A passive sensor data transmission optimization method

Abstract

The application discloses a passive sensor data transmission optimization method, and belongs to the technical field of data transmission optimization, in particular to a passive sensor data transmission optimization method.The application provides a passive sensor data transmission optimization method with high data transmission efficiency, good stability and energy saving nature.The application comprises the following steps: step 1, monitoring the environmental parameters of the passive sensor, and taking the environmental parameters as context information;step 2, performing dynamic redundancy threshold calculation according to the context information;step 3, performing compressed transmission on the data obtained in step 2; and step 4, after the data is compressed and transmitted, obtaining feedback information of transmission quality and feedback information of energy utilization efficiency, and returning the feedback information to step 2 to form a closed-loop control.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data transmission optimization, and particularly relates to a passive sensor data transmission optimization method. BACKGROUND

[0002] With the continuous development of Internet of Things technology, passive sensor networks are increasingly widely used in environmental monitoring, health monitoring, industrial control and other fields. However, traditional passive sensor data transmission technology has many challenges in low-power transmission, redundant data processing, stable transmission under low signal-to-noise ratio, etc. First, most existing compression algorithms use static strategies and cannot dynamically adjust according to changes in the environment and sensor state, resulting in low data transmission efficiency. Second, passive sensors rely on environmental energy such as microwaves as a power source, and how to optimize energy consumption to extend the service life of the device becomes an important problem. In addition, there is a lot of redundant information in sensor data, and existing technologies cannot effectively identify and filter redundant data, causing waste of bandwidth and computing resources.

[0003] Existing related implementation schemes mainly focus on topology optimization and energy efficiency improvement of wireless sensor networks.

[0004] Existing document 1 "Optimization control method and device for wireless sensor network topology" (inventors Zheng Yin et al., patent number CN117412313A) realizes topology optimization and energy efficiency improvement of wireless sensor network through sub-gradient algorithm and block coordinate descent algorithm. This method first selects an initial cluster head node and constructs a transmission power optimization objective function, and uses Lagrange dual problem to solve the maximum transmission power of the node; by iteratively updating the Lagrange multiplier and power parameter, the optimal transmission power and node related factor are determined when the multiplier converges; finally, the cluster head structure is optimized in combination with the Euclidean distance, so that each node in the network realizes efficient data transmission with optimal power and association configuration, significantly reducing the overall energy consumption of the network.

[0005] Existing document 2 "Wireless sensor network data transmission optimization method" (inventor: Li Jianpo, authorized publication number: CN108966354B) improves network energy efficiency and transmission efficiency through a two-stage strategy. In the intra-cluster data transmission stage, the method dynamically allocates transmission priority according to the node's residual energy and data buffer state, optimizes time slot scheduling to reduce data backlog; in the inter-cluster transmission stage, a comprehensive evaluation model is constructed by combining parameters such as relay node energy consumption and transmission distance, and the optimal path is intelligently selected to complete multi-hop forwarding. Through node state sensing and dynamic path planning, this method significantly reduces the overall energy consumption of the network, reduces data transmission delay, and enhances the stability and reliability of the network in complex environments, suitable for energy-limited Internet of Things scenarios.

[0006] Although the prior art has achieved certain results in topology optimization and energy efficiency improvement in the application of passive sensor networks, there are still many deficiencies, especially in dynamic redundant data processing, stable transmission in low signal-to-noise ratio environment and intelligent compression algorithm. SUMMARY

[0007] The present application is aimed at the above problems, and provides a passive sensor data transmission optimization method with high data transmission efficiency, stability and energy saving.

[0008] To achieve the above-mentioned purpose, the present application adopts the following technical solutions, and the present application comprises the following steps:

[0009] Step 1: monitoring the environmental parameters of the passive sensor, the environmental parameters as context information;

[0010] Step 2: calculating the dynamic redundant threshold according to the context information;

[0011] Step 3: compressing and transmitting the data obtained in step 2;

[0012] Step 4: after the data is compressed and transmitted, obtaining the feedback information of transmission quality and the feedback information of energy utilization efficiency, returning the feedback information to step 2 to form a closed loop control.

