Citation
Adnan, Ali Idarous
(2017)
Secure geographic forwarding protocols for wireless sensor networks.
Doctoral thesis, Universiti Putra Malaysia.
Abstract
The Advancement of the micro-electro-mechanical system (MEMS), wireless
communication, and low-power electronic devices has facilitated the development
of multipurpose and low-cost sensor devices. These devices are deployed
in a target region, in hundreds or thousands with an objective of gathering
information and transmit that information through multi-hops, to remote
users/computer using a special pattern of communication known as routing protocol.
Indeed, the overall performance of WSNs’ routing protocol is significantly influenced
by the deployment of security techniques in the routing procedures that
intend to prevent routing attacks. Although, WSNs have been used in many sensitive
applications, the development of reliable security techniques to safeguard
the transmitted information are still a great challenge and unable to adapt to the
resource constraints of the sensor nodes. This is because the existing security
tools such as cryptographic and key management schemes are too expensive in
term of computational resources to be directly integrated into sensor nodes. Furthermore,
the multi-hop communication and the absence of centralized administration
increase WSN vulnerabilities against routing failures and complexities.
For this reasons, inefficient secure routing protocols would cause extreme performance
degradation when subjected to attackers in the communication processes.
In order to enhance network performance and improve the ability of secure routing
protocol on dealing with the existence of attackers, this research work proposes
a Bound collection Geographic Forwarding (BCGF) protocol designed for
WSN. The BCGF helps in reducing the participation of an attacker in the communication
process as well as prevent retransmission of control packets. Thereafter, extensive simulation experiments are carried out to evaluate the performance of
the proposed BCGF compared to the existing secure implicit geographic routing
protocols. The results demonstrate that the BCGF improve network performance
when the protocol is subjected to no attacker and increases packet delivery to the
destination when a single attacker is of concerned.
Furthermore, the Secure Region-based Geographic Routing (SRBGR) protocol is
proposed to increase the number of legitimate responders in the communication
process when the number of Sybil virtual nodes increase in the forwarding allocated
area, in which the number of attacker selection is high and packet drop
is very common. SBRGR proposes an extension of forwarding area beyond allocated
sextant for security purposes. Extensive simulation experiments show that
the proposed protocol achieves a higher performance in minimizing attacker selection
contrary to the other secure protocols.
Moreover, to build a trust communication between neighbouring nodes, during
the routing processes and prevent malicious nodes in dropping packets, a
Light-weight Trust-based Scheme (LTBS) is proposed. LTBS allows each node
to monitor and determine the trustworthiness of its neighbours based on packet
forwarding acknowledge. LTBS encourages cooperation between nodes while
thwarts misbehaving nodes in capturing the communication and create inconsistency
in packet forwarding process. Substantial simulations have been conducted
to evaluate the proposed scheme. The results show that the scheme
achieves better performance in reducing the attackers in their different magnitudes
and severities.
Fortunately, results of the simulation show that the proposed secure routing protocols
provide enhancements in network performance and security. It reduces
the communication overheads, end-to-end delay while improving the packet delivery
to the destination when no attackers are presented in the communication
link. Also, it detects the routing attacks such as black hole and Sybil nodes by
minimizing their selection in routing processes in contrary to the existing secure
routing protocols.
Download File
Additional Metadata
Actions (login required)
|
View Item |