IJSRST162437 | Received : 28 July 2016 | Accepted : 03 August 2016 | July-August 2016 [(2)4: 180-187]
© 2016 IJSRST | Volume 2 | Issue 4 | Print ISSN: 2395-6011 | Online ISSN: 2395-602X
Themed Section: Science and Technology
180
Querying Unintelligible Data on Geospatial Trajectory Database
Dr. K. Sathesh Kumar
1
, Dr. S. Ramkumar
1
1
Assistant Professor, Department of Computer Science and Information Technology, Kalasalingam University, Krishnankoil, Virudhunagar,
Tamil Nadu, India
1
Assistant Professor, Department of Computer Applications, Kalasalingam University, Krishnankoil, Virudhunagar, Tamil Nadu, India
ABSTRACT
Current GPS technologies collect objects and its movement and store the trajectories periodically in the MOD
(Moving Object Database). In such environment, some location errors may arise and some models are unable to
capture the changes in trajectories dynamically. Especially, the uncertainty capturing is a challenging one. In order
to handle these issues in spatial database, the proposed system develops a new trajectory model to handle the
uncertainty and querying on uncertain spatial queries. Initially this develops an adaptable trajectory approach to
provide actual positions and temporal changes in uncertainty along with improbable uncertainty ranges. The next
part of ongoing implementation provides effective spatial query processing with successful indexing process. This
presents the temporal R+ tree indexing with inverted list. This provides an efficient mechanism to evaluate
improbable range objects and its spatial queries using Rife-density trajectories.
Keywords: RFID sensors, R+Tree, GSM, Trajectory Model, Road Network, Dynamic Route Map.
I. INTRODUCTION
The spatial database is defines special data types for
geometric objects and allows to store geometric data
(usually of a geographic nature) in regular database
tables. It provides special functions and indexes for
querying and manipulating that data using something
like Structured Query Language (SQL). While it is often
used as just a storage container for spatial data, it can do
much more than that. Although a spatial database need
not be relational in nature, most of the well-known ones
are [1]. Uncertainty management is a central issue in
trajectory databases. The research interests, optimization
goals, and methodologies in this domain are indeed rich
and diverse. Despite this diversity, these studies are
generally established upon a common principle location
uncertainty is captured by a certain range centered on
the position recorded in the database. This principle was
initially discussed by Poser and Jensen in the database
literature, which is longer than a decade ago [2]. As of
today, GPS is no longer the only primary means for
positioning, yet a wide spectrum of technologies are
being used to produce trajectory data, including RFID
sensors, location estimation with 802.11, smart-phone
sensors, infrared and ultrasonic systems, GSM beacons,
and even vision sensors. These positioning systems
typically yield different characteristics of trajectory data,
which also exhibit various properties as well as degrees
of uncertainty.
Uncertain Spatiotemporal Data: Query processing in
trajectory databases has received significant interest over
the last ten years. Initially, trajectories have been
assumed to be certain, by employing linear or more
complex types of interpolation to handle missing
measurements [12]. These interpolation techniques,
however can lead to impossible patterns of movement as,
for example, a car might be assumed to drive through a
lake. Other solutions such as computing the shortest path
between consecutive locations can produce valid results,
but do not provide probabilities for quantifying the result
quality. As a result, a variety of uncertainty models and
query evaluation techniques has been developed for
moving object trajectories [28].