IEEE Consumer Electronics Magazine - September 2018 - 36

Destined Client
Wireless
Link

SYSTEM OVERVIEW AND ASSUMPTIONS

Internet
NMS

Monitored Region of Interest

Omnidirectional Camera Sensor

360° Camera FOV
Scalar Sensor
FIGURE 3. An overview of the proposed system.

Virtual Compartments

: Scalar Sensor
: SL
Event Radius
Event Area

Camera
Sensor
DOF

Event Point
FIGURE 4. An SL event-detection situation in a multievent occurrence scenario.

again for minimizing redundancy in data transmission. In the
current research, we address a novel distributed approach,
e.g., CMT-MDSD, for affording impressive enhanced event
area coverage while actuating the least number of camera
sensors and, at the same time, lowering the redundancy in
data transmission. Our approach is an extension of the investigative work presented in [16], where we rigorously discussed the methodology involved in the proposed method in a
multievent occurrence scenario.
36 IEEE Consumer Electronics Magazine

A COMPREHENSIVE VIEW
OF THE PROPOSED APPROACH

september 2018

We assume our proposed model to be of the form as portrayed
in Figure 3. The entire network comprises camera sensors and
scalar sensors. The sensors, deployed randomly throughout
the monitored region in fixed positions, are assumed to be
time synchronous, which aims to provide a common time
scale for proper coordination among the local clocks of the
sensors. Similarly, omnidirectional camera sensors with a panoramic view are used.
A node-management server (NMS) is used for the management of, and coordination among, the sensors. It retains
information regarding the sensors deployed in the monitored
region. Furthermore, the NMS broadcasts control messages
to all of the deployed sensors at the needed time. The node
closest to the NMS is selected as the sink node for effectively
retrieving the desired data. In this context, the sink node collaborates with the destined client.
We considered the simultaneous occurrence of multiple
events in our proposed approach. The entire monitored region
is logically segregated into a number of virtual compartments.
An SL is elected from among the scalars so that the measurement with the minimal mean distance from the remaining scalars within a specific compartment is determined as the SL for
that compartment. When multiple events occur, the SLs within
the purview of the prevailing events detect them and act as the
representatives of the scalars in the corresponding virtual
compartments. This scenario is illustrated for four virtual
compartments in Figure 4.
Furthermore, the SLs operate as the vehicles for event
information communication to their corresponding cameras.
Finally, the informed cameras undergo distributed actuation
based on the ascending order of their respective identities
(IDs) while concurrently matching their respective SL IDs.
The following data structures and messages are used in the
proposed CMT-MDSD approach:
▼ Event-detecting SL table (EDSLT): Retains the SL IDs that
detect the prevailing events.
▼ My scalar sensor table (MSST): Contains the IDs of the
scalars lying within the DOFs.
▼ My SL table (MSLT): Retains the IDs of those eventdetecting SLs that are common to the DOFs of recently
actuated camera sensors and prevailing events.
▼ My regional ID table (MRIDT): Holds the IDs of the virtual compartments pertaining to the corresponding monitored region under speculation.
▼ Compartmental scalar ID table (CSIDT): Contains the IDs of
the virtual compartments and their respective scalar sensor IDs.
▼ Waiting list (WL): A repository of the IDs of the camera sensors, in ascending order, that have not yet undergone actuation.
▼ Current actuation list (CAL): Holds the IDs of the actuated
camera sensors.
▼ MYDIST list: Maintained by each of the scalars and cameras that retains its distance from each of the other sensors.

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September 2018