Intelligent cooperative spectrum sensing via hierarchical dirichlet process in cognitive radio networks

Xin Lin Huang, Fei Hu, Jun Wu, Hsiao Hwa Chen, Gang Wang, Tao Jiang

Research output: Contribution to journalArticlepeer-review

64 Citations (Scopus)


Cognitive radio (CR) is a critical technology for improving spectrum utilization and solving the radio spectrum scarcity problem. In CR devices, spectrum sensing is important to implement opportunistic spectrum access. Many spectrum sensing schemes have been proposed, including uncooperative, cooperative, centralized, and distributed algorithms. However, they aimed to obtain a global consensus sensing result, which may not always be possible in large-scale cognitive radio networks (CRNs) due to heterogeneous spectrum availability in different areas. Hence, some new spectrum sensing schemes should be designed to discover idle heterogeneous spectrum in CRNs. In this paper, we propose an intelligent cooperative spectrum sensing algorithm based on a non-parametric Bayesian learning model, namely the hierarchical Dirichlet process, which groups spectrum sensing data without the need to know the number of hidden spectrum states, and discovers a common sparse spectrum within each group. Furthermore, a concisely distributed information exchange scheme is designed, where intra-cluster and inter-cluster spectrum information is shared for global spectrum cognition. Experimental results show that the proposed algorithm can exploit the spatial relationship among sensed data to achieve a better spectrum sensing performance in terms of detection probability and false alarm probability.

Original languageEnglish
Article number6914577
Pages (from-to)771-787
Number of pages17
JournalIEEE Journal on Selected Areas in Communications
Issue number5
Publication statusPublished - 2015 May 1

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Intelligent cooperative spectrum sensing via hierarchical dirichlet process in cognitive radio networks'. Together they form a unique fingerprint.

Cite this