现代雷达
現代雷達
현대뢰체
MODERN RADAR
2009年
12期
40-42,48
,共4页
模糊神经网络%旁瓣抑制%混合学习算法
模糊神經網絡%徬瓣抑製%混閤學習算法
모호신경망락%방판억제%혼합학습산법
neural fuzzy network%sidelobe suppression%hybrid learning algorithm
研究了模糊神经网络在二相码旁瓣抑制中的应用,对网络的学习算法进行了改进,采用梯度下降算法优化规则前件参数,而用最小二乘算法优化规则后件参数.对13位巴克码进行的仿真结果表明,改进的算法具有极快的收敛速度,可获得60 dB以上的输出主副比,提高了雷达的探测性能.
研究瞭模糊神經網絡在二相碼徬瓣抑製中的應用,對網絡的學習算法進行瞭改進,採用梯度下降算法優化規則前件參數,而用最小二乘算法優化規則後件參數.對13位巴剋碼進行的倣真結果錶明,改進的算法具有極快的收斂速度,可穫得60 dB以上的輸齣主副比,提高瞭雷達的探測性能.
연구료모호신경망락재이상마방판억제중적응용,대망락적학습산법진행료개진,채용제도하강산법우화규칙전건삼수,이용최소이승산법우화규칙후건삼수.대13위파극마진행적방진결과표명,개진적산법구유겁쾌적수렴속도,가획득60 dB이상적수출주부비,제고료뢰체적탐측성능.
The application of fuzzy neural network (FNN) for sidelobe suppression in binary-coded radar is discussed in this paper. The learning algorithm is improved through gradient descent algorithm to optimise the parameters of the premise part of the fuzzy rule and adopting the least square algorithm to regulate the parameters of consequent part. Simulation results show that the im-proved algorithm has faster convergence speed, and over 60 dB mainlobe to sidelobe ratio is obtationed. As a result, the detection performance of radar is enhanced.
