Abstract: This paper introduces the principles of emerging sensors such as intelligent sensors, micro-sensors, multi-functional sensors, and multi-parameter sensors, and, based on the latest domestic and international developments, studies the application of these new sensors in signal detection. It points out existing problems in signal detection and outlines future development trends for these new sensors. Keywords: Weak signal detection; New sensors; Micro-sensors; Intelligent sensors; Networked sensors I. Introduction Sensor technology has developed into a new interdisciplinary science. Compared with traditional sensors, new sensors have a much broader scope. The continuous increase in information volume, the increasing variety of information, the continuous improvement in information transmission speed, the accurate extraction and processing of weak signals, and the acquisition of two-dimensional or three-dimensional image signals all place newer and more stringent demands on the application of sensors in signal detection. Many new sensors have emerged, such as networked sensors, multi-functional sensors, multi-parameter sensors, intelligent sensors, solid-state image sensors, ultraviolet fluorescence sensors, biosensors, robotic sensors, amorphous alloy sensors, microwave sensors, superconducting sensors, liquid crystal sensors, and X-ray sensors, playing an irreplaceable role in the field of signal detection. Sensors in the 21st century will inevitably move towards miniaturization, intelligence, and multifunctionality. II. Intelligent Sensors and Signal Detection Since their emergence in the 1980s, intelligent sensors have been a hot topic in the field of detection, now demonstrating their capabilities in distributed real-time detection, network detection, and multi-signal detection. The practical application of optical fiber materials, due to their time-division multiplexing, distributed sensing, high-temperature resistance, and corrosion resistance, has greatly improved the performance of intelligent detection systems through the combination of optical fibers and intelligent sensors. An intelligent sensor is a sensor with information detection, information processing, information memory, logical thinking, and judgment functions. It not only possesses all the functions of traditional sensors but also has many other functions such as data processing, fault diagnosis, nonlinear processing, self-calibration, self-adjustment, and human-machine communication. Its emergence is a result of the combination of microcomputers and sensors. Structurally, an intelligent sensor consists of three parts: a main sensor, auxiliary sensors, and a microcomputer hardware system. Taking an intelligent pressure sensor as an example, the main sensor is a pressure sensor that measures the pressure parameter being measured, while the auxiliary sensors are a temperature sensor and an environmental pressure sensor. When the temperature sensor detects the temperature change of the pressure-sensitive element due to changes in ambient temperature or the temperature of the measured medium during the operation of the main sensor, it corrects and compensates for the measurement error caused by the temperature change. The environmental pressure sensor measures the atmospheric pressure change of the working environment to correct its influence. The microcomputer hardware system amplifies, processes, stores and communicates with the computer for the weak signal output by the sensor. Usually, a traditional detection device can only detect one physical quantity, and its signal conditioning is achieved by several analog electronic circuits connected to the main detection unit; while a smart sensor can achieve all these functions, and is more compact, lower cost and higher performance. Compared with traditional sensors, smart sensors have the following advantages: (1) They have logical thinking and judgment, information processing functions, and can analyze, correct and compensate for errors in the detected values. Smart sensors can linearize nonlinear signals by looking up tables, can easily filter digital signals by using software-developed filters, and can also use software to achieve nonlinear compensation or other more complex environmental factor compensation, thus improving measurement accuracy; (2) They have self-diagnosis and self-calibration functions, which improves reliability. Intelligent sensors can detect the working environment and give alarm signals when environmental conditions approach the critical limit. They can also provide diagnostic information by analyzing the input signal status. When an intelligent sensor cannot work properly due to internal faults, abnormal phenomena or partial faults can be detected through internal testing. (3) Multi-sensor multi-parameter composite measurement can be realized, expanding the detection and application range. The microprocessor makes it easy for intelligent sensors to perform multiple signal operations. Its configuration function can make the same type of sensor work in the best state and perform different tasks in different situations. (4) Detection data can be stored and retrieved, making it convenient to use. Intelligent sensors can store a large amount of information for querying, including the device's historical information, catalog, test results, etc. (5) It has a digital communication interface, which can be directly connected to a computer to exchange information, facilitating information