Please enable JavaScript.
Coggle requires JavaScript to display documents.
Ninth reading - Spatial and Temporal Data and Mobility, Ariana Alvarado…
Ninth reading - Spatial and Temporal Data and Mobility
Complex data management: time, space and multimedia in databases.
Challenge: expansion of mobile devices, including phones.
Motivation
Temporal data
Databases model the present, but storing past data is crucial, integrating it manually.
Spatial data
maps and designs. Initially in files, but insufficient with growth.
Multimedia data
Video and audio data are called continuous because they require recovery at a constant rate.
Mobile databases
They operate offline, with limited storage.
Time in Databases
Databases: present represented, essential past in time bases.
In databases, essential difference between real time (veracity) and transaction time (actuality, automatically generated).
Tuples in temporal relationships have time associated with them, either valid or transaction time. If both are stored, it is called bitemporal.
Time Specification in SQL
Date: Four digits for the year (1-9999), two for the month (1-12), and two for the day (1-31).
Time stamp: Date and time fields, with six fractional digits for seconds.
Time zone: Allows specifying the time zone along with the time.
Temporal Query Languages
Temporal Projection: Tuples inherit times from the original relation.
Temporal Union: Time in the result is the intersection of the times of the original tuples.
Temporal Predicates: "Precedes," "overlaps," and "contains" apply to inter
Temporal Functional Dependencies: Should be handled with caution as they may change over time.
Spatial and Geographic Data
Spatial Data Support:
Fundamental in databases to efficiently store, index and query data based on spatial locations.
Spatial data types:
Computer-aided-design (CAD) data
Construction details for objects and essentials in computer-aided design databases.
Geographic data
Databases for geographic maps, such as roads and land use.
Representation of Geometric Information
Line Segment: Coordinates of extremes, e.g., latitude and longitude on a map.
Linestring: Connected sequence of segments represented by coordinates.
Circular Arcs: Some systems allow representing curves as sequences of arcs.
Listing vertices in order specifies the boundary of the polygonal region.
Alternatively, you can divide a polygon into triangles by triangulation.
An identifier is assigned to curves and polygons.
Each segment of a polyline carries its identifier.
Design Databases
Traditional CAD storage:
CAD stored data in memory and wrote it to the file at the end of the edit.
Disadvantages:
Cost and complexity of transforming data.
Challenges with Large Designs:
For large designs, such as circuits or airplanes.
Motivation for Object-Oriented Databases:
Designers were inspired by the needs of the design.
Design Structure
Objects stored in the design database are primarily geometric.
Two-Dimensional Geometric Objects:
Points, lines, triangles, rectangles and polygons.
Operations on Complex Two-Dimensional Objects:
Union, intersection and difference form objects.
Three-Dimensional Objects:
Spheres, cylinders, cubes.
Geographic Data
Geographic Data:
Spatial in nature, differing from design data.
Common Examples
Maps and satellite images.
Map Information:
Locations, boundaries, rivers, roads.
Applications of Geographic Data
Uses of Geographic Databases:
On-line map services.
Vehicle navigation systems.
Distribution network information for utilities.
Online Mapping Services:
Extensive use in web mapping services.
Online map generation for desired regions.
Benefits of Online Maps:
Easy scalability for zoom in/out.
Storage of detailed road and service information.
Additional Functionalities:
Route guidance, automatic trip planning.
Consultation of information on services such as hotels, gas stations and restaurants.
Representation of Geographic Data
They can be classified into two types:
Raster data
Data in the form of bitmaps or pixels in two or more dimensions.
Common example: satellite image of a region.
Often arranged as tessellations, each covering a fixed area.
Vector data
Use basic geometric objects such as points, lines and polygons.
Flexible representation of geographical and political elements.
Points represented by latitude and longitude.
Spatial Queries
There are queries involving spatial locations.
Nearness queries
Search for objects near a specified location.
Example: Find restaurants within a given distance from a point.
Query for the closest object to a given point.
Region queries
Spatial region queries search for objects in a specified region.
Example: find retail stores within the boundaries of a city.
Indexing of Spatial Data
Traditional indexes are not suitable for spatial data.
They work in one dimension, while spatial data are multidimensional.
k-d Trees
To index spatial data, we use trees such as binaries and B-trees.
These structures divide the space into smaller parts.
Each level of a B-tree divides an interval into several parts.
In a balanced tree, the partition is chosen so that approximately half of the points fall in each part.
Quadtrees
Each node associated with a rectangular region of space.
The top node is linked to the entire space.
Non-leaf nodes are divided into four quadtrees, each with four child nodes.
Leaf nodes contain between zero and a fixed maximum number of points.
PR quadtree stores points and divides the space into regions.
Each node corresponds to a subset of array values.
Multiple occurrences can lead to storage and query inefficiencies.
R-Trees
Stores and organizes objects such as points, line segments and polygons.
Balanced tree structure with leaf nodes containing objects.
Each node has a rectangular bounding box.
Multimedia Data Formats
Multimedia requires compressed storage and transmission.
JPEG is common for images.
MPEG standards for video and audio (MPEG-1, MPEG-2, MPEG-4).
MPEG-1: 30 fps video and audio at ~12.5 MB per minute.
MPEG-2: For digital systems, DVD, ~17 MB per minute.
MPEG-4: Higher compression with variable bandwidth.
MP3, RealAudio, Windows Media Audio for audio.
Similarity-Based Retrieval
Pictorial data
Databases find similar images, like trademarks, to prevent duplicates during registration.
Audio data
Speech-based interfaces analyze user voice input for similarity with stored system data.
Handwritten data.
Handwritten input is matched for similarity to identify stored data or commands in the database.
Continuous-Media Data
Video and audio data are key types.Video and audio data are key types.
Characteristics of continuous-media systems:
Real-time delivery without gaps in audio or video.
Data rate prevents buffer overflow.
Synchronization of distinct data streams (e.g., lips moving with audio).
Data fetching coordinated for real-time delivery.
Cycle period balances memory and disk-arm movement.
Video-on-Demand System Architecture:
Video Server: Stores multimedia data, often in RAID configuration.
Terminals: Devices like PCs or set-top boxes for viewing.
Network: High-capacity network for data transmission.
Multimedia Databases
Includes images, audio and video.
Generally stored outside the database in file systems.
Problems arise with large numbers of multimedia objects.
Useful database for descriptive attributes and multimedia file management.
Mobility and Personal Databases
Rise in portable computing devices (laptops, notebooks, netbooks).
Proliferation of smartphones with computer-like capabilities.
Affordable wireless digital communication infrastructure.
Ariana Alvarado Molina - 2021089068