Sensing data fusion processing system

Abstract

The invention relates to the technical field of signal acquisition and processing, in particular to a sensing data fusion processing system. The reconfigurable front end is connected with a plurality of sensors, the reconfigurable front end is used for extracting output data of each sensor, the reconfigurable front end comprises a reconfigurable preamplifier and an analog-to-digital converter, the reconfigurable preamplifier comprises an operational amplifier and a reconfigurable peripheral circuit, and the peripheral circuit is reconfigured through switching of a switch; the peripheral circuit comprises a multiplexer and an operational amplifier feedback network; the multi-channel gating device is time division multiplexing, is provided with a plurality of channels and is used for being connected with a plurality of sensors, and when the multi-channel gating device is connected with different sensors, the structure of the operational amplifier feedback network is changed through switching of a switch, so that the reconfigurable pre-amplifier changes a transfer function; based on the sensing fusion technology, the designed data processing system can receive various sensor data at the same time and carry out integrated analysis, multi-dimensional information receiving is achieved, and one-stop data management is provided.

Description

Technical Field

[0001] This invention relates to the field of signal acquisition and processing technology, and more specifically to a sensor data fusion processing system. Background Technology

[0002] With the development of Internet of Things (IoT) technology, many target behavior monitoring and status management systems have emerged. By integrating positioning, motion, and physiological parameter acquisition units into wearable devices, the system continuously monitors the location status, activity characteristics, and health parameters of target objects, and uploads the relevant data to the back-end system for analysis and processing via wireless communication.

[0003] However, due to the diverse data to be collected, various types of sensors need to be set up. Since the signal types of various sensors are different, existing technologies often configure independent front-end acquisition circuits and data processing modules for different sensors, resulting in a multi-channel distributed architecture for the overall system. This architecture has a high degree of complexity, which is not conducive to centralized processing and fusion analysis of multi-source data. Furthermore, the distributed data acquisition and processing method usually requires multiple signal reception and parallel processing, which increases the system's computing load and communication resource consumption, resulting in high power consumption and affecting the battery life of wearable devices. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings and deficiencies of existing technologies by providing a rationally designed sensor data fusion processing system that can solve the aforementioned problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: it includes a reconfigurable front-end, a data merging interface, and a data processing module; the reconfigurable front-end is connected to multiple different types of sensors for extracting the output data of each sensor; the reconfigurable front-end includes a reconfigurable preamplifier and an analog-to-digital converter, the reconfigurable preamplifier includes a single operational amplifier and reconfigurable peripheral circuitry, the peripheral circuitry being reconfigured through switching; the peripheral circuitry includes a time-division multiplexer and an operational amplifier feedback network; the multiplexer has multiple input channels, each connected to a corresponding sensor, for cyclically selecting the output signals of different sensors according to a preset period; the operational amplifier feedback network is connected to the output terminal of the multiplexer, the input terminal of the operational amplifier, and the output terminal, respectively, for changing its own circuit structure through switching when different sensors are connected to the multiplexer, thereby changing the reconfigurable front-end. The transfer function of the preamplifier is reconstructed to adapt to sensor output signals of different modes (voltage, current, resistance), and these signals are uniformly converted into voltage signals and output to the analog-to-digital converter (ADC). The input of the ADC is connected to the output of the operational amplifier, and is used to acquire the output signal and quantize it to generate digital code after each input channel switch of the multiplexer and the establishment of the reconstructible preamplifier output. The input of the data merging interface is connected to the output of the ADC, and is used to convert the multi-channel parallel digital code generated by time-division acquisition into a single-channel serial digital code. The data processing module is electrically connected to the switches in the multiplexer, the operational amplifier feedback network, and the data merging interface, respectively. It is used to output control signals to regulate the channel selection of the multiplexer and the circuit reconstruction of the operational amplifier feedback network. At the same time, it parses and processes the serial digital code output by the data merging interface, and has a built-in dynamic power consumption management strategy to regulate the opening and closing of the sensor and the corresponding channel.

