Information Technology Reference
In-Depth Information
applications or framework components with a database view of all sensor network
systems. Most others follow a centralized broker structure in which a central entity
takes care of interactions with different SAN systems.
Table 5.2 provides a summary of the functional features of the surveyed
approaches. Most notably, nearly all proposals fail to address accountability and
access control of service interactions between application and SANs as well as
privacy, trust, and reputability of offered SAN services and information. Only
Urban-sensing considers access control, privacy, and data integrity as fundamental
issue that needs to be addressed as part of the architecture, but current results are
limited to mere conceptual discussions. Nearly all approaches provide some way of
service discovery to the applications. Service composition , if provided is prelimi-
nary static, mostly based on information available to application at service discov-
ery time. Most approaches do not address issues of mobility and sudden service
unavailability with respect to longer-lasting queries to ensure the continuity of
request information and actuation services. Similarly, closely related mechanisms
for ensuring quality of information (QoI) and actuation (QoA) are not provided
by nearly all solutions. Only SWE and SenseWeb enable QoI attributes to be
attached as metadata to sensor readings or information which can be used as selec-
tion criteria during service discovery. Few proposals allow intermediate in-flow
processing services to be accommodated between the SANs and applications. This
is essential to facilitate high-level composition of context information and semantic
adaptation to be performed, which can even involve information coming from dif-
ferent sensor networks. The surveyed approaches provide limited or no support for
resource arbitration . In particular for actuation services, mechanisms to manage
concurrent access of resources are essential, but all proposals fail to explicitly
address these requirements with specific solutions. Finally, many of the surveyed
approaches fall short in addressing management support within the framework.
Only CoBis, JWebDust, and USN provide some management tools within their
framework, which address some aspects of the overall system.
5.2.2
Prevailing Sensor Networks Integration Frameworks
SNIFs can be classified according to the system architecture into two main groups:
server - client SNIFs and peer-to-peer SNIFs. The former type of integration frame-
works can be described as a central system which requires data owners to register
their data sources to one central server. These sensing resources are updated peri-
odically to let the server know about their availability. When an application submits
a query to search for a service, the central server analyzes the query and finds the
appropriate sensor network, and then produces a response. The latter class of SNIFs
adopts P2P techniques where each WSN with a gateway acts as a peer. The main
goal of P2P overlay is to treat the underlying heterogeneous WSNs as a single uni-
fied network in which users can send queries without considering the details of the
network. User peers communicate with gateway peers in a P2P approach.
Search WWH ::




Custom Search