Driver’s ECG Signal Detection and Transmission by Impulse-Radio-Based Human Body Communication Technology

In this study, we developed a wearable electrocardiogram (ECG) sensor with human body communication (HBC) technology for vital data transmission in a car. The ECG signals were modulated with wideband pulse signals between 10 and 60 MHz based on an impulse radio (IR) scheme, which provides a data rate as high as 1.25 Mbps. To apply the HBC-based wearable ECG in a moving car, we employed a moving average algorithm to reduce the noises produced in the received ECG signals due to the car’s movement. As a result, we con(cid:28)rmed a real-time ECG transmission of driver in the moving car with a reasonable accuracy.


INTRODUCTION
There is a wide-scale demand for health-state monitoring in various environments [1] [2].One of them is the monitoring of driver's health-state for safe driving.In this scenario, some vital sign sensors are placed on the driver's body to collect signals such as ECG, blood presser and pulse rate.
The ECG sensor may be embedded in the driver's seat belt, and the access point may be embedded in the steering wheel so that the driver unconsciously wears the sensors and send the data to an access point by a wireless or HBC technology [3]- [5].The car's control unit can then analyze the driver's    Its spectrum shape is formed between 10 and 60 MHz with appropriate ltering.In order to transmit the digitized data bits, we employ multiple pulse positions to represent bits 1 and 0 , which is actually a pulse position modulation (PPM) scheme at a date rate of 1.25 Mbps.The HBC receiver employs an envelope detector for demodulation.The received signal is ltered and amplied, and is then adjusted to an adequate level by an automatic gain controller for demodulation.Table 1 summarizes the basic specications of the IR-type HBC transceiver.

VERIFICATION EXPERIMENT
The validity of the developed HBC-based wearable ECG was veried by comparison of the measured results between our wearable ECG and a commercial available Holter ECG

ECG TRANSMISSION IN CAR
After the performance validation for our developed wearable ECG, we applied it in a moving car to acquire and transmit the driver's ECG signal to a personal computer in the front of the car.Fig. 3 shows the view of the experiment.The wearable ECG was set on the chest of the driver.When the driver's hand touched the receiver electrodes, his ECG signal was detected and sent to the receiver and then displayed on the PC screen in a real time.
Fig. 4 shows two examples of acquired ECG signals.One is that when the car stopped, and the other is that when the car ran at a speed of about 30 km/h.As can be seen in the gures, not only the Q-, R-, S-waves but also the P-and Twaves were observed in the ECG signals even when the car was running.The results demonstrate a sucient feasibility of our wearable ECG for driver's ECG monitoring.However, it was also found that more noise components appeared in the ECG signals when driving.The P-wave and T-wave were especially aected by the noises.By performing a Fourier    The future subject is to further improve the ltering algorithm for removing various possible noises in driving.

Fig. 1
Fig. 1 shows the structure of our developed HBC-based wearable ECG.The ECG sensor is composited of two 3 cm × 3 cm square electrodes.The two electrodes are attached to the chest for ECG signal sensing.A ground electrode is further attached to the human body as a reference.The ECG signals acquired by the two sensing electrodes are ltered and dierentially amplied in the ECG detector, and then converted to digital signals by an analog-to-digital (AD) converter.The AD-converted ECG signals are transmitted by the HBC transmitter to a HBC receiver through the hu-

Fig. 2
shows the view of ECG acquisition and transmission experiment.This experiment was conducted for three persons (A, B and C).For the acquired ECG signals from the three persons, we made a heart rate variability analysis, and extracted the inter-beat (RR) interval (RRI) and the number that the RR dierence is larger than 50 ms per minute (RR50).Table 2 and 3 compare the RRI and RR50 between our wearable ECG and the Holter ECG.As can be seen, the correlation coecient is higher than 0.953 for RRI, and the average dierence is only 7.0% for RR50.Moreover, we also compared the low-frequency (0.05 -0.15 Hz) to high-frequency (0.15 -0.40 Hz) (LF/HF) ratio for the acquired ECG signals.The LF/HF ratio is usually used as an index of cardiac sympathetic modulation.

Figure 3 :Figure 4 :
Figure 3: ECG acquisition and transmission in a moving car.

Figure 6 :
Figure 6: ECG signal processed by moving average lter when driving.

Table 2 :
Comparison of correlation coecient of RRI

Table 3 :
Comparison of RR50 (Sample/min) (Fukuda Denshi Co. Ltd.).We set both our wearable ECG and the Holter ECG on the chest, and acquired the ECG signal simultaneously.

Table 4
compares the LF/HF ratio between our wearable ECG and the Holter ECG.Again, the average dierence is only 4.4%.Such a good correlation and low dierences suggest that our wearable ECG has the same performance as commercial Holter ECG.