Application of clustering structure in the hierarchical spectrum sharing network based on cognitive radio

This paper firstly analyzes the impact which clustering and non-clustering structure is on sensing spectrum expenditure. Furthermore, the reason that clustering structure is adopted in hierarchical spectrum sharing network is also illustrated by numerical simulation. Secondly, a clustering algorithm, which can reduce detection expenditure, is proposed based on the similarity of the spectrum in such architecture. Additionally, clustering stability based on this algorithm is explored. Theoretical analysis and simulation result indicates that clustering stability is ensured only if threshold K is chosen appropriately.