[0013] As a preferred scheme, the environmental parameters in step 1 of the present application include signal-to-noise ratio (SSNR), multipath fading coefficient, environmental temperature and environmental humidity; the signal-to-noise ratio is the power ratio between the received signal and the environmental noise when the passive sensor sends to the base station or the receiving end; the multipath fading coefficient is the fading degree of the received signal caused by the multipath propagation characteristics in the channel where the passive sensor is located.

[0014] The multipath fading coefficient can be obtained by the receiving circuit of the passive sensor.

[0015] As another preferred scheme, the passive sensor of the present application converts the 2.4GHz microwave energy transmitted by the base station into direct current through a microwave rectifier circuit (it is a conventional technology to convert radio frequency energy into direct current by using a microwave rectifier circuit, and the present application improves the data transmission and redundancy optimization of the passive sensor), and the relationship between the conversion efficiency η and the SSNR is:

[0016]

[0017] wherein, is the received power; P tx is the transmitted power; k is a constant coefficient, which comprehensively reflects the conversion efficiency of the rectifier circuit, antenna coupling loss and device characteristics.

[0018] The environmental parameter vector is:

[0019] X = [SSNR, T env ,H env ,vibration freq_peak ,vibration band_energy ]

[0020] T env is the ambient temperature parameter, H env is the ambient humidity parameter, vibration freq_peak is the vibration peak frequency, vibration band_energy is the vibration bandwidth energy; the ambient parameter vector is used as context information.

[0021] vibration freq_peak is the main vibration frequency detected by the passive vibration sensor on the target to be measured, which can be monitored by the MEMS accelerometer sensing element in the sensor.

[0022] vibration band_energy is the energy accumulation of the target component vibration signal in a certain frequency band range by the passive vibration sensor, which can be obtained by accumulating the energy of the frequency band by the same set of MEMS sensing elements as the vibration peak frequency through digital signal processing algorithm.

[0023] As another preferred solution, the calibration and correction process of k in the application is as follows:

[0024] Hardware actual measurement calibration: in a complete transmitting-receiving system, multiple sets of input / output power and SSNR data are measured, and the optimal k is obtained by formula

[0025]

[0026] least squares or curve fitting method;

[0027] Ambient and device correction: repeat the calibration under different temperature and humidity, device aging degree and antenna matching conditions, and obtain the corresponding correction coefficient or correction function to dynamically calibrate k.

[0028] Due to different hardware designs and environmental factors, hardware measurement and environmental correction are required to determine the efficiency.

[0029] As another preferred solution, step 2 of the application includes the relationship between environmental parameters and redundancy threshold, the hierarchical encoding and redundancy calculation, and the dynamic adjustment of redundancy threshold; the setting of redundancy threshold is determined by environmental parameters, and the adjustment strategy of redundancy threshold is affected by redundancy calculation; the redundancy threshold is adjusted according to the deviation between redundancy and current redundancy threshold through the calculation of data redundancy; the updated redundancy threshold guides the data compression and transmission strategy.

[0030] The environmental parameters (such as signal-to-noise ratio, temperature and humidity, etc.) are used to set the redundancy threshold T0. These parameters affect the initial threshold setting. Through the calculation of data redundancy, the system adjusts the redundancy threshold according to the deviation of redundancy and the current threshold, realizing intelligent adjustment. The updated redundancy threshold guides the data compression and transmission strategy, ensuring the transmission efficiency and stability in different environments. This closed loop iterates continuously, environmental changes affect the redundancy threshold, redundancy calculation feedback adjusts the threshold, and finally optimizes the data transmission efficiency.

