Biomedical Engineering Reference
In-Depth Information
viding global information. This aspect will reduce the uncertainties about the
physical world-state comprising the artificial based system, the involved human
and its environment. The only apriori defined constraint in this model procedure
is what sensing parameters are to be measured without necessarily having the
knowledge about where, when or even if, they will occur. The challenge regarding
this approach is concerning how multi-sensor data are integrated with human per-
ception information, that are able to effectively be integrated in artificial world
models. This approach will add a new dimension to the process, that we will
denote the perception based sensors or artificial human sensors .
The increased interest on human-like and intelligence based system, based on
powerful computer platforms, that allow for complex data processing algorithms,
can be incorporated on a wider scale in new measurement methods. Virtual model
based solutions are evolving even more realistically in describing the environment
by providing an interactive ability in various applications, Ahmed (2007), Valdes
(2003).
The model procedure enables an extensive development of a symbiotic con-
cept involving both human and artificial sensor partnership based systems. This
approach may include the use of human capability as an explicit multi-sensor system,
as well as the use of human perception. This will act as an implicit behavioural and
contextual measuring unit for indicating behavioural qualities in the environment,
e.g., situation assessments.
4.5
SIGNAL PROCESSING
The signal processing mechanisms are generally performed in an electronic pro-
cessor unit or external computer. The main task for a human based sensor system
is to transform the measured sensing parameters into a representative qualitative
value, describing the measuring object of interest. The procedure forms the basis
of subsequent decisions, from sensor signals to an information status.
More or less advanced calculations are used and different techniques are
utilised for the realisation of signal processing. Normally, the reason to conduct
such a process includes the improvement of the interpretability regarding the
environment of interest. Sensor devices in general have shortcomings that will
cause imperfections of the rendered data. Under normal circumstances, many
of these imperfections can be suppressed by using appropriate signal processing
techniques. By incorporating statistical operations, and by analysing signals over
time, it becomes possible to get proper indications when significant deviations
from normal conditions occur. In the case of using several sensors, a combined
analysis of multi-sensor signals may reveal hidden patterns that can be extracted
and correlated to significant properties of measured samples.
Advanced signal processing is a useful tool for enhancing the information that
can be comprised in the sensor data. The obtained information can enhance the
usability and performance of the overall system, in order to achieve an effective
Search WWH ::




Custom Search