Wireless sensor networks (WSN) behave as a digital skin, providing a virtual layer where the information about the physical world can be accessed by any computational system. As a result, they are an invaluable resource for realizing the vision of the Internet of Things (IoT). Many applications of sensor networks require secure communication. Thus establishing a secure channel between any two sensor-nodes in WSNs/IESNs is important for many applications, such as secure data exchange, secure data aggregation, and secure routing. In this paper, first we show why existing three party key establishment schemes cannot be easily applied to IESN. Second we propose an extension of traditional three-party key establishment schemes (such as SNEP, BBF and OR). The method provides DoS and Sybil attack resistance and benefits from low communication cost, independence of prior sensor deployment knowledge and support for node mobility. In comparison to the previous well-known three-party schemes, our extension not only fixes DoS vulnerability, but also provides some other advantages such as significant efficiency. The proposed key establishment scheme can be used not only for establishing shared key between any two sensors, but it is applicable for establishing shared secret between any two entities/ things in the context of IoT.

The proposed system introduces new social network privacy management models and it measures their human effects. Here it introduces a mechanism using clustering techniques which helps users to group their friends using policy management. Then it introduces new privacy management model which will give policies to other friends to find similar friends in the network. And thereby explored various ways that help users to find example friends. In addition, it will help to find privacy management models which can be further enhanced and also it helps to detect privacy sentiment of user. Assistant friend grouping will be done for effective friendship establishment. In a network user privacy will be maintained by setting privacy techniques. Privacy management models can be routinely customized to the privacy sentiment and done according to the needs of the user.