[0031] As another preferred solution, the relationship between the environmental parameters and the redundancy threshold of the present application is as follows:

[0032] The initial redundancy threshold is set as T0, and the relationship between the dynamic redundancy threshold T and the environmental parameter vector X is as follows:

[0033] T = T0 x f(X)

[0034]

[0035] Where f(X) is a mapping function, representing how the environmental parameter vector X affects the redundancy threshold T; α SSNR , α T and α H control the weight of SSNR, temperature and humidity on the redundancy threshold, respectively. ref , T ref and H ref are reference values.

[0036] As another preferred solution, the SSNR ref of the present application is selected as the signal-to-noise ratio value measured in an environment without signal interference or small background noise. For example, the signal-to-noise ratio measured in an ideal experimental environment can be selected, or a standard value determined when designing the sensor.

[0037] T ref : is the temperature value measured under standard experimental environment conditions, which is the ideal temperature value set when testing the sensor. For example, 25℃ (room temperature) or other preset values.

[0038] H ref : Select the standard humidity value measured in the laboratory conditions. For example, relative humidity 50%, or the standard humidity in the test environment.

[0039] As another preferred solution, the hierarchical encoding and redundancy calculation part of the present application is as follows:

[0040] The detection data of the passive sensor is D = [d1, d2, …, d n ], which is divided into multiple levels according to the redundancy degree, and each level of data redundancy has different threshold; the redundancy R of the kth level isk For:

[0041]

[0042] wherein n k represents the number of data points of the kth level, d i is the original data point collected, is the prediction or expectation value of the current level, the redundancy R k represents the redundancy degree of the data of the level.

[0043] D = [d1, d2, …, d n ], representing a series of data detected by the passive sensor during operation, such as temperature, humidity, and vibration amplitude.

[0044] The division into multiple levels according to the redundancy degree means that the data is divided into several levels according to the size of the prediction residual error, and more redundancy is provided for the data level with larger error, and higher compression is provided for the level with smaller error.

[0045] As another preferred solution, the dynamic adjustment of the redundancy threshold value part of the application includes:

[0046] The adjustment rule of the redundancy threshold value T k is represented by the following formula:

[0047]

[0048] wherein T k base is the basic redundancy threshold value of the level; a k is the adjustment coefficient of the level, controlling the sensitivity of the redundancy threshold value to environmental changes; T ref and H ref are standard values set; a T and a H are adjustment coefficients, controlling the degree of influence of temperature and humidity changes on the threshold value.

[0049] The system dynamically adjusts the redundancy threshold value according to environmental conditions (such as signal-to-noise ratio, temperature and humidity, etc.). For high-redundancy layer data (such as important temperature and humidity change data), the redundancy threshold value is low (such as T k _high base = 0.1), bas e ensures that these key data can maintain a high transmission quality and avoid losing important information. For low-redundancy layer data (such as non-critical data), the redundancy threshold value will be higher (such as T k _low base = 0.5), allowing more compression of data during transmission to reduce bandwidth occupation and energy consumption. The a kIt can be set to 0.1 (low sensitivity), and α can be set for high redundancy layers. T =0.8, α H =0.8 (highly affected by environmental parameters). α of the low redundancy layer k It can be set to 0.3 (high sensitivity), and α can be set for low redundancy layers. T =0.2, α H =0.2 (Environmental parameters have little impact).

[0050] As another preferred embodiment, when SSNR < 20 dB, T k base Increase by 20% to 30%; when SSNR > 25dB, T k base A decrease of 10% to 15%.

[0051] As another preferred embodiment, the R described in this invention k >T k (High redundancy layer): Differential coding + Huffman static dictionary compression is used, with a compression ratio of 5:1;

[0052] R k ≤T k (Low redundancy layer): Directly discard adjacent duplicate data points and only retain data with a change of more than ±0.5%.

[0053] As another preferred embodiment, in step 3 of the present invention, the data compression and transmission of the passive temperature sensor is as follows:

[0054] The original temperature data sequence is D = [d1, d2, ..., d n The differential encoding process is as follows:

[0055] Δd i =d i -d i-1 i = 2, 3, ..., n

[0056] Where, d i For the i-th data point, d i-1 For the previous data point, Δd i The difference between data points is used to transform the original data sequence into a difference sequence in order to reduce the amount of data transmitted.