[0006] The operational amplifier feedback network includes two output lines. One end of each line is connected to the two output ports of the multiplexer, and the other end is connected to the non-inverting input and the inverting input of the operational amplifier, respectively.

[0007] A switch and a capacitor are connected in series on the line between the non-inverting input of the operational amplifier and the multiplexer. The switch is located on one side of the multiplexer, and the node between the switch and the capacitor is also grounded through the switch. The non-inverting input of the operational amplifier is also connected to a capacitor. The other end of the capacitor is divided into two paths: one is grounded through the switch, and the other is connected to the output port of the operational amplifier through the switch.

[0008] A switch two and a capacitor two are connected in series on the line between the inverting input terminal of the operational amplifier and the multiplexer. The switch two is located on one side of the multiplexer, and the node between the switch two and the capacitor two is also grounded through the switch three. The inverting input terminal of the operational amplifier is also connected to the capacitor four. The other end of the capacitor four is divided into two paths, one of which is grounded through the switch seven and the other of which is grounded through the switch six.

[0009] The analog-to-digital converter is a ∑Δ analog-to-digital converter; the single-channel serial digital code output by the data merging interface includes the original sensor acquisition data and a flag bit. The flag bit is concatenated after the original acquisition data and is used to distinguish the sensor type and signal mode corresponding to the current data. The bit width of the flag bit is flexibly configured according to the number of connected sensors.

[0010] The output of the data merging interface is connected to a wireless digital communication module. The wireless digital communication module completes wireless data transmission and reception through an antenna, realizing information interaction between the system and external devices. At the same time, it supports accessing large models deployed in the cloud via the Internet, realizing in-depth analysis functions based on multi-sensor fusion data.

[0011] The dynamic power consumption management strategy of the data processing module is used to match and enable the sensors required for the corresponding functions according to the current working state of the system, and to turn off the sensors and corresponding input channels that are not necessary in the current working state, thereby reducing the system's operating power consumption.

[0012] The multiple sensors include one or more of the following: sound sensor, motion sensor, heart rate sensor, temperature sensor, and GPS positioning sensor. The system is integrated into a smart pet collar for pet status monitoring, health assessment, and emotion analysis.

[0013] The beneficial effects of the present invention after adopting the above structure are:

[0014] 1. Based on sensor fusion technology, this application converts the parallel digital codes acquired by multiple time-division multiplexing into a single-channel serial digital code with sensor type flag bits, realizing the unified encapsulation and transmission of multi-sensor data. It eliminates the need for multi-channel parallel communication, significantly reducing the consumption of communication resources and the computational load of the data processing module.

[0015] 2. Based on time-division multiplexing and reconfigurable technology, this application uses a switchable operational amplifier feedback network to synchronously change the transfer function and amplification type of the preamplifier according to the type of sensor connected. Without modifying the hardware circuit, it can adapt to sensor signals with different output modes such as voltage, current, and resistance, and uniformly convert them into standard voltage signals for subsequent quantization processing. This solves the problems of customized front-end circuit design for different types of sensors, poor system compatibility, and insufficient scalability in the prior art.

[0016] 3. This application can flexibly utilize sensors based on the current working status of the smart collar. On the one hand, time-division multiplexing technology converts multi-channel parallel acquisition into single-channel time-division acquisition, reducing the continuous power consumption caused by multi-channel parallel acquisition and processing. At the same time, the simplified architecture of the single operational amplifier significantly reduces the static power consumption of the analog front end. On the other hand, the system's data processing module has a built-in dynamic power management strategy, which can flexibly enable or disable corresponding sensors and data channels and shut down unnecessary peripheral modules in the current scenario based on the current working status and functional requirements of the device. This effectively solves the core pain points of existing wearable monitoring devices, such as excessive power consumption and insufficient battery life caused by the continuous parallel operation of multiple sensors, and significantly extends the single-use time of the device. Attached Figure Description

[0017] Figure 1 This is a system block diagram of the present invention;

[0018] Figure 2 This is a schematic diagram illustrating the implementation of the present invention.

