Abstract

The high-speed development of the Internet of Things and robotic systems created growing need for energy-efficient signal processing algorithms. Conventional methods are inefficient in optimizing use of resources which translates to unnecessary power consumption and processing delay. Current methods such as conventional FFT based signal processing, simple threshold-based decision-making and non-adaptive communication protocols are hindered by heavy computational overhead and lack of dynamic adaptability. Use of independent machine learning models like simple CNNs and SVMs has proven to be inadequate in managing intricate real time robotic tasks effectively. This disadvantage calls for creation of more enhanced techniques that utilize deep learning and enhanced data transmission processes. This research suggests optimized algorithm for energy-efficient signal processing in IoT-based robotic systems. It combines state-of-the-art preprocessing, feature learning and AI-driven optimization methods including Long Short-Term Memory networks and Graph Neural Networks. Designed system includes energy-efficient communication protocol to reduce data transmission overhead while supporting real-time decision-making. With the help of these innovations system improves robotic operation accuracy and reliability with minimal power consumption. Performance is assessed based on normal performance measures such as accuracy, latency and power efficiency. Experimental outcomes show drastic improvement with accuracy of 98.5 percent, lower latency and higher power efficiency in comparison to traditional approach proving validity of proposed method. Proposed framework has the potential to aid design of sustainable and intelligent robotic systems for use in different industrial applications.