[0057] On the basis of differential coding, further compression is performed using a Huffman static dictionary. The frequency of the data points after differential coding is counted, and the data points are converted to binary codes through frequency mapping to complete data compression. The compressed data is transmitted through the communication link between the passive temperature sensor and the receiving end. In this process, important data will maintain a lower compression ratio to ensure accurate transmission, while non-critical data will use a higher compression ratio to improve transmission efficiency. For data in the low redundancy layer, the overall energy efficiency of the system is further optimized by adjusting the transmission frequency, bandwidth and power during data transmission.

[0058] High redundancy layer (important data): This layer contains data that is critical to the application, such as the amplitude of temperature and humidity changes, or other key environmental parameters. These data points are assigned a lower redundancy threshold to ensure high-fidelity transmission. Since these data are crucial to the function of the system, they need to be transmitted at a lower compression ratio to ensure data integrity and accuracy. Even in cases where environmental conditions change significantly, the system will ensure that these data are transmitted stably, thereby ensuring the performance and reliability of the system.

[0059] Low redundancy layer (non-critical data): This layer contains data that is less important to the application and can tolerate larger errors or have less impact on the system. These data points are assigned a higher redundancy threshold, allowing for higher compression ratios to save bandwidth and energy. By using stronger compression algorithms, the system can efficiently reduce the amount of non-critical data transmitted, thereby optimizing resource usage and reducing unnecessary data transmission.

[0060] As another preferred solution, the data compression and transmission of the passive vibration sensor in step 3 of the present application is as follows:

[0061] Time-domain vibration data x n The formula for DCT transformation is:

[0062]

[0063] where X k is the transformed frequency domain data; the low-frequency coefficients (i.e., the part with smaller k) usually contain more signal energy, reflecting the main characteristics of the data. Therefore, the probability distribution of these frequency domain coefficients is constructed by counting their frequency of occurrence.

[0064] P(X k ) is the probability of the kth low-frequency coefficient X k , representing the frequency of the kth coefficient appearing in all frequency domain coefficients;

[0065] The P(X k ) distribution satisfies:

[0066]

[0067] Arithmetic coding divides the entire interval [0,1) into sub-intervals, each of which represents a coefficient. Initialization: low = 0, high = 1;

[0068] For each coefficient X k , the corresponding interval is [low k , high k ], and the interval is updated:

[0069]

[0070] Continue this operation until all coefficients are encoded. Finally, output the midpoint of the interval as the encoding result:

[0071] Encoded output = Binary representation of the midpoint of [low, high].

[0072] As another preferred solution, the data layer compression and transmission of the passive vibration sensor according to the present application:

[0073] If SSNR > 25 dB, the first eight DCT coefficients are retained:

[0074] X k = [X0, X1, …, X7] (SSNR > 25 dB)

[0075] If SSNR < 20 dB, the first twelve DCT coefficients are retained:

[0076] X k = [X0, X1, …, X 11 ] (SSNR < 20 dB).

[0077] Passive sensors can be classified according to their types and data acquisition characteristics. For different types of sensors, the data compression and transmission process is different. The following is a description of the process for different types of sensors:

[0078] Passive temperature sensor compression and transmission process:

[0079] Passive temperature sensors are usually used to monitor relatively slow-changing environmental parameters, such as temperature. The data collected by this type of sensor changes less and has a low fluctuation amplitude, so the process is relatively simple. Mainly using differential encoding and Huffman static dictionary compression strategy.

[0080] Differential encoding: By recording the difference between adjacent data points instead of transmitting the raw data values directly, the amount of data transmitted is reduced.

[0081] Huffman encoding: By assigning shorter codes to symbols with higher frequencies, the data volume is further reduced, optimizing data transmission efficiency.