[0019] Explanation of reference numerals in the attached figures:

[0020] 1. Reconfigurable front end; 2. Data merging interface; 3. Wireless digital communication module; 4. Data processing module; 11. Multiplexer; 12. Switch 1; 13. Switch 2; 14. Switch 3; 15. Switch 4; 110. Switch 5; 111. Switch 6; 112. Switch 7; 113. Switch 8; 114. Operational amplifier; 115. Analog-to-digital converter; 16. Capacitor 1; 17. Capacitor 2; 18. Capacitor 3; 19. Capacitor 4; 30. Operational amplifier feedback network. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] See Figures 1-2As shown, this embodiment is a system for pet collars. It includes a data processing module 4 and a reconfigurable front-end 1, which is connected to multiple sensors. The reconfigurable front-end 1 is used to extract the output data of each sensor. It includes a reconfigurable preamplifier and an analog-to-digital converter 115. The reconfigurable preamplifier includes an operational amplifier 114 and reconfigurable peripheral circuitry. The peripheral circuitry is reconfigured by switching. The peripheral circuitry includes a multiplexer 11 and an operational amplifier feedback network 30. The multiplexer 11 is time-division multiplexed and has multiple channels for connecting multiple sensors. Each channel has two switches connected to the sensors. The conduction of the switches in each channel is controlled by the data processing module 4. Figure 2 For example, there are four sensors, namely A is a sound sensor, B is a motion sensor, C is a heart rate sensor, and D is a temperature sensor. The multiplexer 11 cyclically connects to the sensors according to a preset cycle. When different sensors are connected to the multiplexer 11, the operational amplifier feedback network 30 changes its structure by switching, so that the reconfigurable preamplifier changes its transfer function, which can adapt to the output signals of sensors with different modes such as voltage, current or resistance, and convert them into a unified voltage signal. The analog-to-digital converter 115 collects and quantizes the signal to form a digital code.

[0023] The operational amplifier feedback network 30 includes two output lines. One end of each line is connected to the two output ports of the multiplexer 11, and the other end is connected to the non-inverting and inverting inputs of the operational amplifier 114.

[0024] On the line connecting the non-inverting input terminal of the operational amplifier 114 and the multiplexer 11, there is a switch 12 and a capacitor 16. The switch 12 is located on one side of the multiplexer 11, and the switch 12 and the capacitor 16 are grounded through a switch 4 15.

[0025] The non-inverting input terminal of operational amplifier 114 is also connected to capacitor 318. There are two paths connected to the rear of capacitor 318. One path is grounded through switch 813, and the other path is connected to the output port of operational amplifier 114 through switch 510.

[0026] On the line connecting the inverting input terminal of the operational amplifier 114 and the multiplexer 11, there is a second switch 13 and a second capacitor 17. The second switch 13 is located on one side of the multiplexer 11, and the connection between the second switch 13 and the second capacitor 17 is also grounded through the third switch 14.

[0027] The inverting input terminal of operational amplifier 114 is also connected to capacitor 419. There are two connections on the back of capacitor 419: one is grounded through switch 712, and the other is grounded through switch 611.

[0028] The analog-to-digital converter 115 is a ∑Δ analog-to-digital converter. Its output is a digital code. Each time the multiplexer 11 switches the input channel and the reconfigurable preamplifier output is established, the analog-to-digital converter 115 acquires its output signal and quantizes it to generate a digital code, which is then sent to the data merging interface 2 connected to it. The data merging interface 2 converts the multi-line parallel digital code acquired from multiple lines into a unit serial digital code. In addition to the original input data, the unit serial digital code also contains a flag bit to distinguish the current data type. The flag bit data is concatenated after the original input data, and its bit width can be determined according to the number of connected sensors, which has a certain degree of flexibility.

