Databases Reference
In-Depth Information
FIGURE 9.3: Sensor node placement in a Cartesian grid. Each node is allo-
cated to a specific grid area. (This figure is copyright of and reproduced with
permission of Civil-Comp Ltd. Previously published in [96].)
For complex event detection (multiple sensory schemes), each DHGN sub-
net is mapped to a sensor node using a clustered configuration. The detection
scheme comprises a collection of wireless sensor nodes and a sink. We exam-
ine the deployment of the WSN in a two-dimensional plane with n sensors,
represented by a set W = (w
1
, w
2
, . . . , w
n
), where w
i
denotes the ith sensor.
The sensors are uniformly placed in a grid-like area, A = (x×y), where x
represents the x-axis coordinate of the grid area, and y represents the y-axis
coordinate of the grid area. Each sensor node is assigned to a specific grid
area, as shown in Figure 9.3. The location of each sensor node is represented
by the coordinates of its grid area (x
i
, y
i
).
For the communication model, a single-hop mechanism for data transmis-
sion from the sensor node to the sink is proposed. The “autosend” approach is
used to minimize errors associated with the loss of packets during data trans-
mission. Because of the front-end processing approach, the proposed scheme
does not involve massive transmissions of sensor readings from the sensor
nodes to the sink. Previous research has shown that the single-hop mechanism
is the most suitable approach for the DHGN deployment. Communication be-
tween the sink and the sensor nodes is performed using a broadcast method.
9.1.3 Dimensionality Reduction in Sensory Data
Event detection involves the recognition of significant changes or abnormal-
ities in sensory readings. In heterogeneous sensor networks, sensory readings
are of different types and values, e.g., temperature, light intensity, and wind
speed. In the DHGN implementation, these data must be pre-processed and
transformed into an acceptable format while maintaining the values of the