[0082] Passive humidity sensor compression transmission process:

[0083] The data collected by the passive humidity sensor usually changes slowly, similar to the temperature sensor. The processing of humidity data can also use differential encoding and static compression strategy. The specific processing steps are as follows:

[0084] Differential encoding: Similar to the temperature sensor, the humidity data changes less, and differential encoding is used to reduce the data volume.

[0085] Huffman encoding: Further compress the differential encoded data, and optimize the data transmission through Huffman static dictionary.

[0086] Passive vibration sensor compression transmission process:

[0087] Passive vibration sensors are usually used to monitor relatively fast-changing environmental parameters, such as vibration. Due to the frequent changes and high-frequency information contained in vibration data, the processing process is relatively complex. Discrete cosine transform (DCT) and arithmetic coding techniques are mainly used.

[0088] DCT (Discrete Cosine Transform): Convert time-domain data to frequency-domain data, remove low-energy high-frequency parts, and reduce data volume.

[0089] Arithmetic coding: Further compress the frequency-domain data, convert it to compact binary code by estimating the probability distribution of each coefficient, and reduce the transmission volume.

[0090] As another preferred solution, the way to obtain the feedback information of transmission quality in step 4 of the present application is to obtain the feedback information of transmission quality through the base station side or passive sensor side detection module.

[0091] As another preferred solution, the feedback information of transmission quality includes bit error rate BER, packet loss rate PLR and link quality indicator LQI.

[0092] Secondly, the way to obtain the feedback information of energy utilization efficiency in step 4 of the present application is to obtain the feedback information of energy utilization efficiency through the energy detection module.

[0093] In addition, the energy detection module of the present application is used to monitor the output power of the microwave rectifier circuit.

[0094] Advantages of the present application.

[0095] The present application can flexibly adjust the data compression strategy according to the real-time environmental changes and the feedback of the passive sensor state, thereby effectively improving the data transmission efficiency and reducing the waste of bandwidth and computing resources caused by redundant data, and ensuring the efficiency, stability and energy saving of passive sensor data transmission in complex environments.

[0096] The present application can maintain the stability and reliability of data transmission in a low signal-to-noise ratio environment through accurate redundant data identification and filtering mechanism, avoiding the bandwidth waste and network instability caused by improper processing of redundant data in traditional technology. That is, assuming that the data of the sensor is D=[d1, d2, …, d n ], it is divided into multiple levels according to the degree of redundancy, and each level of data redundancy has different threshold. The redundancy R k of the kth level is defined as:

[0097]

[0098] Where, n k represents the number of data points in the kth level, d i is the original data point collected, is the prediction or expectation value of the current level. The redundancy R k represents the degree of data redundancy of this level.

[0099] The adjustment rule of the redundancy threshold T k is represented by the following formula:

[0100]

[0101] Where, T k basek is the basic redundancy threshold of this level; α k is the adjustment coefficient of this level, which controls the sensitivity of the redundancy threshold to environmental changes; T ref and H ref are standard values; α T and α H are adjustment coefficients, which control the influence degree of temperature and humidity changes on the threshold.

[0102] The present application optimizes the data compression process through hierarchical encoding and data compression, further improves the efficiency of data transmission, and overcomes the problems of insufficient stability and lack of adaptability of compression strategy in the prior art.

[0103] According to the sensor type and environmental changes, the application adopts hierarchical coding technology to process data according to redundancy. Different compression strategies are applied to each data layer according to its importance and redundancy degree, thereby optimizing the use of transmission bandwidth while ensuring data integrity and accuracy. That is, passive temperature sensors are usually used to monitor relatively slow-changing environmental parameters, such as temperature. The collected data of such sensors changes less and has lower fluctuation amplitude, so the processing process is relatively simple. Mainly adopts differential encoding and Huffman static dictionary compression strategy. Passive vibration sensor compression transmission processing: Passive vibration sensors are usually used to monitor relatively fast-changing environmental parameters, such as vibration. Since vibration data changes frequently and contains more high-frequency information, the processing process is relatively complex. Mainly adopts discrete cosine transform (DCT) and arithmetic coding technology. In hierarchical coding, the redundancy threshold is dynamically adjusted according to the real-time environmental parameter changes. For each level, the adjustment rule of the redundancy threshold T k

