|
117 | 117 |
|
118 | 118 |  |
119 | 119 |
|
120 | | -系统 $H(z)=1+z^{-1}+z^{-2}$. 线性相位系统 $N=3$. |
121 | | -$$ |
122 | | -\theta(\omega)=-\frac{N-1}{2}\omega\\ |
123 | | -\Rightarrow \theta(\omega)=-\omega |
124 | | -$$ |
| 120 | +### 12 |
125 | 121 |
|
126 | | -$$ |
127 | | -\theta(0)=\arg H(e^{j0})=\arg 3=0\\ |
128 | | -\theta(\pi)=\arg H(e^{j\pi})=\arg 1=0\not=-\pi |
129 | | -$$ |
| 122 | +Given the following signal graph for an FFT algorithm |
| 123 | + |
| 124 | + |
| 125 | + |
| 126 | +1. The graph represents a decimation-in-<u>time</u> algorithm. |
| 127 | + |
| 128 | +2. Complete the graph with arrowed lines and coefficients. |
| 129 | + |
| 130 | +  |
| 131 | + |
| 132 | +3. Using the above graph, compute the 4-point DFT of the sequence $x(n)=\delta(n)+\delta(n-1)$. |
| 133 | + $$ |
| 134 | + X(k)=\sum_{n=0}^{3} x(n)W_{4}^{nk} |
| 135 | + $$ |
| 136 | + |
| 137 | + - $X(0)=x(0)+x(1)=2$. |
| 138 | + - $X(1)=x(0)+x(1)W_{4}^1=x(0)+jx(1)=1+j$. |
| 139 | + - $X(2)=x(0)+x(1)W_{4}^2=x(0)-x(1)=0$. |
| 140 | + - $X(3)=x(0)+x(1)W_{4}^3=1-j$. |
| 141 | + |
| 142 | +### 13 |
| 143 | + |
| 144 | +Use Butterworth filter with impulse invariance to design a low-pass filter with |
| 145 | + |
| 146 | +- Passband edge frequency: $0.2\pi$ |
| 147 | + |
| 148 | + Passband attenuation: $1\mathrm{~dB}$ |
| 149 | + |
| 150 | +- Stopband edge frequency: $0.3\pi$ |
| 151 | + |
| 152 | + Stopband attenuation: $10\mathrm{~dB}$ |
| 153 | + |
| 154 | +The sampling interval is $T=1\mathrm{~s}$. |
| 155 | + |
| 156 | +根据 $\omega=\Omega T$ 可得 $\Omega_p=0.2\pi,\Omega_{st}=0.3\pi$,因为低通滤波器频带位于 $[-\pi/T,\pi/T]$ 之间,所以没有混叠。 |
| 157 | + |
| 158 | +计算巴特沃斯滤波器阶数和中心频率: |
| 159 | + |
| 160 | +- $N\ge \lg \left.\left(\frac{10^{0.1A_s}-1}{10^{0.1R_p}-1}\right)\right/2\lg (\Omega_{st}/\Omega_p)$. 得到 $4.38$ 取 $N=5$. |
| 161 | +- $\Omega_c=\Omega_p/\sqrt[2N]{10^{0.1R_p}-1}=0.719$. |
| 162 | + |
| 163 | +确定归一化巴特沃斯滤波器 $H_{an}(s)$,去归一化得到 $H_{a}(s/\Omega_c)$. |
| 164 | + |
| 165 | +使用冲激响应不变法,设计数字滤波器。 |
| 166 | + |
| 167 | +### 14 |
| 168 | + |
| 169 | +1. Derive the decimation-in-time FFT algorithm with $N=4$ and plot its signal flow graph. |
| 170 | + $$ |
| 171 | + \begin{aligned} |
| 172 | + X[k]&=\sum_{n=0}^{N-1} x[n] W_{N}^{nk}\\ |
| 173 | + &=\sum_{m=0}^{N/2-1} (x[2m] W_{N}^{2mk}+x[2m+1]W_{N}^{2mk+k})\\ |
| 174 | + &=X_1[k]+W_{N}^{k} X_1[k] |
| 175 | + \end{aligned} |
| 176 | + $$ |
| 177 | + |
| 178 | + $$ |
| 179 | + X[k+N/2]=X_1[k]-W_{N}^{k} X_1[k] |
| 180 | + $$ |
| 181 | + |
| 182 | +  |
130 | 183 |
|
0 commit comments