[0029] A wireless digital communication module 3 is connected to the data merging interface 2. The wireless digital communication module 3 is used for information interaction between the system and the outside world, and completes wireless data transmission and reception through the antenna. In this embodiment, the wireless digital communication module 3 is used for information interaction between the system and the outside world. At the same time, it can access the large model deployed in the cloud through the Internet to realize more complex functions such as pet emotion analysis and health status assessment. The data processing module 4 is used to receive the unit serial digital code output by the data merging interface 2, extract sensor and communication data from it and process it. At the same time, it uses a dynamic power consumption management strategy to schedule peripherals.

[0030] The above data processing includes: filtering and noise reduction of sensor data to improve information recognition rate; storing input data for subsequent statistics and chart drawing; and filtering received data and uploading data containing high-value information to the cloud via wireless digital communication module 3 for deeper analysis.

[0031] The aforementioned scheduling includes: controlling the operating status of sensors in the system based on sensor output data and communication data, and periodically shutting down unused sensors and their corresponding interfaces, thereby reducing system power consumption and increasing battery life.

[0032] When a specific sensor signal is input, the switch will periodically change under the control of clock signals ctrl1 and ctrl2. Taking the input sensor signal as a heart rate sensor C as an example, its specific working process is as follows: The data processing module 4 outputs a gating signal, enabling the multiplexer 11 to connect the heart rate sensor channel. When ctrl1 is high and ctrl2 is low, the reconfigurable front-end 1 operational amplifier operates in the sampling stage. The input voltage signal is converted into charge and stored in capacitors 16 and 17. The amount of stored charge is:

[0033] (1)

[0034] Wherein, the input voltage is Vin, the capacitance of capacitor 16 and capacitor 17 is C1, and the capacitance of capacitor 18 and capacitor 19 is C2. When ctrl1 is low and ctrl2 is high, the reconfigurable front-end 1 op-amp operates in the amplification stage. The charge stored in the sampling stage is transferred to capacitor 18 and capacitor 19, and a certain voltage value is output at the output terminal of operational amplifier 114. The voltage at the output terminal of operational amplifier 114 is:

[0035] (2)

[0036] The voltage output is then fed into a subsequent ΣΔ analog-to-digital converter, where it is converted into digital code. This digital code is then sent to a subsequent data processing module for further data processing.

[0037] Regarding data merging, due to the different transduction mechanisms of various sensors, the output signal modes will have different types. For example, sound sensors typically use piezoelectric materials or variable capacitors to receive sound pressure and convert it into a voltage signal output; PPGs used for pulse detection and accelerometers based on optical interferometry use photodiodes to convert light signals into current signals output; while temperature measurement often uses the temperature characteristics of thermistors or thermoelectric potentials to convert temperature information into resistance or voltage signals output; mid-to-low frequency analog front-ends are often implemented based on operational amplifiers and their feedback networks. Assuming the input signal of the front-end is... To adapt to the analog-to-digital converter after the front-end amplifier, it is desired to convert the output signal into voltage. When the feedback coefficient of the feedback network is... At that time, the front-end output voltage The relationship between the input signal and the input signal can be expressed as:

[0038] (3)

