1. Introduction The periodic contraction and relaxation of the ventricles in the human heart cause the aorta to contract and relax, resulting in blood pressure propagating as a wave from the aortic root along the entire arterial system. This wave is called the pulse wave. The comprehensive information presented by the pulse wave, including its morphology, intensity, rate, and rhythm, largely reflects many physiological and pathological blood flow characteristics in the human cardiovascular system. Traditional pulse measurement uses pulse diagnosis, and traditional Chinese medicine pulse diagnosis is a highly effective application of pulse measurement in traditional Chinese medicine. However, it is greatly affected by human factors, resulting in low measurement accuracy. Noninvasive measurements, also known as indirect measurements, are characterized by the probe not penetrating the body and causing no trauma. They are usually performed externally, especially on the body surface, to indirectly measure physiological and biochemical parameters. Biomedical sensors are key devices for acquiring biological information and converting it into easily measurable and processable signals. A photoelectric pulse sensor is a pulse sensor made based on the photoplethysmography method. It indirectly detects the pulse signal by monitoring the light transmittance at the fingertip. Photoelectric pulse sensors have advantages such as simple structure, non-destructive operation, and good repeatability. This paper discusses the design and implementation of a photoelectric pulse sensor. 2. Principle and Structure of Photoelectric Pulse Sensor 2.1 Principle of Photoelectric Pulse Sensor According to Lambert-Beer's Law, the absorbance of a substance at a certain wavelength is directly proportional to its concentration. When light of a constant wavelength shines on human tissue, the light intensity measured after absorption, reflection, and attenuation by the tissue will, to some extent, reflect the structural characteristics of the irradiated tissue. The pulse is mainly generated by the dilation and contraction of human arteries. The fingertip has a high content of arterial components, and its thickness is relatively thin compared to other human tissues. Therefore, the light intensity detected after passing through the finger is relatively large. Thus, the measurement site of a photoelectric pulse sensor is usually the fingertip. A schematic diagram of the change in light absorption in a finger. Sensor Technology Dai Junwei et al.: Development and Noise Analysis of a Photoelectric Pulse Sensor. Finger tissue can be divided into non-blood tissues such as skin, muscle, and bone, and blood tissues. The light absorption of non-blood tissues is constant. In blood, the pulsation of venous blood is very weak compared to arterial blood and can be ignored. Therefore, it can be considered that the change in light after passing through the finger is only caused by the filling of arterial blood. Thus, under the illumination of a light source of constant wavelength, the pulse signal of the human body can be indirectly measured by detecting the light intensity transmitted through the finger. 2.2 Structure of a Photoelectric Pulse Sensor. In addition to being absorbed by the finger tissue, part of the light emitted from the light source is diffusely reflected back by the blood. The rest is transmitted out. Photoelectric pulse sensors can be divided into two types according to the light receiving method: transmissive and reflective. In the transmissive type, the emitting light source and the photosensitive receiving device are equidistant and symmetrically arranged, receiving transmitted light. This method can reflect the temporal relationship of heart rhythm well, but cannot accurately measure changes in blood volume. In the reflective type, the emitting light source and the photosensitive device are located on the same side, receiving light diffusely reflected back from the blood. This signal can accurately measure changes in intravascular volume. This article discusses the transmissive pulse sensor, focusing on the measurement of pulse signals. 3. Fabrication of Photoelectric Pulse Sensors 3.1 Photosensitive Devices Photoelectric pulse sensors have various implementation methods due to the use of different photosensitive elements. The main photosensitive elements include photoresistors, photodiodes, phototransistors, and silicon photovoltaic cells. In traditional photoelectric pulse sensor designs, independent photosensitive elements are typically used, utilizing the photoelectric effect of semiconductors to change the output current. However, the output current of these photosensitive elements is usually extremely low, making them susceptible to external interference. Furthermore, they place stringent requirements on the subsequent amplifier, necessitating a low no-load current output to avoid interference with pulse signal measurement. This means that ordinary amplifiers cannot be directly used at the back end of the photosensitive element. In this paper, a novel photosensitive element, OPT101, is used. This element integrates the photosensitive component and amplifier into a single chip. This integrated design effectively overcomes the influence of the no-load current output of the back-end operational amplifier on the output current of the photosensitive component. Moreover, the chip's output voltage signal can be adjusted via an external precision resistor, facilitating chip adaptation to the overall circuit design. Simultaneously, the integrated design reduces system power consumption. 3.2 Emission Light Source The photoelectric pulse sensor mainly consists of a light source, a photosensitive device, and corresponding signal conditioning and control circuitry. To fully utilize the device's performance, the selection of the light source and photosensitive element is comprehensively considered. The wavelength of the light source should fall within the wavelength range where the photosensitive element has high detection sensitivity. Figure 4 shows the optical wavelength response curve of the OPT101. The pulse signal is mainly caused by the filling of arterial blood. Changes in the content of reduced hemoglobin (Hb) and oxyhemoglobin (HbO2) in the blood will cause changes in transmittance. When the light absorption of oxyhemoglobin and reduced hemoglobin is equal, the intensity of transmitted light will mainly be caused by the contraction and dilation of arterial blood vessels. At this time, the pulse signal can be reflected more accurately. Figure 5 shows the light absorption curve of hemoglobin. It can be seen from the figure that the absorption coefficients of HbO2 and Hb in the blood for different wavelengths of light are significantly different, and the two curves have several different intersection points. Considering that the light absorption rate of hemoglobin is relatively low at a wavelength of 805 nm, the light intensity transmitted through the finger is relatively large, which is beneficial to the reception of the photosensitive device. Therefore, the wavelength of the emitting light source is selected as 805 nm. 3.3 Constant current source control circuit In the process of pulse signal measurement, in order to minimize the influence of the power supply fluctuation of the light source on the measured pulse signal, a constant current circuit [4] is needed to control the stable power supply of the light source so that the light intensity emitted by the light source is constant during the pulse measurement process. Figure 6 shows the constant current source circuit. In the circuit, the voltage value across R1 is always equal to the voltage value of the Zener diode D1. Therefore, the current value flowing through R1 is constant. The control makes the transistor Q1 in the amplification state. Then the current value flowing through the light-emitting diode D3 is constant. Therefore, the light-emitting diode D3 can output light with a stable intensity. 