Energy-Efficient Multi-layer Adaptive Transmission with Entropy-Weighted TOPSIS for Diverse Contents
摘要
This paper investigates content-aware link selection and power control in multi-link wireless communication systems. To capture the diverse requirements of text, audio, and picture transmissions, we employ entropy-weighted TOPSIS with monotone benefit-type transformations, which enables interpretable weighting of error-related and throughput-related metrics. On top of this outer-layer selection, we introduce an inner-layer proportional power control scheme that tracks content-specific target SNR values in the dB domain. The layered design ensures that link selection and power control remain decoupled yet complementary, achieving both quality adaptation and energy efficiency. Extensive simulations demonstrate that the proposed framework improves the average goodput across diverse content types, reduces the frequency of link switching for enhanced stability, and simultaneously achieves stable SNR tracking with lower transmit power compared to baseline approaches.