management. For example, it can remotely control the detection system and track the measurement working mode, and can also transmit measurement data to remote users. III. Multifunctional Sensors and Signal Detection Traditional single sensors can only measure one physical quantity. However, in many fields, in order to comprehensively and accurately reflect the object and environment, it is often necessary to measure multiple physical quantities at the same time. Multifunctional sensors are composed of several sensitive elements, which are not only small in size but also powerful in function. Based on the different physical or chemical effects and different characteristics of a sensitive element, multiple functions of the sensor can be realized. With the development of sensor and microfabrication technology, people can fabricate several sensitive elements on the same material or silicon wafer to make integrated multifunctional sensors. Multifunctional sensors mainly have the following different implementation principles and structural forms: (1) Several different sensitive elements are combined together to form a sensor that can measure several parameters at the same time. Each sensitive element is independent. For example, by combining temperature and humidity sensitive elements, temperature and humidity can be measured at the same time; (2) Several different sensitive elements are fabricated on the same silicon wafer to make an integrated multifunctional sensor. This type of sensor has high integration and small size. Since it is integrated on a chip, the working conditions of each sensitive element are the same, and it is easy to realize compensation and correction. This is the development direction of multifunctional sensors; (3) Different information is obtained by using the different effects of the same sensitive element. For example, when a coil is used as a sensitive element, it exhibits different capacitance and inductance under the action of materials with different permeability or dielectric constant; (4) The same sensitive element exhibits different characteristics under different excitations. For example, when the sensor is subjected to different excitation voltages or currents, or operates at different temperatures, its characteristics are different, and sometimes it can be equivalent to several different sensors. Some multifunctional sensors detect several pieces of information mixed together, and it is necessary to use signal processing methods to separate the various information. Multifunctional sensors are a new development direction in sensor technology, and many scholars are actively exploring this field. For example, several sensors can be reasonably combined to form a new sensor (such as a composite sensor composed of sensors that measure hydraulic pressure and differential pressure). The miniature three-terminal digital sensor is a type of sensor that uses z-elements and is composed of photosensitive elements, humidity-sensitive elements and magnetic-sensitive elements, used to measure a variety of high-precision and small-sized signals. It can output not only analog signals, but also frequency signals and digital signals. Starting from simulating the senses of organisms in nature, a series of multifunctional sensors with tactile, visual, auditory, thermal and the latest achievement—olfactory sense—have also been developed and applied. In the field of multifunctional tactile sensors alone, there are already multifunctional sensors such as artificial skin tactile sensors made of PVDF materials, non-contact sensing skin systems, and pressure-sensitive conductive rubber tactile sensors. Among them, the non-contact sensing skin system developed by MERRITT Systems in the United States employs non-contact ultrasonic sensors, infrared radiation proximity sensors, thin-film capacitive touch sensors, temperature and gas sensors, etc. By inserting multiple intelligent sensors into the flexible circuitry of the sensing skin, the robot's need to detect objects can be met, avoiding unnecessary contact and collisions. In the current development of artificial sensory systems, the development of artificial olfaction (i.e., the electronic nose) is far less satisfactory than other senses. The sensory signals received by olfaction are not singular; they are usually composed of hundreds to thousands of chemical substances, making the signal processing within the olfactory system extremely complex. The electronic nose uses a cross-selected sensor array and related data processing technology, combining a gas sensor array and artificial nerves to solve this problem. The electronic nose consists of an array of partially dedifferentiated gas-sensitive sensors and an appropriate pattern recognition system, representing an effective combination of gas-sensitive sensor technology and information processing technology. Gas sensors offer advantages such as small size, low power consumption, and ease of signal acquisition and processing. Gases or odors pass through a gas sensor array and are input to the signal preprocessing section of an electronic nose system, where the array response patterns are pre-processed and features are extracted. The pattern recognition section uses algorithms such as correlation methods, least squares methods, clustering methods, and principal component analysis to perform qualitative and quantitative identification of gases/odors. Materials science provides atomic, molecular, supramolecular, and biomimetic