[0039] in, For the amplifier's forward gain; from equation (3), it can be seen that if we want the front end to be in different modes... Both can output voltage signals, as long as you ensure Dimensions and As long as they are the same, and The dimensions are related to the structure of the feedback network; for example, when the feedback is voltage-current negative feedback, The dimension of is conductance, and the amplifier's amplification type is transimpedance, i.e., input current and output voltage; when the feedback is voltage-to-voltage negative feedback, The dimension is 1, and the amplifier's amplification type is voltage, i.e., input voltage and output voltage. For the input mode being resistive, one end of the resistor can be connected to a constant voltage, and the other end to the virtual ground of the op-amp input. Through the resistor's VI characteristic, it can be made to output a current related to its resistance value. In this way, a transimpedance amplifier front-end can be used to detect resistive mode signals. Based on the above analysis, it is possible to change the structure of the feedback network by switching the switch, thereby changing... The system amplifies different modal signals such as voltage, current, and resistance using only one operational amplifier. Based on this, and utilizing multiplexing technology, different sensors can be connected to the system at different times, while simultaneously changing the amplification type at the front end. This enables time-division multiplexing, reducing the number of front-end channels from the original total number of sensors in the system to one. Assuming the number of input channels is... After time-division multiplexing, each pair of front-end nodes... After one sampling of the input signal, the output signal can be expressed as:

[0040]

[0041] (4)

[0042] In formula (4) Let be the unit impulse function in the discrete domain. For the first Sensor input signals from each channel, For reconfigurable front-end amplifiers receiving the first The gain of each channel; as shown in equation (4), by sequentially switching multiple sensor channels and simultaneously reconstructing the front-end amplifier to match the mode of the input signal of that channel, if the sampling period of the front end is Then each time Afterwards, the front end completes a recording of the signals from all input channels. The output data, after being extracted by the front end, is uniformly converted into voltage mode signals in the discrete time domain. Considering that the front end will start recording again from the first channel after completing a recording of the signals from all input channels, and repeat this cycle, equation (4) can be modified as follows:

[0043] (5)

[0044] At this point, a sampling rate can be used. The analog-to-digital converter completes the quantization and encoding of the voltage signal, and then the parallel-to-serial conversion technology is used to complete the data extraction of all sensors using a single serial data channel;

[0045] In terms of data processing, the smart collar offers diverse functions, but not every function requires the use of all sensors. To reduce energy waste from sensors in different operating modes, the processing module can flexibly utilize sensors based on the collar's current operating status. For sensors not needed in the current operating state, the processing module will disable their enable switches, thereby saving system power consumption and improving battery life. For example, when monitoring only the pet's health, the processing module will enable the temperature and heart rate sensors while disabling other sensors. Only when the owner wants to use specific collar functions, such as obtaining information about the pet's location and movement, will the processing module enable the corresponding GPS module and accelerometer. Furthermore, the data processing module will comprehensively process the data collected by each sensor, combining information from different dimensions of the current environment reflected by each sensor to improve system accuracy and the number of supported functions. For example, when analyzing the emotions contained in the pet's vocalizations, the processing module will combine information such as the pet's current heart rate, body temperature, and movement obtained from the heart rate, temperature, and accelerometer sensors to help determine the pet's emotions.

[0046] The installation, connection, or setting methods of the components not detailed above are all common mechanical methods, and the specific structure, model, and coefficient indicators of all their components are their own technologies. As long as they can achieve their beneficial effects, they can be implemented, so they will not be elaborated further.

[0047] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of the invention should be included within the protection scope of the invention. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A sensor data fusion processing system, characterized in that: It includes a reconfigurable front end (1), a data merging interface (2), and a data processing module (4); the reconfigurable front end (1) is connected to multiple different types of sensors to extract the output data of each sensor; the reconfigurable front end (1) includes a reconfigurable preamplifier and an analog-to-digital converter (115); the reconfigurable preamplifier includes a single operational amplifier (114) and reconfigurable peripheral circuits; the peripheral circuits are reconfigured by switching; the peripheral circuits include a time-division multiplexer (11) and an operational amplifier feedback network (30); the multiplexer (11) is provided with multiple input channels, each input channel is connected to a corresponding sensor, and is used to cyclically select the output signals of different sensors according to a preset period; the operational amplifier feedback network (30) is connected to the output terminal of the multiplexer (11), the input terminal and the output terminal of the operational amplifier (114), and is used to change its own circuit structure by switching when the multiplexer (11) is connected to different sensors, so as to change the transmission of the reconfigurable preamplifier. The function adapts to the output signals of sensors with different modes of voltage, current, and resistance, and converts them into voltage signals for output to the analog-to-digital converter (115). The input terminal of the analog-to-digital converter (115) is connected to the output terminal of the operational amplifier (114) and is used to collect the output signal and quantize it to generate digital code after the multiplexer (11) completes one input channel switching and the reconfigurable preamplifier output is established. The input terminal of the data merging interface (2) is connected to the output terminal of the analog-to-digital converter (115) and is used to convert the multi-channel parallel digital code generated by time-division acquisition into a single-channel serial digital code. The data processing module (4) is electrically connected to the multiplexer (11), the switches in the operational amplifier feedback network (30), and the data merging interface (2) respectively. It is used to output control signals to regulate the channel selection of the multiplexer (11) and the circuit reconstruction of the operational amplifier feedback network (30). At the same time, it parses and processes the serial digital code output by the data merging interface (2) and has a built-in dynamic power consumption management strategy to regulate the opening and closing of the sensor and the corresponding channel.

