基于卡尔曼自回归的LTE天线端口数检测算法

在长期演进(long term evolution， LTE)系统中，传统天线端口数检测使用盲检测的方式分别对1，2和4端口数进行解码，直至物理广播信道系统消息成功通过循环冗余码校验，该方法会产生大量的计算冗余和时延。针对这一问题，提出一种改进的卡尔曼自回归天线端口数检测算法，该算法通过提取不同天线端口对应的小区参考信号得到信道状态信息，并将信道状态的相位信息进行卡尔曼自回归拟合，将自回归拟合后的相位与接收信号相位作差得到的平均值与预设判决门限进行比较，以得到天线端口数判决结果。理论分析与仿真结果表明，改进算法相较于传统盲检测算法节省的时间开销可达49%。在相同信噪比下，相较于其他优化后的门限判决算法，改进算法最多提高约10%的检测成功率且具有更优的抗频偏性能。

LTE antenna port number detection algorithm based on Kalman autoregression filtering

In long term evolution (LTE) system, the traditional antenna port number detection algorithm used blind detection to decode the 1, 2, and 4 antenna ports respectively until the system message in physical broadcast channel successfully passed the cyclic redundancy check. This method generated a large amount of computational redundancy and delay. To solve this problem, an improved Kalman autoregressive antenna port number detection algorithm was proposed, which obtained channel state information by extracting cell reference signals corresponding to different antenna ports, and carried out Kalman autoregressive fitting on the phase of channel state information. Finally, the average value of phase difference between autoregressive fitting and received signal is compared with the preset decision threshold, and the decision result of the number of antenna ports is obtained. Theoretical analysis and simulation results showed that the proposed algorithm saved up to 49% of time cost compared with the traditional blind detection algorithms. With the same signal to noise ratio (SNR), compared with other threshold decision algorithms, the improved algorithm can improve the detection success rate by up to 10% and has better anti-frequency bias performance.