[0104]

[0105] Wherein, T k basek is the basic redundancy threshold of the level, and α k is the adjustment coefficient of the level, which controls the sensitivity of the redundancy threshold to environmental changes. T ref and H ref are standard values set by the system, which are usually measured under ideal conditions without special environmental changes. α T and α H are adjustment coefficients that control the influence of temperature and humidity changes on the threshold.

[0106] In the data compression of the vibration sensor, the application combines the SSNR dynamic adjustment of the number of retained DCT coefficients, and determines which frequency domain coefficients to retain according to the signal-to-noise ratio. In this way, a small amount of low-frequency components are retained under high SSNR conditions, and more frequency domain coefficients are retained under low SSNR conditions (traditional DCT does not consider the influence of environmental changes on compression efficiency).

[0107] The application can effectively identify and filter redundant data, realize dynamic identification and filtering of redundant data, and avoid wasting bandwidth and computing resources. That is, hierarchical redundancy quantization (step 2 core):

[0108] The sensor data is divided into multiple levels according to the predicted residual, and the redundancy of each level is calculated:

[0109]

[0110] Wherein Sliding window mean prediction is adopted. When R​k When the current dynamic threshold Tk is lower, the layer data is determined as redundant data.

[0111] Environment adaptive threshold adjustment:

[0112] According to real-time signal-to-noise ratio (SSNR) to dynamically correct the threshold:

[0113]

[0114] When SSNR < 20 dB, T k base Up 20% to 30%, relax the redundancy determination standard to ensure transmission stability; when SSNR > 25 dB, T k base Down 10% to 15%, implement more stringent redundancy filtering.

[0115] Hierarchical compression execution:

[0116] R k > T k (high redundancy layer): differential encoding + Huffman static dictionary compression is adopted, and the compression ratio is 5:1;

[0117] R k ≤ T k (low redundancy layer): adjacent repeated data points are directly discarded, and only data with a change of more than ±0.5% is retained. BRIEF DESCRIPTION OF DRAWINGS

[0118] The present application will be further described below in conjunction with the drawings and specific embodiments. The scope of protection of the present application is not limited to the following content.

[0119] Figure 1 is a flow chart of the data transmission optimization system of the present application. DETAILED DESCRIPTION

[0120] As shown in the figure, in a passive sensor network, data collection is the starting point of the entire system. Each passive sensor is responsible for real-time monitoring of specific environmental parameters, including temperature, humidity, and vibration parameters. In terms of power supply, the passive sensor works by receiving microwave energy from the base station and feeds back the collected data to the base station in the form of wireless signals. Since the passive sensor completely relies on external microwave power supply, the stability of energy supply is highly coupled with data transmission efficiency, and the influence of energy dependence on data integrity needs to be reduced through dynamic optimization.

[0121] In order to reasonably optimize data acquisition, the application adopts context-aware modeling. In this stage, first, the base station monitors the environmental parameters in real time. The key parameters include signal-to-noise ratio (SSNR), multipath fading coefficient, environmental temperature and humidity. Changes in the environment will directly affect the data quality of the sensor, and the signal-to-noise ratio (SSNR) is particularly important. Low signal-to-noise ratio environments often lead to data transmission errors.

[0122] The passive sensor converts the 2.4GHz microwave energy transmitted by the base station into direct current through a microwave rectifier circuit, and the conversion efficiency η is related to the SSNR:

[0123]

[0124] wherein, is the received power, P tx is the transmitted power. The formula reveals the inherent coupling between energy supply and data transmission quality, which needs to be decoupled through a dynamic threshold compensation mechanism.

