Empirical Analysis of Blockchain and Machine Learning Models for QoS Aware Security Architectures in Wireless Networks

Providing node and network level security in wireless environments requires implementation of highly complex algorithms. These algorithms include encryption models, privacy preservation models, secret sharing models, hashing algorithms, etc. Implementation and deployment of these models requires additional computational power, which results in increased end-to-end delay, increased energy consumption, reduced throughput, and increased routing load. Due to which overall Quality of Service (QoS) of the network is impacted. To improve the QoS, machine learning optimization models are implemented. These models aim at reducing redundancies during network communications, thereby assisting the network to have better QoS performance. It has been observed that blockchain-based networks tend to have better security performance than their non-blockchain counterparts. Thus, in this text an indepth survey of such blockchain based networks is done, wherein each network is compared in terms of security and QoS performance. Due to a wide variety of applications for these networks, it is necessary for researchers and network designers to select the most optimum blockchain implementation for their network. This text will assist such network designers to select application specific blockchain models for high security and QoS performance. Along with this, the underlying text also compares various machine learning models and evaluates them in terms of computational complexity and QoS performance. Finally, this text recommends the best combination of blockchain and machine learning models that can be combined in order to improve overall system performance. By referring this text, system designers will be able to improve their network performance by integrating blockchain and machine learning optimized solutions to their networks.