研究生时的现代信号处理的作业，包括随机梯度算法，相对梯度算法，最大似然（ML）准则和最大后验概率（MAP）准则，多姿态，多角度，有不同光照,有信道编码，调制，信道估计等，结合PCA的尺度不变特征变换(SIFT)算法，是一种双隐层反向传播神经网络。- Modern signal processing jobs when the graduate, Including stochastic gradient algorithm, the relative gradient algorithm, Maximum Likelihood (ML) criteria and maximum a posteriori (MAP) criterion, Much posture, multi-angle, have different light, Channel coding, modulation, channel estimation, Combined with PCA scale invariant feature transform (SIFT) algorithm, Is a two hidden layer back propagation neural network.