"""
Cabin Breath Alcohol Sensor - Multi-Point Sampling System
多点采样座舱呼吸酒精传感器模拟
"""

import numpy as np
from typing import Tuple, List
from dataclasses import dataclass

@dataclass
class SensorReading:
    """单传感器读数"""
    sensor_id: str           # 传感器位置
    alcohol_ppm: float       # 酒精蒸汽浓度 (ppm)
    timestamp: float
    temperature: float       # 摄氏度
    humidity: float          # 相对湿度 %
    airflow_speed: float     # 气流速度 m/s

class CabinAlcoholDetector:
    """
    多点座舱酒精检测系统
    
    传感器部署：
    - 驾驶员方向盘柱（近距离呼吸区）
    - 副驾驶侧（区分乘客来源）
    - 后视镜区域（环境基线）
    - HVAC 出风口（通风系统影响）
    
    核心算法：
    1. 梯度分析：驾驶员侧 vs 其他位置的浓度差
    2. 时序模式：呼吸节律相关的浓度波动
    3. 环境补偿：温度/湿度/气流修正
    """
    
    def __init__(self):
        self.sensors = ['steering', 'passenger', 'mirror', 'hvac']
        self.driver_threshold_ppm = 15.0  # 驾驶员呼吸区阈值
        self.gradient_threshold = 5.0      # 梯度差阈值
        self.breath_pattern_freq = (0.1, 0.5)  # 呼吸频率 Hz
        
    def analyze(self, readings: List[SensorReading]) -> dict:
        """
        分析多点传感器读数，判定驾驶员是否饮酒
        
        Returns:
            dict: {
                'driver_alcohol_detected': bool,
                'estimated_bac': float,  # g/dL
                'confidence': float,
                'contamination_flag': bool,
            }
        """
        # 按传感器分组
        by_sensor = {}
        for r in readings:
            if r.sensor_id not in by_sensor:
                by_sensor[r.sensor_id] = []
            by_sensor[r.sensor_id].append(r)
        
        # 1. 梯度分析
        driver_avg = np.mean([r.alcohol_ppm for r in by_sensor.get('steering', [])])
        passenger_avg = np.mean([r.alcohol_ppm for r in by_sensor.get('passenger', [])])
        env_avg = np.mean([r.alcohol_ppm for r in by_sensor.get('mirror', [])])
        
        gradient = driver_avg - max(passenger_avg, env_avg)
        
        # 2. 呼吸节律检测（驾驶员侧浓度波动）
        driver_series = [r.alcohol_ppm for r in by_sensor.get('steering', [])]
        if len(driver_series) > 10:
            # FFT 检测呼吸频率分量
            freqs = np.fft.rfftfreq(len(driver_series), d=0.1)
            spectrum = np.abs(np.fft.rfft(driver_series))
            breath_mask = (freqs >= self.breath_pattern_freq[0]) & \
                          (freqs <= self.breath_pattern_freq[1])
            breath_power = np.sum(spectrum[breath_mask])
            total_power = np.sum(spectrum) + 1e-10
            breath_ratio = breath_power / total_power
        else:
            breath_ratio = 0
        
        # 3. 环境污染检测
        contamination = env_avg > 10.0  # 环境基线过高
        
        # 4. BAC 估计（简化模型）
        # 经验公式：BAC ≈ (driver_ppm - env_ppm) * calibration_factor
        calibration = 0.003  # ppm → g/dL
        estimated_bac = max(0, (driver_avg - env_avg) * calibration)
        
        # 5. 综合判定
        is_detected = (gradient > self.gradient_threshold and 
                       breath_ratio > 0.15 and
                       not contamination and
                       estimated_bac > 0.02)
        
        confidence = min(1.0, gradient / 20.0 * 0.4 + breath_ratio * 2.0 * 0.6)
        
        return {
            'driver_alcohol_detected': is_detected,
            'estimated_bac': round(estimated_bac, 4),
            'confidence': round(confidence, 3),
            'contamination_flag': contamination,
            'gradient': round(gradient, 2),
            'breath_ratio': round(breath_ratio, 3),
        }


# 测试
if __name__ == "__main__":
    detector = CabinAlcoholDetector()
    
    # 场景1：驾驶员饮酒（BAC ~0.08）
    readings_driver = [
        SensorReading('steering', 25 + 5*np.sin(2*np.pi*0.3*t*0.1) + np.random.randn()*2, 
                      t*0.1, 22, 45, 0.5)
        for t in range(50)
    ]
    readings_passenger = [
        SensorReading('passenger', 8 + np.random.randn()*1, t*0.1, 22, 45, 0.3)
        for t in range(50)
    ]
    readings_env = [
        SensorReading('mirror', 5 + np.random.randn()*1, t*0.1, 22, 45, 0.2)
        for t in range(50)
    ]
    
    result = detector.analyze(readings_driver + readings_passenger + readings_env)
    print("场景1 - 驾驶员饮酒:")
    print(f"  检测到: {result['driver_alcohol_detected']}")
    print(f"  估计BAC: {result['estimated_bac']:.4f}")
    print(f"  置信度: {result['confidence']:.1%}")
    
    # 场景2：乘客饮酒，驾驶员清醒
    readings_driver2 = [
        SensorReading('steering', 7 + np.random.randn()*1, t*0.1, 22, 45, 0.5)
        for t in range(50)
    ]
    readings_passenger2 = [
        SensorReading('passenger', 30 + np.random.randn()*3, t*0.1, 22, 45, 0.3)
        for t in range(50)
    ]
    readings_env2 = [
        SensorReading('mirror', 12 + np.random.randn()*2, t*0.1, 22, 45, 0.2)
        for t in range(50)
    ]
    
    result2 = detector.analyze(readings_driver2 + readings_passenger2 + readings_env2)
    print("\n场景2 - 乘客饮酒，驾驶员清醒:")
    print(f"  检测到: {result2['driver_alcohol_detected']}")
    print(f"  估计BAC: {result2['estimated_bac']:.4f}")
    print(f"  环境污染: {result2['contamination_flag']}")
    
    # 场景3：酒精洗手液
    readings_driver3 = [
        SensorReading('steering', 80 + np.random.randn()*5, t*0.1, 22, 45, 0.5)
        for t in range(50)
    ]
    readings_env3 = [
        SensorReading('mirror', 75 + np.random.randn()*5, t*0.1, 22, 45, 0.2)
        for t in range(50)
    ]
    
    result3 = detector.analyze(readings_driver3 + readings_env3)
    print("\n场景3 - 酒精洗手液污染:")
    print(f"  检测到: {result3['driver_alcohol_detected']}")
    print(f"  环境污染: {result3['contamination_flag']}")