With the ever-growing number of base stations(BSs)and user equipments(UEs)in ultra-dense networks(UDN),reusing the same pilot sequences among the cells is inevitable.With pilot reuse scheme,the channel estimation obta...With the ever-growing number of base stations(BSs)and user equipments(UEs)in ultra-dense networks(UDN),reusing the same pilot sequences among the cells is inevitable.With pilot reuse scheme,the channel estimation obtained at a BS contains not only the desired channel-state information(CSI)but also interference from neighboring cells,which can severely degrade CSI estimation performance and adversely affect communication performance.In this paper we consider a pilot contamination avoidance based on pilot pattern design for UDN where the pilot reuse employed and the interfering users from neighboring cells may be not at lower power levels at the BS compared to the in-cell users.We present a novel statistical interference model of sub-carriers to describe the non-deterministic interference from neighboring cells.Then,we provide a pilot pattern design model with non-uniform pilot distribution.Based on this,a pilot contamination avoidance based on pilot pattern design is proposed where pilot reuse scheme and the non-deterministic interference from neighboring cells are taken into consideration.Unlike existing interference mitigation approaches,the proposed method eliminates interference through the method of interference avoidance and can be applied to different kinds of channel estimation algorithms.Simulation results showed that the proposed approach can effectively avoid the interference and ensure the accuracy of channel estimation.展开更多
正交频分复用(OFDM,orthogonal frequency division multiplexing)是物联网的物理层关键技术之一,导频设计和信道估计是OFDM系统的关键问题。针对物联网通信场景复杂多样导致固定导频方案性能较差的问题,提出了一种基于图信号处理(GSP,g...正交频分复用(OFDM,orthogonal frequency division multiplexing)是物联网的物理层关键技术之一,导频设计和信道估计是OFDM系统的关键问题。针对物联网通信场景复杂多样导致固定导频方案性能较差的问题,提出了一种基于图信号处理(GSP,graph signal processing)的导频设计和信道估计方法。首先,将时频资源块建模为图信号,将信道估计问题转换为图信号的采样重建问题。进而考虑时频双选衰落的影响,设计加权图邻接矩阵,构造基于时频位置的图拓扑结构。在此基础上,基于图信号采样理论优化导频位置,提出一种基于加权图拓扑构造的导频图案设计方法。同时,基于图信号重建方法进行信道信息重建,提出基于图平滑性约束的信道估计方案。仿真结果表明,所提方法相较于传统方案在双选信道的高速场景下可取得更高的信道估计精度,在低速场景下则可有效节约导频资源。展开更多
基金This work was supported in part by the Chongqing Research Program of Basic Research and Frontier Technology under Grant cstc2019jcyj-msxmX0233in part by Science and Technology Research Program of Chongqing Education Commission of China under Grant KJQN201901125,Grant KJQN201901103in part by the Scientific Research Foundation of Chongqing University of Technology under Grant 2019ZD42,Grant 2019ZD63.
文摘With the ever-growing number of base stations(BSs)and user equipments(UEs)in ultra-dense networks(UDN),reusing the same pilot sequences among the cells is inevitable.With pilot reuse scheme,the channel estimation obtained at a BS contains not only the desired channel-state information(CSI)but also interference from neighboring cells,which can severely degrade CSI estimation performance and adversely affect communication performance.In this paper we consider a pilot contamination avoidance based on pilot pattern design for UDN where the pilot reuse employed and the interfering users from neighboring cells may be not at lower power levels at the BS compared to the in-cell users.We present a novel statistical interference model of sub-carriers to describe the non-deterministic interference from neighboring cells.Then,we provide a pilot pattern design model with non-uniform pilot distribution.Based on this,a pilot contamination avoidance based on pilot pattern design is proposed where pilot reuse scheme and the non-deterministic interference from neighboring cells are taken into consideration.Unlike existing interference mitigation approaches,the proposed method eliminates interference through the method of interference avoidance and can be applied to different kinds of channel estimation algorithms.Simulation results showed that the proposed approach can effectively avoid the interference and ensure the accuracy of channel estimation.
文摘正交频分复用(OFDM,orthogonal frequency division multiplexing)是物联网的物理层关键技术之一,导频设计和信道估计是OFDM系统的关键问题。针对物联网通信场景复杂多样导致固定导频方案性能较差的问题,提出了一种基于图信号处理(GSP,graph signal processing)的导频设计和信道估计方法。首先,将时频资源块建模为图信号,将信道估计问题转换为图信号的采样重建问题。进而考虑时频双选衰落的影响,设计加权图邻接矩阵,构造基于时频位置的图拓扑结构。在此基础上,基于图信号采样理论优化导频位置,提出一种基于加权图拓扑构造的导频图案设计方法。同时,基于图信号重建方法进行信道信息重建,提出基于图平滑性约束的信道估计方案。仿真结果表明,所提方法相较于传统方案在双选信道的高速场景下可取得更高的信道估计精度,在低速场景下则可有效节约导频资源。
