[Todos] Charla JAVA: Small Programmable Object Tecnology Sun Project

[Irene Loiseau]

CHARLA: Small Programmable Object Tecnology Sun Project

En el marco de los Sun Java TechDays el martes 24 de Octubre de
15 a 16:00hs (aula a confirmar) dictará una charla la Dra.
Ángela Caicedo quien entre otras cosas se referirá al proyecto Sun SPOT
(http://sunspotworld.com/).

Ángela es una de las representantes de SUN encargada de difundir
mundialmente tecnología Java. En particular, está especializada en Java ME
y es desarrolladora dentro del proyecto Sun SPOT avocado a simplificar el
desarrollo de sistemas que incluyan sensado inalámbrico.

Las aplicaciones de estos sistemas van desde el sensado remoto de
variables ambientales hasta la organización de comunidades de pequeños
robots.
La charla cubrirá la esencia de la tecnología, demostraciones y como
desarrollar en ella.
También hablará sobre Java  ME (variante de Java dedicada a dispositivos
subPC) y como desarrollar usando netbeans.

Quedan todos invitados.

------------------------------------------------
Project Sun SPOT:
Programming the Real-world

By simplifying the development of wireless transducer applications, the
Sun SPOT System from Sun Labs will help transform the potential of
wireless sensors into real-world products.
The possibilities for wireless sensors have excited scientists and
researchers, the business community, military and government
officials, and consumers alike for many years. The potential
applications for wireless sensors and transducers (sensors combined with
actuator mechanisms) are limited only by the imagination. Sensing alone is
not enough for many applications-the ability to act on the sensory data is
also required. A small, battery-powered platform
capable of sensing and actuation is needed. Just a few examples:

"Smart dust" networks of tiny wireless microelectromechanical sensors
(MEMS) could track and report on everything from cold areas in rooms to
enemy movements in a military operation.

Wireless sensors could catch manufacturing defects by sensing
out-of-range vibrations in industrial equipment, monitor patient
movements in a hospital room, track fertilizer runoff from farms into
lakes, or ascertain the origins of a gunshot. Used in conjunction with
RFID tags and actuation mechanisms, they could intelligently control
light, heat, and appliances in buildings, making rooms "come alive" just
before people enter.

"Gesture interfaces" harnessing wireless sensors could make it
possible to control devices by hand gestures, eliminating the
distraction of finding and turning dials and knobs.

"Swarms" or large numbers of autonomously functioning vehicles could be
deployed to carry out a prescribed mission and respond as a group to
high-level management commands.
Technological Obstacles to Widespread Adoption

The technology for wireless sensors is compelling, but on the whole there
is still a wide gap between vision and real-world applications. Wireless
sensors have remained primitive and difficult to program-not ready for
mass commercial deployment. There are several technical
challenges that must be overcome:
* Current development tools for creating and investigating wireless sensor
and transducer systems are difficult to use and unproductive. * Security
is a critical issue for many wireless sensor applications, but
implementing effective security mechanisms within tight resource
constraints at an affordable cost can be difficult and complex.
* More powerful processing capability is needed close to the sensor for
signal analysis and control, and libraries are needed for
investigating all aspects of wireless transducer applications-from
hardware all the way through to network layers.
* Unique characteristics of these new small devices present challenges for
networking, requiring new ways for devices to communicate with each other
and the Internet. Current standards don't apply, so new standards are
needed which support communication both between the new devices themselves
as well as between the new devices and devices
using today's standards.

Sun SPOT: Simplified Development of Wireless Transducers Using Java
Technology

Researchers at Sun Microsystems Laboratories (Sun Labs) are developing a
system that takes major strides toward solving the key challenges that are
inhibiting development of wireless sensor and transducer
applications.

Based on a 32 bit ARM CPU and an 11 channel 2.4GHz radio, Sun SPOT
radically simplifies the process of developing wireless sensor and
transducer applications. The platform enables developers to build
wireless transducer applications in Java using a sensor board for I/O, an
802.15.4 radio for wireless communication, and use familiar
Integrated Development Environments (IDEs), such as NetBeans to write code.

The Sun SPOT system uses Java technology to up-level programming.
Developers can write a program in Java, load it on a wireless sensor
device, run it, debug it, as well as access low-level mechanisms-with
standard Java IDEs. The inherent portability of Java makes it simpler to
migrate applications among platforms and enables developers to
build new wireless sensors devices using off-the-shelf hardware
components. Java also eliminates or streamlines many of the low-level
tasks of traditional development languages such as C, and for the
millions of developers who already write code in Java there is little
additional learning curve for building wireless sensor/transducer
programs.

The Sun SPOT system features the "Squawk VM," a small J2ME virtual machine
(VM) written almost entirely in Java. It provides the ability to run
wireless transducer applications "on the metal," (directly on the CPU
without any underlying OS), saving overhead and improving
performance. End users also gain the flexibility to experiment with
different implementations of low-level services, such as networking
protocols, which are typically buried inside an OS. A set of Java
libraries under development will provide access to the sensors, the I/O
pins on the sensor application board, and the integrated on-board radio.
By running multiple applications on the one virtual machine, and by using
a more compact representation of class files, the Squawk VM makes better
use of the small memory space available on SPOT
devices.
_______________________________________________


Irene Loiseau
Departamento de Computación
Facultad de Ciencias Exactas y Naturales
Universidad de Buenos Aires
Pabellón I- Ciudad Universitaria
1428 Buenos Aires - ARGENTINA
TE/FAX: 54 11 4576 3359
TE: 54 11 4576 3390/96 int 711
e-mail: irene en dc.uba.ar