Home > Publications > Doctoral dissertation & Master thesis
Domestic Conference
Title Security Enhanced Key Schemes based on Clustering Routing Protocols in Wireless Sensor Networks
Degree Ph.D.
Author Saewoom Lee
Advisor Kiseon Kim
Graduation Date 2015.08.25 File
    Date 2017-03-06 12:07

Wireless sensor networks (WSNs) can be applied to a wide range of applications for safety monitoring of special spaces and buildings, traffic monitoring, environmental pollutant tracking, ocean and wildlife monitoring, and many military applications. Generally, WSNs are composed of a huge number of sensor nodes. In WSNs, sensor node with limited battery power senses phenomena. After sensing a phenomena, they transmit the data to the sink node, which collects all information. As WSNs are moving towards widespread deployment, various approaches have been proposed based on clustering routing protocol which has the advantage of node scalability and energy efficiency.

Based on clustering routing protocol, the first research issue is to propose a hybrid key scheme that uses both symmetric and public key schemes in order to take the advantage of the rapid calculation times of the symmetric key scheme and flexible key management of the public key scheme. By considering a hybrid key scheme, we can offer security into the TEEN protocol. Usually, the encrypted data is transmitted from the sending node, and then it decrypts at the receiving node. In hybrid key scheme, two different key schemes are used depending on the types of communication for the TEEN protocol. That is, the symmetric key scheme is applied to intracluster communication, and the public key scheme is used for intercluster communication. Also, to provide the assurance of the identities among communication nodes, we create hashed value generated from the hash function. This hashed value is used to authenticate the origin of the messages as a message authentication code. The simulation results show that network lifetime of the proposed hybrid key scheme decreases about 8% than the TEEN protocol and about 4% compared with the TEEN protocol with symmetric key scheme.

The next research issue is to propose a key renewal scheme with sensor authentication for clustered WSNs. In clustered sensor networks, member nodes as well as cluster heads (CHs) are likely to be compromised since they are deployed in an unattended environment. If a member node is compromised, its data can be revealed to the attacker, and the attacker can send falsified data to the BS via the compromised node. However, the compromising of CHs has a more severe impact on the network than that of the member nodes because CHs gather data from every member node. Therefore, a key renewal scheme that deals with the confirmation of sensor nodes is required. To apply the key renewal scheme with sensor authentication, we authenticate sensor nodes during the key establishment procedure first and then update the symmetric key periodically using hash functions. The simulation results show that the proposed key renewal scheme lengthens the network lifetime and has shorter activity time of attackers in spite of increasing the number of compromised nodes.

The last research issue is to propose energy efficient securing key setup procedure based on clustering routing protocol. By inserting mobile sensors into the traditional WSNs, they can connect separated networks and support disparate missions. In spite of these advantages, security in MWSNs becomes more complicated due to the mobile network topology. Mobile sensors are likely to be attacked by malicious attackers. Thus, all the key information can be revealed to the attackers through captured sensors. Also, they consumes more energy because they join or leave the cluster frequently. To apply energy efficient key setup procedure for mobile sensors, we reduce the number of transmission during the key initialization and update steps. Based on this proposed scheme, the simulation results show that proposed scheme seems to provide better network lifetime, confidentiality of the network, and data aggregation accuracy when the number of nodes is increasing.

광주과학기술원 한·러 MT-IT 융합기술연구센터 광주과학기술원정보통신공학부