1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
| """
CPD 儿童遗留检测：120GHz FMCW 雷达信号处理流程
基于相位变化检测呼吸与心跳
硬件参数（参考 indie.inc 120GHz 模块）:
- 中心频率: 120 GHz
- 带宽: 10 GHz
- 距离分辨率: c/(2*B) = 3e8/(2*10e9) = 0.015m = 1.5cm
- 帧率: 20 Hz
- 功耗: < 10 mW（待机模式）
"""
import numpy as np
from scipy.signal import butter, filtfilt, find_peaks
class CPD_Radar_Processor:
"""
120GHz FMCW 雷达 CPD 检测处理器
数据流:
ADC采样 → 距离FFT → 相位提取 → 呼吸带通滤波 → 心跳带通滤波
→ 活体判定 → CPD报警
"""
def __init__(self, config: dict = None):
self.config = config or {
'sample_rate': 20,
'range_bins': 64,
'breath_low': 0.15,
'breath_high': 0.5,
'heart_low': 0.8,
'heart_high': 2.0,
'detection_window': 10,
'min_breath_amp': 0.3,
'min_heart_amp': 0.05,
}
self.breath_b, self.breath_a = butter(
2, [self.config['breath_low'], self.config['breath_high']],
btype='band', fs=self.config['sample_rate']
)
self.heart_b, self.heart_a = butter(
2, [self.config['heart_low'], self.config['heart_high']],
btype='band', fs=self.config['sample_rate']
)
def extract_phase(self, range_fft: np.ndarray) -> np.ndarray:
"""
从距离FFT结果提取相位
Args:
range_fft: shape=(N_frames, N_bins), complex
Returns:
phase: shape=(N_frames,) 目标距离门的相位序列
"""
energy = np.abs(range_fft).mean(axis=0)
target_bin = np.argmax(energy)
phase = np.unwrap(np.angle(range_fft[:, target_bin]))
wavelength = 2.5e-3
displacement = phase * wavelength / (4 * np.pi)
return displacement
def detect_vital_signs(self, displacement: np.ndarray) -> dict:
"""
检测呼吸和心跳
Args:
displacement: 位移序列 (米), shape=(N,)
Returns:
result: 包含呼吸率、心率、是否活体
"""
breath_signal = filtfilt(
self.breath_b, self.breath_a, displacement
)
residual = displacement - breath_signal
heart_signal = filtfilt(
self.heart_b, self.heart_a, residual
)
breath_freq = self._dominant_freq(
breath_signal, self.config['sample_rate']
)
heart_freq = self._dominant_freq(
heart_signal, self.config['sample_rate']
)
breath_amp = np.std(breath_signal) * 1e3
heart_amp = np.std(heart_signal) * 1e3
is_alive = (
breath_amp > self.config['min_breath_amp'] and
heart_amp > self.config['min_heart_amp'] and
self.config['breath_low'] <= breath_freq <= self.config['breath_high'] and
self.config['heart_low'] <= heart_freq <= self.config['heart_high']
)
return {
'breath_rate_bpm': breath_freq * 60,
'heart_rate_bpm': heart_freq * 60,
'breath_amp_mm': breath_amp,
'heart_amp_mm': heart_amp,
'is_alive': is_alive,
'target_bin': np.argmax(np.abs(
np.fft.fft(displacement)
))
}
def _dominant_freq(self, signal: np.ndarray, fs: float) -> float:
"""计算主频率"""
if len(signal) < 4:
return 0.0
fft = np.fft.rfft(signal)
freqs = np.fft.rfftfreq(len(signal), 1/fs)
return freqs[np.argmax(np.abs(fft))]
def classify_occupant(self, vital_signs: dict,
radar_cross_section: float) -> str:
"""
基于生命体征 + RCS 分类乘员
Args:
vital_signs: detect_vital_signs 输出
radar_cross_section: 目标RCS (dBsm)
Returns:
classification: 'adult' | 'child' | 'pet' | 'empty' | 'object'
"""
if not vital_signs['is_alive']:
if radar_cross_section > -20:
return 'object'
return 'empty'
hr = vital_signs['heart_rate_bpm']
if radar_cross_section > -10:
return 'adult'
elif radar_cross_section > -25:
if hr > 80:
return 'child'
else:
return 'pet'
else:
return 'child'
if __name__ == "__main__":
processor = CPD_Radar_Processor()
np.random.seed(42)
n_frames = 200
n_bins = 64
t = np.linspace(0, 10, n_frames)
breath = 0.001 * np.sin(2 * np.pi * 0.3 * t)
heartbeat = 0.0002 * np.sin(2 * np.pi * 1.2 * t)
noise = np.random.normal(0, 0.0001, n_frames)
displacement = breath + heartbeat + noise
phase = displacement * 4 * np.pi / 2.5e-3
range_data = np.zeros((n_frames, n_bins), dtype=complex)
range_data[:, 32] = np.exp(1j * phase)
disp = processor.extract_phase(range_data)
result = processor.detect_vital_signs(disp)
classification = processor.classify_occupant(result, -15)
print("=== 120GHz CPD 检测结果 ===")
print(f"呼吸率: {result['breath_rate_bpm']:.1f} BPM")
print(f"心率: {result['heart_rate_bpm']:.1f} BPM")
print(f"呼吸幅度: {result['breath_amp_mm']:.3f} mm")
print(f"心跳幅度: {result['heart_amp_mm']:.3f} mm")
print(f"活体检测: {'是' if result['is_alive'] else '否'}")
print(f"分类结果: {classification}")
print(f"\n检测窗口: {processor.config['detection_window']}s")
print(f"距离分辨率: 1.5 cm (120GHz, 10GHz带宽)")
|