无线通信信噪比估计算法研究与实现

发布时间：2018-04-02 16:47

本文选题：无线通信　切入点：信噪比估计　出处：《电子科技大学》2014年硕士论文

【摘要】：信噪比是无线通信信道环境和通信质量的有效评估指标,并可作为指导编解码以及数字解调算法选择和优化的重要依据。在衡量信道质量的参数中,信噪比实时测量性好,并且与信道的误码率、误帧率直接相关。本文首先基于高斯信道研究了最大似然估计、二阶四阶矩估计、高阶累积量估计以及最小均方误差估计信噪比估计算法在不同频偏、数据长度、采样倍数以及有无定时的情况下的估计性能,并分析总结出影响算法性能的影响因子。仿真时最大似然估计可以工作于多倍采样的情况下,在有频偏时性能下降;二阶四阶矩和高阶累积量估计法在信号定时理想的情况下具有良好的估计性能,增大数据量可以提高算法估计性能的稳定性;最小均方误差法在理想高斯白噪声下对理想定时后的信号才具有良好的估计性能,并且以上信噪比估计算法主要针对PSK调制信号具有较好的估计性能。针对以上分析结果,本文提出了基于信号频谱自身归一化的改进频偏估计算法,信号带宽、符号率估计算法以及信噪比估计算法。改进的频偏估计、信号带宽、符号率估计算法能很好的消除频偏、采样倍数对信噪比估计性能的影响,改进的信噪比估计算法适用于QAM和PSK调制,估计性能几乎不受频偏和采样倍数的影响,能很好的估计1-30d B范围的信噪比值。然后,基于DSP平台在不同实现条件下进行了以上算法的全数字实现。最大似然估计实际实现的过程中由于噪声的非完全高斯性,估计性能较差;二阶四阶矩和高阶累积量估计法在在信号定时恢复后能较好的估计信噪比值;最小均方误差法实际实现时性能不佳;改进的信噪比估计算法实现时具有良好的估计性能。改进的频偏估计算法,信号带宽、符号率估计算法能很好的估计发送端调制信号的频偏、带宽及符号率。最后,将信噪比估计算法应用于64QAM的自适应选择载波恢复算法中,使得解调性能得到了优化,稳定性提高,解调星座点更集中。
[Abstract]:Signal-to-noise ratio (SNR) is an effective evaluation index for wireless communication channel environment and communication quality, and can be used as an important basis for the selection and optimization of codec and digital demodulation algorithms.Among the parameters used to measure the channel quality, the SNR can be measured in real time, and it is directly related to the bit error rate (BER) and frame error rate (FER) of the channel.In this paper, the maximum likelihood estimation, the second order fourth moment estimation, the high order cumulant estimation and the minimum mean square error estimation SNR estimation algorithm are studied based on Gao Si channel at different frequency offset and data length.The performance of the algorithm is estimated by sampling multiple and with or without timing, and the factors that affect the performance of the algorithm are analyzed and summarized.In simulation, the maximum likelihood estimation can work in the case of multiple sampling, and the performance of the second order fourth moment and high order cumulant estimation method can be reduced when the frequency offset is high, and the second order fourth moment and the high order cumulant estimation method have good estimation performance in the case of ideal signal timing.Increasing the amount of data can improve the stability of the estimation performance of the algorithm, and the minimum mean square error method has good estimation performance for the signal after ideal timing under ideal Gao Si white noise.And the above SNR estimation algorithms have better estimation performance for PSK modulation signals.Based on the above analysis results, an improved frequency offset estimation algorithm, a signal bandwidth estimation algorithm, a symbol rate estimation algorithm and a signal-to-noise ratio estimation algorithm are proposed based on the normalization of the signal spectrum itself.The improved frequency offset estimation, signal bandwidth and symbol rate estimation algorithm can eliminate the frequency offset and the influence of sampling multiple on the SNR estimation performance. The improved SNR estimation algorithm is suitable for QAM and PSK modulation.The estimation performance is almost independent of the frequency offset and sampling multiple, and can estimate the SNR in the range of 1-30 dB.Then, the full digital implementation of the above algorithm is carried out based on the DSP platform under different implementation conditions.In the process of maximum likelihood estimation, the estimation performance is poor due to the incomplete Gao Si of noise, and the second order fourth moment and high order cumulant estimation method can estimate the signal-to-noise ratio better after the signal timing recovery.The minimum mean square error method has poor performance in practice and the improved SNR estimation algorithm has good estimation performance.The improved frequency offset estimation algorithm, signal bandwidth, symbol rate estimation algorithm can well estimate the frequency offset, bandwidth and symbol rate of the modulation signal at the transmitter.Finally, the SNR estimation algorithm is applied to the adaptive selective carrier recovery algorithm of 64QAM. The demodulation performance is optimized, the stability is improved, and the demodulation constellation is more concentrated.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【参考文献】