The difficulty of privacy protection in cyber-physical installations encompasses several sectors and calls for methods like encryption, hashing, secure routing, obfuscation, and data exchange, among others. To create a privacy preservation model for cyber physical deployments, it is advised that data privacy, location privacy, temporal privacy, node privacy, route privacy, and other types of privacy be taken into account. Consideration must also be given to other types of privacy, such as temporal privacy. The computationally challenging process of incorporating these models into any wireless network also affects quality of service (QoS) variables including end-to-end latency, throughput, energy use, and packet delivery ratio. The best privacy models must be used by network designers and should have the least negative influence on these quality-of-service characteristics. The designers used common privacy models for the goal of protecting cyber-physical infrastructure in order to achieve this. The limitations of these installations' interconnection and interface-ability are not taken into account in this. As a result, even while network security has increased, the network's overall quality of service has dropped. The many state-of-the-art methods for preserving privacy in cyber-physical deployments without compromising their performance in terms of quality of service are examined and analyzed in this research. Lowering the likelihood that such circumstances might arise is the aim of this investigation and review. These models are rated according to how much privacy they provide, how long it takes from start to finish to transfer data, how much energy they use, and how fast their networks are. In order to maximize privacy while maintaining a high degree of service performance, the comparison will assist network designers and researchers in selecting the optimal models for their particular deployments. Additionally, the author of this book offers a variety of tactics that, when used together, might improve each reader's performance. This study also provides a range of tried-and-true machine learning approaches that networks may take into account and examine in order to enhance their privacy performance.