Sec-TEEN: Secure Threshold sensitive Energy Efficient sensor Network protocol Ibrahim Alkhori, Tamer Abukhalil & Abdel-shakour A. Abuznied Department of Computer Science and Engineering University of Bridgeport, Bridgeport, Connecticut Introduction Converting TEEN into SecTEEN Wireless sensor Networks (WSN’s) as illustrated in fig. 1, are composed of a huge number of sensor nodes with limited energy resources. Designing an energy efficient protocol is a key issue that needs to be dealt with in order to expand the life span of the entire network. We propose to analyze the performance of LEACH (Low Energy Adaptive Clustering Hierarchy) as well the TEEN (Threshold sensitive Energy Efficient sensor Network) protocols. Further we discuss the security features of SEC-LEACH and we propose adopting the authentication mechanism of SEC-LEACH to the TEEN. We show improvement by running experimental testing with multiple rounds. The new proposed SEC-TEEN improves the network security in order to make it more robust to external threats. In our proposal, we used the security techniques given by Sec-LEACH to design our secure TEEN protocol. In other words, we propose to generate a large pool of S keys and their ids prior to network deployment. Each node is then assigned a ring of m keys drawn from the pool pseudo-randomly, without replacement. The new TEEN algorithm can then be run with the modifications described below. When a selfelected CH broadcasts its adv message, it includes the ids of the keys in its key ring. The remaining nodes now cluster around the closest CH with whom they share a key. During the selfelection, CH broadcasts its id and a nonce. Then each ordinary node in the network computes the set of key ids, chooses the closest CH with whom it shares a key, and sends a CH a join-req message, protected by a MAC. The MAC is generated, and includes the nonce from CH’s broadcast to prevent replay attacks, as well as the id r of the key chosen to protect this link (so that the receiving CH knows which key to use to verify the MAC). At the end of the setup phase, the CHs send the time slot schedule to the nodes that chose to join their clusters. In the steadystate phase, the CH broadcasts to its members along with its adv message the values of hard threshold and soft threshold. Fig. 1. Four main parts for a typical WSN. Simulation Model Sec-LEACH vs. TEEN When nodes communicate in a sensor network, LEACH balances the draining of energy during that communication. In a wireless sensor network, LEACH rotates CHs randomly among all sensors in the network in order to distribute the energy consumption among all the sensors. LEACH works in rounds. In each round, LEACH elects CHs using a distributed algorithm and then dynamically clusters the remaining sensors around the CHs. Consequently, Sensor-to-BS communication takes place using this clustering structure for the rest of the current round. Secure Low Energy Adaptive Clustering Hierarchy protocol (Sec-LEACH) is the same as LEACH protocol but with some modification to improve the level of security and to protect the network from many kinds of attacks such as sinkhole and selective forwarding attacks. SecLEACH randomly assigns each sensor with a ring of keys taken from key pool pseudo that have been generated at the time when the network is deployed. In TEEN routing protocol, nodes are arranged in hierarchical clustering scheme in which some nodes act as 1st level and 2nd level cluster heads. After forming the cluster head, it gets the attribute from the user. Once the attribute is received the cluster head broadcasts the attribute. These are hard and soft thresholds for sensed attributes (values) to its cluster members. Hard threshold is the minimum possible value of an attribute to trigger a sensor node to switch on its transmitter and transmit to the cluster head. The sensor nodes start sensing and transmitting the sensed data when it exceeds hard threshold. Nodes transmit only when the sensed attribute is in the range of their target. By doing that we are reducing the number of transmissions significantly. In terms of Data Overload, Sec-TEEN works with extra conditions for sending data at each node, such condition reduces the total transactions required in the network communications. This leads to highly reduced data overload compared with Sec-LEACH. This trend can be exhibited by analyzing the following graph; Referring to Fig. 3 below and after applying both algorithms on the network, we were able to achieve better results in terms of energy saving. When comparing the two protocols, we found that the SecTEEN performs much better than Sec-LEACH. SecTEEN has significant lower values of the energy consumed by the nodes than what is consumed by the Sec-LEACH. However, the two protocols have a remarkably considerable variation in such values in the case of energy saving. Fig. 3. Total energy consumption occur in Sec-LEACH and Sec-TEEN after 500 rounds, for network size (100). Our experiment shows that the variation of energy consumption is very significant for the simulation network of 100 sensors. With a 0.5 Joule energy initially given to each node in both protocols, we can tell that nodes in the Sec-TEEN tend to consume their energy at a slower rate if compared to the SecLEACH nodes. Consequently, as can be seen from the Fig. 3, a WSN of 100 nodes using Sec-TEEN have consumed all of their energy at time 246. However, the nodes have consumed the energy at time 224 for the SecLEACH. This improvement comes from the nature of how Sec-TEEN works. Using the hard as well as soft threshold values improved the performance of SecTEEN. Data are not sent when the sensed data do not belong to the range of desired values which saves the energy needed transmission. According to the energy saving analysis, we can easily figure out that the number of alive nodes at any point in Sec-TEEN will be much higher than Sec-LEACH. This can be depicted in Fig. 4, below; keeping in mind, the number of alive nodes depends on the energy consumption by the network. Fig. 2. Total data overhead in Sec-LEACH and SecTEEN after 500 rounds for network size (100 nodes). Our experiment shows that for the network of size 100 nodes, both Sec-LEACH and Sec-TEEN have different volume of data being transmitted between nodes to cluster heads and vice versa. Referring to Fig. 2, we find that both of the SecLEACH and Sec-TEEN produce different total data overhead in the specified network. From this point, when it comes to our Sec-TEEN protocol, it remarkably involves fewer amounts of data sent between nodes in the system. It is noticed from Figure 2 that after a period of time both protocols will start having a steady phase in terms of data overhead and that obviously is due to the energy being completely consumed by all of the nodes. The highest data volume sent by the Sec-TEEN was 1.76 x 107 bits where it was 3.1 x 107 bits in the case of Sec-LEACH. Fig. 4. Alive nodes in Sec-LEACH and Sec-TEEN after 500 rounds, for network size of 100 sensors. Conclusion We have proved using a simulation for WSN that SecTEEN has improved the network in terms of energy consumption when compared to Sec-LEACH. This was achieved by reducing the data overhead and reducing the transmission of packets by having soft & hard thresholds.