Abstract: This paper designs the cooperative spectrum prediction strategy in multiple Primary Users（PU） and multiple Secondary Users（SU） Cognitive Radio Network（CRN）. The problem of cooperative spectrum prediction strategy is formulated as an optimization problem. The objective of this problem is maximize the throughput of the whole CRN, the number of SU that participating the cooperation and the fusion threshold are the optimization parameters. An exhaustive searching algorithm is proposed to solve this problem. This paper is the first work to study the design of cooperative spectrum prediction strategy. This paper further proposes the group-based cooperative spectrum prediction strategy and design the frame structure accordingly. Simulation results show that the proposed group-based cooperative spectrum prediction achieves higher throughput while consumes less energy in cooperation, when meet the requirement of the probability of miss detection. The efficiency is improved under the group-based cooperative spectrum prediction.

Key words: cooperation spectrum prediction, group-based cooperative spectrum prediction, cognitive radio network, fusion threshold

摘要： 在多主用户（PU）、多认知用户（SU）的认知无线网络（CRN）场景中，研究了频谱预测的协作策略设计问题。将频谱预测的协作策略设计问题归纳为一个优化问题，以整个CRN的吞吐率最大为优化目标，对参与协作的SU数量和数据融合门限进行优化，并提出一种搜索算法求解此优化问题，这是首次对协作频谱预测中的协作策略进行研究。在此基础上，提出了基于分组的协作频谱预测策略，并设计其帧结构。仿真结果表明，这种基于分组的协作频谱预测策略在满足虚警概率要求的前提下，能够以更低的协作预测能耗，达到更高的CRN吞吐率，从而提高了协作频谱预测的效率。

关键词: 协作频谱预测, 基于分组的协作频谱预测, 认知无线网络, 数据融合门限