Wireless Sensor Networks (WSNs) play a vital role in military spectrum sensing applications, such as surveillance of emitters and electronic warfare. However, the inherent information security challenges in WSNs, particularly concerning stream ciphers based on Linear Feedback Shift Registers (LFSRs), necessitate enhanced security measures. This research proposes an intrusion prevention system designed to enhance algebraic immunity of a geffe generator-based stream cipher for secure data transmission in military WSNs. The Geffe Generator utilizes three LFSRs to produce a keystream, and this work addresses its limitations related to correlation and vulnerability to brute force and correlation attacks. The proposed solution involves incorporating a σ-LFSR into the Geffe Generator and introducing a modified shrinking generator to increase non-linearity of generated keystream. MATLAB simulations were conducted to evaluate the proposed model's period, complexity, and correlation with individual keystreams. The results demonstrate that the proposed model exhibits a significantly higher period and reduced correlation compared to conventional Geffe Generators, making it a promising solution for enhancing the security of WSNs in military spectrum sensing applications.