2. The sensor data fusion processing system according to claim 1, characterized in that: The operational amplifier feedback network (30) includes two output lines. One end of each line is connected to the two output ports of the multiplexer (11), and the other end is connected to the non-inverting input and the inverting input of the operational amplifier (114).

3. The sensor data fusion processing system according to claim 2, characterized in that: On the line connecting the non-inverting input terminal of the operational amplifier (114) and the multiplexer (11), a switch (12) and a capacitor (16) are connected in series. The switch (12) is located on one side of the multiplexer (11), and the node between the switch (12) and the capacitor (16) is also grounded through the switch (4) (15). The non-inverting input terminal of the operational amplifier (114) is also connected to the capacitor (3) (18). The other end of the capacitor (3) (18) is divided into two paths. One path is grounded through the switch (8) (113), and the other path is connected to the output port of the operational amplifier (114) through the switch (5) (110).

4. The sensor data fusion processing system according to claim 2, characterized in that: On the line connecting the inverting input of the operational amplifier (114) and the multiplexer (11), a second switch (13) and a second capacitor (17) are connected in series. The second switch (13) is located on one side of the multiplexer (11), and the node between the second switch (13) and the second capacitor (17) is also grounded through the third switch (14). The inverting input of the operational amplifier (114) is also connected to the fourth capacitor (19). The other end of the fourth capacitor (19) is divided into two paths, one of which is grounded through the seventh switch (112), and the other is grounded through the sixth switch (111).

5. The sensor data fusion processing system according to claim 1, characterized in that: The analog-to-digital converter (115) is a ∑Δ analog-to-digital converter; the single-channel serial digital code output by the data merging interface (2) includes the original sensor acquisition data and a flag bit. The flag bit is concatenated after the original acquisition data and is used to distinguish the sensor type and signal mode corresponding to the current data. The bit width of the flag bit is flexibly configured according to the number of connected sensors.

6. The sensor data fusion processing system according to claim 1, characterized in that: The output end of the data merging interface (2) is connected to a wireless digital communication module (3). The wireless digital communication module (3) completes wireless data transmission and reception through an antenna, realizes information interaction between the system and external devices, and supports access to large models deployed in the cloud via the Internet to realize in-depth analysis based on multi-sensor fusion data.

7. The sensor data fusion processing system according to claim 1, characterized in that: The dynamic power consumption management strategy of the data processing module (4) is used to match the sensors required to enable the corresponding functions according to the current working state of the system, and to turn off the unnecessary sensors and corresponding input channels in the current working state, thereby reducing the system's operating power consumption.

8. The sensor data fusion processing system according to claim 1, characterized in that: The multiple sensors include one or more of the following: sound sensor, motion sensor, heart rate sensor, temperature sensor, and GPS positioning sensor. The system is integrated into a smart pet collar for pet status monitoring, health assessment, and emotion analysis.

Images