In this Paper, we use probability theory to analyse the challenge of trust management. In particular, Bayesian likelihood outperforms rule-based methods for describing confidence in sparse, self-organizing networks. In order to accurately estimate confidence among confused ideas, three methods are used. Bayesian inference, Bayes networks and the Dempster-Shafer model (DST). To begin, let's imagine a mobile dedicated network (MANET) in which previously established relationships between trusted insiders are used for network penetration purposes. To determine which nodes in a MANET can be trusted based on their historical behaviour, we use Bayesian inference. The reliability of each node is determined based on its next probability. Direct and indirect evidence forms the basis of DST confidence assessment. A trust management system eliminates the threat posed by low-trust malicious nodes. We demonstrate that the suggested schemes provide better dynamic dependability, throughput, end-to-end latency, etc. under unfavourable circumstances than the state-of-the-art systems. To implement Manet's trust management tactics, we resort to the Bayesian networks framework. Examining the level of certainty that can be assigned to each node in a Bayesian network is essential when trying to simulate one. As a result, the bad actors will be contained inside the trust management system. In an effort to make spectrum sensing and data transmission safer, we examine several methods of providing trust management in ad hoc networks with CR capabilities. Using a consensus-based weighted approach, we increase the trustworthiness of collaborative spectrum sensing. The dangers to network security from NFV are also covered.

Cloud computing, MANET, NFV.