A space-efficient and multi-objective case-based reasoning in Cognitive Radio

摘要

Cognitive Radios (CRs) are intelligent mobile systems that can adapt to changing network environments. Adaptivity is achieved through reasoning and learning from past experiences. The conventional case-based reasoning (CBR) method requires a large storage space for the cases; however, embedded system have limited storage space. This work proposes a novel CBR method for improving the storage space efficiency. The CBR method is based on the Divide-and-Conquer technique. Our experiments show that the method can reach higher accuracy with a maximum of 797% improvement.