Experimental Validation of Decision Tree-Based Adaptive Modulation and Coding Prediction for Multiple Users in Dynamic Wireless Environment

The increasing complexity of modern wireless
communication environments necessitates efficient and accurate
modulation and code rate classification to optimize system
performance. This paper investigates the use of a Decision Tree
(DT) algorithm for adaptive modulation and coding scheme
(MCS) prediction in multi-user scenarios. Leveraging real-time
datasets collected from a Software Defined Radio (SDR)
platform using USRP N210 and GNU Radio, the proposed
approach classifies modulation type and code rate based on key
channel parameters, including user distance from the base
station, channel noise, and signal-to-noise ratio (SNR). The DT
algorithm’s performance is evaluated using both Gini Index and
Entropy criteria, with results analyzed in terms of predicted
values, confusion matrices, and classification accuracy.
Experimental findings demonstrate that the Gini Index yields
superior accuracy compared to Entropy, achieving up to 100%
classification performance for certain parameters.
Furthermore, higher user counts (50 users) enhance prediction
accuracy due to richer training data. These results confirm the
potential of decision tree-based classification for real-time
adaptive MCS selection in dynamic wireless environments