超密集网络中基于Stackelberg博弈的非统一定价功率控制

Non-unified pricing power control based on Stackelberg game in ultra-dense network

针对超密集网小区同频部署中产生的区间干扰问题,提出基于Stackelberg博弈的非统一定价功率控制方案。首先,建立基于Stackelberg博弈的非统一定价功率控制模型,并通过该模型求解出基于价格的最优发射功率,此发射功率是关于干扰价格的函数;接着通过引入拉格朗日函数求解出相应基站的最优干扰价格,由控制器把干扰价格发送给相应的基站,并由基站调整自己的发射功率值去减弱对当前用户的干扰。仿真结果表明,与基于Stackelberg博弈的统一定价功率控制方案相比,所提方案在系统平均中断概率上平均下降了3个百分点。与基于基站权重的功率控制方案相比,在基站数量低于105时,所提方案在平均中断概率平均上上升了1.4个百分点;而当基站数超过105时,所提方案在平均中断概率上平均下降了1.6个百分点。此外,与这两个方案相比,所提方案在系统平均吞吐量上分别提升了12个百分点和10.5个百分点;在系统平均频谱效率上分别提升了9个百分点和8.5个百分点;在系统平均功率效率上分别提升了13个百分点和12个百分点。实验结果表明所提方案能够在部署更多基站情况下更好地提高蜂窝系统的性能。

Aiming at the problem of inter-zone interference in co-frequency deployment of ultra-dense network cells,a non-uniform pricing power control scheme based on Stackelberg game was proposed. Firstly,a non-uniform pricing power control model based on Stackelberg game was established. Then the optimal price based transmission power was obtained by this model,which is a function of interference price. Secondly,the Lagrange function was introduced to solve the problem of the optimal interference price of the corresponding base station, and the controller sent the interference price to the corresponding base station,and the base station adjusted its own transmission power value to weaken the interference to the current users. The simulation results show that compared with the unified pricing power control scheme based on Stackelberg game,the proposed scheme reduces the average outage probability of the system by an average of 3 percentage points.Compared with the power control scheme based on the weight of the base station,when the number of base stations is less than105,the proposed scheme increases the average outage probability by an average of 1. 4 percentage points; when the number of base stations exceeds 105,the proposed scheme reduces the average outage probability by an average of 1. 6 percentage points. In addition,compared with the two scheme mentioned above,the proposed scheme increases the average throughput of the system by 12 percentage points and 10. 5 percentage points respectively,the average spectrum efficiency of the system is increased by 9 percentage points and 8. 5 percentage points respectively,and the average power efficiency of the system is improved by 13 percentage points and 12 percentage points respectively. The experimental results show that the proposed solution can better improve the performance of the cellular system when more base stations are deployed.