中文摘要：

构造认知用户的网络模型为马尔科夫随机场，应用BP算法和加权BP算法来协助分布式网络中的决策融合，利用加权的BP算法更有效地解决阴影衰落和恶意节点所导致的问题。这种方法的性能优于现有的分布式网络中的大数判决等其他多数算法的性能。采用MATLAB进行仿真，验证了分析结果。

英文摘要：

Model the network of secondary users as Markov random fields and to apply the belief propagation algorithm and the weighted belief propagation algorithm to facilitate decision fusion in a distributed manner, and use weighted BP to solve the shadow fading and malicious nodes more effectively. The performance of this method is superior to the existing major voting algorithm and the most of the others algorithms in distributed cognitive radio networks. Finally, a simulation with MATLAB has been done and verifies the results.