3.4 Pulse signal conditioning circuit The pulse signal output in the chip OPT101 is a mixed voltage signal of DC and AC superposition. The AC signal contains pulse information. Therefore, the signal conditioning circuit first filters out the superposition DC signal and then amplifies the AC signal. The DC signal can be filtered out by a capacitor. However, the capacitor may cause partial distortion of the pulse signal while blocking DC. Ideally, a subtractor should be used to filter out most of the DC level. Since the transmittance of different subjects' fingers varies, the measured DC level will also differ. Therefore, a corresponding DC level filter is needed. This paper uses a controllable DC level output and a subtractor to extract the pulse signal. After obtaining the AC signal containing the pulse signal, the pulse signal can be extracted using a simple amplification circuit and a low-pass filter circuit. 4. Experimental Measurement and Noise Analysis of the Photoelectric Pulse Sensor During the measurement process, the pulse signal measured at the front end is very weak and easily affected by external environmental interference. Therefore, it is necessary to analyze the interference noise of the pulse sensor and reduce interference from the technical perspective of photoelectric pulse sensor design to ensure accurate pulse signal measurement. The interference of the photoelectric pulse sensor mainly includes ambient light interference, electromagnetic interference, and motion noise during the measurement process. The following analysis, combined with experimental measurements, further examines these factors. 4.1 Influence of Ambient Light on Pulse Sensor Measurement In photoelectric pulse sensors, the light signal received by the photosensitive device includes not only the transmitted light signal of the pulse information but also the background light signal of the measurement environment. Since the light intensity change caused by arterial pulsation is much weaker than the change in background light, it is necessary to maintain a constant background light during the measurement process to reduce background light interference. The background light in the measurement environment includes ambient light and secondary reflection light caused during the measurement process. To reduce the influence of ambient light on pulse signal measurement, and considering the ease of use of the sensor, a sealed finger-cot packaging method is adopted. The entire shell uses an opaque medium and color to minimize the influence of external ambient light. To avoid the influence of secondary reflection light during the measurement process, a light-absorbing material is coated on the inner surface of the finger-cot sensor, which effectively reduces the interference of secondary reflection light. As can be clearly seen from the graph in Figure 7, the pulse waveform measured by the pulse sensor with the finger-cot shell is smoother. This is because the ambient light in the finger-cuff type pulse sensor is largely unaffected by external ambient light during the measurement process, and it effectively reduces secondary reflections, ensuring that the wavelength of light illuminating the finger is singular. Therefore, the obtained pulse signal is more stable, without significant overlapping clutter signals, and can well reflect the characteristics of the pulse waveform. 4.2 The Influence of Electromagnetic Interference on the Pulse Sensor The electrical signal containing pulse information obtained through photoelectric conversion is generally weak and easily affected by external electromagnetic signals. In traditional photoelectric pulse sensor circuits, since the photosensitive device and the first-stage amplifier circuit are separate, the signal transmission process is easily affected by external electromagnetic interference. Electromagnetic shielding is usually used in the first-stage amplifier circuit to eliminate electromagnetic interference. This system uses a novel photosensitive device, integrating the photosensitive device and the first-stage amplifier circuit within the chip, effectively suppressing the interference of external electromagnetic signals on the original pulse signal. Power frequency interference is the most common type of interference in circuits. Since the pulse signal changes slowly, it is particularly susceptible to interference from power frequency signals. Therefore, suppressing power frequency signal interference is one of the main measures to ensure the accuracy of pulse signal measurement. The frequency range of pulse signals is typically between 0.13 and 30 Hz, which is less than the power frequency of 50 Hz. Therefore, a low-pass filter can effectively filter out power frequency interference, which is easily implemented in signal conditioning circuits. Simultaneously, pulse modulation of the light source can be performed in the control circuit. This not only reduces system power consumption but also reduces external electromagnetic interference to a certain extent. After pulse signal data acquisition, power frequency interference can be further filtered out using data processing methods. 4.3 Motion Noise During Measurement During the measurement process, the finger and the photoelectric pulse sensor may typically move relative to each other, causing errors in pulse measurement. Motion noise errors can be reduced in two ways: first, by improving the mechanical resistance to motion of the finger-type sensor, for example, by making the finger sleeve fit more tightly on the finger and less prone to loosening; second, by reducing errors through algorithms from the perspective of pulse signal processing. Currently, the first approach is mainly used in sensor design. 5 Conclusion Non-invasive monitoring technology will be an important direction for the future development of medical engineering, and the human pulse signal contains rich physiological information, which has gradually attracted great interest from clinicians. Photoplethysmography (PPG) is an effective method for measuring pulse signals. It can also be used to measure physiological signals such as blood oxygen saturation, oxygen partial pressure, and cardiac output, providing strong technical support for clinical diagnosis. Recently, Japanese researchers have proposed indirectly measuring blood pressure by analyzing the correlation between pulse wave velocity and blood pressure, demonstrating the promising future of pulse detection applications. Key technologies in pulse detection include sensor design and the extraction of weak signals from the sensor output. This paper presents a preliminary exploration of pulse sensor design and yields promising experimental results. Experiments show that the proposed method can effectively measure pulse signals, providing a favorable foundation for further pulse information extraction.