| 摘要: |
| 智能反射面(Intelligent Reflective Surface,IRS)是未来6G通信增强网络覆盖和信道容量的潜在技术之一。然而由于智能反射面的无源设计使得大量的反射单元不具备发送、接收或处理信号的能力，对下行链路低复杂度的信道估计造成较大挑战。考虑IRS辅助的多天线系统下行链路信道估计，采用GaussSeidel（GS）迭代方法避免系统矩阵求逆，大大降低估计的复杂度。相比最小二乘（Least Squares,LS）算法，该算法计算复杂度阶数从基站天线数的立方降到基站天线数的平方。为了进一步评估性能，推导了相应的信道估计参数的CramerRao下界，最后通过仿真验证了所提算法的有效性。仿真结果表明,只需要较少的迭代数，GS就可以实现与LS算法相同的估计误差性能。 |
| 关键词: 多天线系统 下行链路 信道估计 智能反射面（IRS） |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目(61501315，62072325)；山西省基础研究计划面上项目(20210302123205) |
|
| Low-complexity channel estimation for IRS-aided multi-antenna downlink systems |
| LI Suyue,HAO Hongting,WANG Anhong |
| (School of Electronic and Information Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China) |
| Abstract: |
| Intelligent Reflective Surface(IRS) is one of the potential technologies for future 6G communications to enhance network coverage and channel capacity.However,due to the passive design of intelligent reflectors,a large number of reflection units do not have the ability to transmit,receive or process signals,which poses a great challenge to the low-complexity channel estimation of downlink.To extend the IRS single-antenna system model to multi-antenna system,Gauss-Seidel(GS) iterative method is adopted to avoid matrix inversion,which greatly reduces the complexity of estimation.Compared with the Least Squares(LS) algorithm,the computational complexity order of the proposed method is reduced from the base station antenna numbers cube to the base station antenna numbers square.In order to further evaluate the performance,the Cramer-Rao lower bound(CRLB) of the corresponding channel estimation parameters is derived,and the effectiveness of the proposed algorithm is verified by simulations.Simulation results show that GS can achieve the same error estimation performance as LS algorithm with fewer iterations. |
| Key words: multi-antenna system downlink channel estimation intelligent reflective surface(IRS) |
