       T         2
           t 1 (T - ( x  t ))                 T -t 1 ( x - t )  (T - ( x  t ))
                                    2                           2                 2
L2  2                             dx                                          dx
          0T                2T                t     T               2T
                    t  x 1 (T - ( y  t ))
                                              2
           t 1
L3  6                                      dydx
          0T        x     T          2T
                     T -t 1 ( y - ( x  t ))  (T - ( y  t ))
                                             2                    2
           t 1
    6                                                         dydx
          0T        tx T                       2T
                          T -t 1 ( x - t )  ( y - ( x  t ))  (T - ( y  t ))
                                          2                   2                 2
           T -2 t 1
    2                                                                         dydx
          t       T      tx T                          2T
L4  ...                                                                                31
Summary of Contributions
   We have prototyped an event-driven
    instant messaging application over
    integrated telecomm and datacomm
    networks.
   Sensing/detecting capability analysis




                                            32
        “Where are you?”



   location management problem
   localization problem



                                  tseng:33
               “Where Are You?”
   a question used in our daily life

   Essence of the question:
     mapping the location L of an identity ID
     L = a physical location, a relative location,
      a logical area
       e.g., 45.7th mile of highway I-95, at the north of
        Miami, under the table
     ID = a person, an object, a phone number


                                                     yctseng: 34
    Location Mgt. in Cellular Network
 ID = Mobile Station
  ISDN (MSISDN)
 L = location area




                                    yctseng: 35
      Location Mgt. in IP Networks
 ID = IP address
 L = subnet




                                     yctseng: 36
    Location Mgt. in Ad Hoc Networks
 ID = host
 L = physical location
 assumption: availability of GPS
 Example: virtual home region




                                    yctseng: 37
             One More Example
   Grid’s Location Service (GLS)
     distributed location servers




                                     yctseng: 38
A Short Summary: the Mapping




                               yctseng: 39
                Future Trend
 standardization:
  OSA/OMA
 hybrid location
  database




                               yctseng: 40
Positioning Techniques




                         tseng:41
        Measurement of Distance
   Time of arrival      Time difference of
                          arrival




                                          yctseng: 42
         Measuring Distance by
          RF Signal Strength
   Path loss formula



   An experiment by IEEE 802.11b




                                    yctseng: 43
       Positioning by Trilateration
   ideal situation      real situation




                                           yctseng: 44
Fingerprint Location Matching




                                yctseng: 45
           Pattern Matching Approach


training point                                                    Phase 2
                                         Phase 1              real-time phase
      (x1, y1, <ss1, ss2, ..., ssn>1) off-line phase
                                                                 <ss1, ss2, ..., ssn>


 (x2, y2, <ss1, ss2, ..., ssn>2)

                                                       Positioning
                                        Database         Model
                                                                             final point!!


                                                                     (x,y)
      (xm, ym, <ss1, ss2, ..., ssn>m)




                                                                                     yctseng: 46
            Conclusions
 sensor network applications
 sensor network + instant messaging
 location management
 localization




                                       yctseng: 47

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:6
posted:10/20/2011
language:English
pages:47