ID 原文 译文
20535 该方法首先采用稀疏滤波器设计算法得到满足频域性能设计要求的FIR陷波原始滤波器系数， First, a sparse FIR notch benchmark filter that fulfills frequency response specifications is obtainedfrom the sparse filter design algorithm.
20536 然后对其进行CSD编码，并分析CSD编码量化系数集中所有的2项子式和孤子的灵敏度， Then, each quantized filter coefficient is represented in CanonicalSigned Digit (CSD). The sensitivities of all weight-two subexpressions and isolated nonzero digits of thequantized coefficient set are analyzed.
20537 最后根据灵敏度的大小依次选择合理的2项子式或孤子直接合成滤波器系数集。 Finally, the filter coefficient set with lower implementation cost isconstructed by iteratively admitting subexpressions and isolated nonzero digits according to their sensitivities.
20538 仿真结果表明，新算法设计实现的FIR陷波滤波器比已有的低复杂度设计方法最多可减少51%的加法器，有效地降低了硬件实现复杂度，大大节省了硬件资源。 Simulation results show that the proposed algorithm can save about 51% of adder compared with other low complexity filter design algorithms, which reduces effectively the implementation complexity and saves greatly the hardware cost.
20539 针对机载多输入多输出(MIMO)雷达杂波抑制问题，该文提出一种改进的基于知识辅助的空时自适应信号处理算法(KA-STAP)。 Focusing on the clutter suppression problem of the airborne Multiple Input Multiple Output (MIMO)radar, an improved method based on Knowledge-Aided Space-Time Adaptive signal Processing (KA-STAP)algorithm is proposed.
20540 根据杂波在空时2维平面的先验分布离线构造杂波子空间，以此替代基于扁长椭球波函数(PSWF)估计的杂波子空间，避免了复杂运算。 The clutter subspace is constructed offline according to the prior distribution of the clutter in the space-time plane, to replace that of estimation based on the Prolate Spheroidal Wave Function(PSWF), so that complex operations are avoided.
20541 仿真结果表明，所提方法不仅能减小运算量，还能获得更深的零陷以及更优的旁瓣性能。 Simulation results show that the proposed approach can notonly reduce the computational complexity, but can obtain deeper notch and better side-lobe performance.
20542 针对滤波器组系统硬件实现时原型滤波器的有限字长效应问题，该文研究如何改善FIR原型滤波器由信号量化引起的舍入噪声，即降低舍入噪声增益，提出一种FIR滤波器优化结构。 For the problem of the finite word length effect of prototype filters in hardware implementation ofthe filter bank system, this paper studies how to improve the performance of roundoff noise caused by signal quantization for the FIR prototype filter, that is, to reduce the roundoff noise gain. An FIR filter optimization structure is proposed.
20543 通过分析舍入噪声来源，利用多项式参数化方法对舍入噪声增益表达式进行推导。 By analyzing the source of roundoff noise, a polynomial parameterization method is usedto derive the roundoff noise gain expression.
20544 仿真实例证明，在不同字长约束条件下所提结构滤波器的幅频相频响应与理想状态基本吻合； The simulation example shows that the amplitude-frequency andphase-frequency response of the proposed structure filter are basically consistent with the ideal state underdifferent constraint of word length.