基于混合粒子滤波的载波估计算法

针对粒子滤波载波估计算法的高复杂度、粒子退化及贫化问题,提出了一种基于混合粒子滤波的载波估计方法.该方法引入多阶马尔科夫模型,采用多个非零均值高斯分布的加权和来近似重要性函数的最佳选择,并根据最大后验概率准则规范粒子的迭代计算.仿真结果表明,在非高斯噪声环境下,低轨卫星通信TDMA/DEQPSK(Time Division Multiple Address/Differential Quadrature Phase Shift Keying)数据帧非合作接收载波估计时,与基于经典粒子滤波的载波估计算法相比,提高了粒子"效率",在误码性能相当的情况下,有效降低了计算复杂度.

Carrier estimation algorithm based on novel hybrid particle filtering

To reduce the computation complexity and improve the weight degeneracy and loss of diversity about the particles in particle-filtering (PF) based carrier estimation, a novel hybrid particle filter based carrier estimation algorithm was proposed and the method for assigning particle weights of the proposed PF algorithm (PPF) was derived. In this algorithm, the m-order Monte Carlo Markov chain was introduced and the optimal importance function-the posterior importance function was approximated by a weighted sum of non-zero mean Gaussian distributions and the iterative computation of the particles was constrained by the maximum a posterior (MAP) criterion. In non-Gaussian environment, by applying the algorithm in the non-cooperative receiver of the time division multiple address/differential quadrature phase shift keying(TDMA/DEQPSK) data frame in low earth orbit (LEO) satellite communication system, it is found that the efficiency of the particles is improved and the computational complexity is reduced compared to the carrier estimation algorithm based on the traditional PF.