The IoT wave is on rise and it is considered as the biggest world changing computing ecosystem after the invention of Internet where the meaning of lifestyle is expected to be changed. IoT is now diffusing pervasively in most areas of life like smart home, smart cities, smart irrigation, smart healthcare etc. The concerned industry is trying to reap maximum benefits from this regime without putting extra efforts or investing much to make the related infrastructure secure and trustworthy. IoT end device, a.k.a smart object, is one component of this ecosystem, responsible to interact with the physical environment and gather the data, along with communication technologies, processing capabilities like fog or cloud computing and applications to interact with the device (s). It is possibility that such devices can be faulty, compromised or misbehaving because of internal or external factors like hardware malfunctioning or cyber-attacks. In this situation the data gathered and transferred by such devices can be disaster and challenging in decision making specifically in an area where the human life is involved like IoT healthcare. We have proposed a mathematical model to estimate the trust of such devices. Trust on IoT devices and gathered data from such trusted devices will boost the confidence of end users on this new computing regime; especially in healthcare environment. The estimated trust status (trusted, uncertain, and untrustworthy) will be saved in a database or CSV file with a timestamp to be used as reputation by healthcare applications. Patients are assigned their SOI based on their specific diagnoses and procedures performed during their medical encounter. Similarly, for a patient with heart diseases or having hypertension can be considered in extreme category with a value of γ = 1 if there is some deviation of readings. 

A vehicular ad hoc network (VANET) is the emerging technology that allows the drivers to keep the road safety throughout the journey. In VANETs, vehicles can collaborate with each other by exchanging the messages. When these messages are incorrect, drivers will have to face many serious problems which include traffic congestion and minor to fatal road accidents. Therefore, drivers need a method which provides the correct decision using risk analysis calculated from the vehicle context. For this purpose, we propose a new contextual risk-based decision methodology for vehicular networks. This methodology can be used to provide robust and reliable decisions.

Wireless Sensor Networks (WSNs) are typically formed by collaboration of the large amount of partial sensor nodes, which are connected through wireless medium. In wireless sensor network, security is an essential aspect because of its usage in applications like monitoring, tracking, controlling, surveillance etc. Secure communication is extremely crucial in delivering vital information accurately and on the time through resource constraint sensor nodes. In this paper, our contribution is threefold. Firstly, we have summarized the network layer routing attacks on WSNs. Secondly, we have provided taxonomy of secure routing protocols of WSNs. Thirdly, we have provided a qualitative comparison of existing secure routing protocols. Results show that most of the existing secure routing schemes are not very efficient due to various reasons like high-energy consumption, and large communication overhead. 

Validation and verification of security policies is a critical and important task to ensure that access control policies are error free. The two most common problems present in access control policies are: inconsistencies and incompleteness. In order to detect such problems, various access control policy validation mechanisms are proposed by the researchers. However, comprehensive analysis and evaluation of the existing access control policy validation techniques is missing in the literature. In this paper, we have provided a first detailed survey of this domain and presented the taxonomy of the access control policy validation mechanisms. Furthermore, we have provided a qualitative comparison and trend analysis of the existing schemes. From this survey, we found that only few validation mechanisms exist that can handle both inconsistency and incompleteness problem. Also, most of the policy validation techniques are inefficient in handling continuous values and Boolean expressions.