舰船电子对抗

随着移动通信技术的发展,通信业务需求增加,无蜂窝网络陷入频谱资源短缺的问题。为解决该问题,通过将认知技术引入到无蜂窝网络形成认知无蜂窝系统,次级无蜂窝网络可以利用频谱感知手段接入主级网络频谱,从而实现自身传输。然而现有的认知无蜂窝系统功率分配研究中均假设次级发射机能够准确感知主级用户频谱空洞,未考虑次级用户频谱感知非理想问题。针对该问题,考虑了非理想频谱感知,提出了一种认知无蜂窝系统功率分配方法。由于非理想频谱感知存在虚警及漏检概率,用户速率建模变得复杂且主次级网络间存在干扰,使得该功率分配优化问题难以采用传统凸优化算法进行求解。此外,随着次级用户的数目增多,问题复杂度急剧上升。因此,提出了一种基于多智能体平均场SAC(MAMFSAC)的功率分配算法。

With the development of mobile communication technology,the demand of communication service increases,and cell-free networks fall into the problem of shortage of spectrum resources.In order to solve this problem,cognitive cell-free system is formed by introducing cognitive technology into cell-free network.Secondary cell-free network can use spectrum sensing to access the primary network spectrum,and its own transmission is realized.However,the existing studies on power allocation of cognitive cell-free systems assume that secondary transmitters can accurately perceive the spectrum hole of the primary user,and the non-ideal spectrum perception of the secondary user is not considered.To solve the problem,this paper considers the non-ideal spectrum sensing,and proposes a power allocation method for cognitive cell-free system.Because of the false alarm and missed detection probability of non-ideal spectrum sensing,the user rate modeling becomes complicated and there is interference between the primary and secondary networks,which make the power allocation optimization problem difficult to be solved by the traditional convex optimization algorithm.In addition,as the number of secondary users increases,the complexity of the problem increases dramatically.Therefore,a power allocation algorithm based on multi-agent mean field soft actor critic(MAMFSAC) is proposed.