structures, enabling the design of high-performance novel sensors; microstructure transducers and integrated data preprocessing circuit systems in electronic technology facilitate signal processing; and information theory enables electronic noses to better analyze complex data and compare and distinguish it with standards. Electronic noses have broad potential applications, such as odor identification, quantitative detection of individual molecule concentrations in complex environments, and comprehensive parameter measurement of combustible gases, volatile organic compounds, or toxic mixtures mixed in the air. IV. Networked Sensors With the rapid development and widespread application of computer technology, network technology, and communication technology, networked testing technology, which combines automatic testing technology with these technologies, has emerged. Networked testing systems enable remote testing of large and complex systems, representing an inevitable trend in testing in the information age. In measurement and control systems, sensors are indispensable for information acquisition, and they too must adapt to this trend, leading to the concept of networked sensors. Networked sensors refer to sensors that implement the TCP/IP protocol at the field level, enabling field measurement and control data to access the network locally and be published and shared in real time within the network's reach. The goal of designing networked sensors is to adopt standard network protocols while organically combining sensor and network technologies through a modular structure. The analog signal output by the sensing element is converted from digital signal to digital signal and processed. Then, the network processing device encapsulates it into data frames according to the program settings and network protocol (TCP/IP), adds the destination address, and transmits it to the network through the network interface. Conversely, the network processor can receive data and commands from other nodes on the network, enabling operations on its own node. In this way, the sensor becomes an independent node in the measurement and control network. For a networked sensor to function as an independent node, possessing the configurability and interoperability of a network node, achieving local network connection, and even "plug and play," the key is the standardization of the network interface. In 1994, IEEE and NIST (the National Institute of Standards and Technology) jointly initiated and developed the "Smart Sensor Interface Standard" (IEEE 1451). This standard uses a common A/D or D/A converter as the sensor's I/O interface, converting analog signals from various sensors into data in a standard format. This data, along with a small memory—the Sensor Electronic Data Sheet (TEDS)—connects to the standard-defined processor target model—the Network Capable Application Processor (NCAP). This allows data to access the network according to network protocols. Using a common processor and digital-to-analog converter, it is not limited to specific sensors or networks. This processor with a standardized interface enables various common sensors to access the network and become independent nodes, possessing network node configurability and interoperability. V. Microsensors and Signal Detection Currently, sensor manufacturing is shifting from traditional structural design and production to microstructure design based on micromachining technology and simulation programs. In this era of rapid information expansion, the amount of information that needs to be collected and processed is increasing, placing higher demands on sensor performance (such as accuracy, reliability, and sensitivity). Simultaneously, for ease of system integration, sensors should also have standard output formats. Traditional, less functional, and larger sensors are struggling to meet these requirements and are gradually being replaced by various high-performance microsensors primarily made of silicon. The sensing elements of microsensors are typically on the micrometer scale and are fabricated using micromachining technology, including processes such as photolithography, etching, deposition, bonding, and encapsulation. Utilizing anisotropic etching, sacrificial layer technology, and LIGA processes, three-dimensional microstructures with significant differences between layers can be manufactured, including movable diaphragms, cantilever beams, bridges, as well as grooves, pores, and cones. VI. Conclusion As sensors evolve towards miniaturization, intelligence, networking, and multifunctionality, their applications in signal detection will become more widespread and feature new characteristics. It's clear that it's difficult to definitively categorize these emerging sensors as micro-sensors, intelligent sensors, or multifunctional sensors. Therefore, comprehensive development, multifunctionality, miniaturization, and intelligence represent the overall development trend of sensors. Currently, sensor technology is developing towards miniaturization, integration, multifunctionality, intelligent systematization, interactivity, and visibility. For edge sensor detection mechanisms and technologies arising from interdisciplinary research, the types of detected signals will become increasingly diverse, detection functions increasingly powerful, and detection accuracy increasingly higher. As the "eyes" of modern science, sensors will undoubtedly help and guide scientists to advance with leaps and bounds.