Wireless communication networks are rapidly evolving toward high data rates, ultra-low latency, and enhanced connectivity, which has led to increased concerns over both security and energy consumption. Conventional cryptographic techniques, although reliable, introduce additional processing overhead and may not always be feasible for resource-constrained devices such as sensors, Internet of Things (IoT) nodes, or mobile terminals. At the same time, purely physical-layer techniques often improve secrecy but compromise energy efficiency by requiring additional power for artificial noise or complex beamforming. This paper addresses these challenges by presenting an Adaptive Pre-Processing Filter (APPF) mechanism designed to jointly optimize secrecy capacity and power consumption in wireless systems. The APPF operates at the transmitter side, dynamically tuning its filtering coefficients according to instantaneous channel state information (CSI) while considering both the legitimate receiver and potential eavesdroppers. The key idea is to maximize the secrecy capacity—the difference between the capacities of the legitimate and eavesdropper channels—while simultaneously ensuring that the transmit power remains within a predefined efficiency budget. A mathematical framework for the APPF is formulated, and an iterative optimization algorithm is proposed to achieve real-time adaptation. The approach is validated through analytical modeling and MATLAB-based simulations, showing that APPFs provide up to 28% improvement in energy efficiency compared with traditional artificial noise methods, without sacrificing secrecy. Moreover, APPFs are shown to enhance resilience against eavesdropping attempts under fading channels, making them highly suitable for emerging green communication systems. The results highlight the potential of APPFs in bridging the gap between information-theoretic secrecy and sustainable power usage, offering a practical and scalable solution for 5G, beyond-5G, and IoT-based wireless networks. The findings demonstrate that adaptive filtering is a promising alternative to conventional security mechanisms and could play a pivotal role in next-generation wireless architectures.