Cognitive radio networks (CRNs) are a potential remedy for the approaching spectrum shortage and the inefficient use of licenced frequency bands. However, the dynamic and shared nature of CRNs makes them vulnerable to various security threats, including Byzantine attacks. Consequently, this study proposes a method for spectrum sensing that requires little energy for detecting Byzantine attacks in CRN using machine learning. Cognitive radio networks are susceptible to malicious activities, such as Byzantine attacks, which can significantly degrade network performance. The proposed strategy utilizes machine learning techniques to accurately detect and classify Byzantine attacks, enabling efficient spectrum sensing and improving overall network security. By incorporating energy efficiency considerations, the strategy optimizes the resource allocation process, minimizing energy consumption while maintaining high detection accuracy. The investigational conclusions show that the suggested method is effective in identifying Byzantine assaults, which improves the stability and dependability of cognitive radio networks. Cognitive radio networks face potential threats from malicious activities, particularly Byzantine attacks, which can severely impact network performance. The proposed strategy leverages machine learning techniques to effectively identify and categorize Byzantine attacks, thereby enabling efficient spectrum sensing and enhancing overall network security. By leveraging the power of machine learning algorithms, the strategy accurately distinguishes between genuine and compromised nodes during spectrum sensing. It reduces the impact of Byzantine attacks on spectrum decision-making processes and helps maintain the integrity and reliability of spectrum allocation in cognitive radio networks. This ensures the integrity and authenticity of spectrum sensing information, safeguarding the network against unauthorized access and interference.