[0125] Context-aware real-time feedback of the current environmental state dynamically adjusts the processing strategy of the passive sensor data. Assuming that X represents a set of key parameter vectors in the environment:

[0126] X = [SSNR, T env , H env , vibration freq_peak , vibration band_energy ]

[0127] These environmental parameters, i.e. environmental temperature parameter, environmental relative humidity parameter, vibration peak frequency, vibration bandwidth energy, will be input into the system model as context information to help judge the redundancy, transmission priority and compression strategy of the data. Based on the passive sensor ID, the system automatically identifies the type of passive sensor and loads the feature model suitable for different passive sensors. For temperature sensors, the temperature fluctuation model is loaded; for vibration sensors, the vibration frequency change model is loaded.

[0128] In the dynamic redundancy threshold calculation step, context-aware dynamic threshold layered coding (CDT-LC) is used as the core of the data redundancy awareness mechanism, which aims to dynamically adjust the data redundancy threshold by real-time feedback of environmental changes and sensor state, thereby realizing efficient data compression and transmission.

[0129] First, the system acquires environmental parameters related to data transmission through the base station. Key parameters include signal-to-noise ratio (SSNR), multipath fading coefficient, ambient temperature, and humidity. These environmental parameters directly affect the validity and redundancy of the data.

[0130] An initial redundancy threshold is set to T0, and this threshold is adjusted based on real-time environmental feedback. Assume the relationship between the dynamic redundancy threshold T and the environmental parameter X is as follows:

[0131] T = T0 × f(X)

[0132] Here, f(X) is a mapping function that represents how environmental parameters affect the redundancy threshold. For example, when the SSNR is low, the redundancy threshold can be appropriately relaxed to ensure stable data transmission; while changes in temperature and humidity will affect the fluctuation range of the data, thus affecting the setting of the redundancy threshold.

[0133] One of the core features of CDT-LC is layered coding, which allows the system to divide data into different layers based on its importance and redundancy. At each layer, the system compresses data using different redundancy thresholds, thus prioritizing the transmission of important data.

[0134] Assume the sensor data is D = [d1, d2, ..., d n The data is divided into multiple levels based on its redundancy level, with each level having a different threshold for redundancy. The redundancy level R of the k-th level is... k Defined as:

[0135]

[0136] Where, n k d represents the number of data points at level k. i The raw data points collected, This represents the predicted or expected value at the current level. Redundancy R k This indicates the degree of data redundancy at this level.

[0137] In hierarchical coding, the redundancy threshold is dynamically adjusted based on real-time changes in environmental parameters. For each level, the redundancy threshold T... k The adjustment rules can be expressed by the following formula:

[0138]

[0139] Among them, T k base α is the basic redundancy threshold for this level. k This is the adjustment coefficient for this level, controlling the sensitivity of the redundancy threshold to environmental changes. T ref and H refThese are standard values ​​set by the system. Normally, these values ​​are measured under ideal conditions without special environmental changes. α T and α H It is an adjustment coefficient that controls the degree to which changes in temperature and humidity affect the threshold.

[0140] When the SSNR is low, the redundancy threshold is appropriately increased to ensure effective data compression and transmission even under poor signal conditions. Changes in temperature and humidity also affect the dynamic adjustment of the redundancy threshold, enabling the system to be flexibly optimized under different environmental conditions.

[0141] After calculating data redundancy, the next task is to compress and transmit the effective data. Different compression algorithms are selected based on sensor type, data characteristics, and environmental conditions to optimize transmission efficiency, save bandwidth, reduce energy consumption, and ensure the data maintains its integrity and stability during transmission.

[0142] Temperature data collected by sensors (low-frequency, slowly varying data) typically changes slowly and fluctuates little. Therefore, for temperature data, a combination of differential coding and a Huffman static dictionary is used for compression. The core of differential coding is to record the differences between adjacent data points, rather than directly transmitting the raw data values. Because temperature data changes slowly, the differences between adjacent data points are usually small, which can significantly reduce the amount of data transmitted.

[0143] Assume the original temperature data sequence is D = [d1, d2, ..., d n The differential encoding process is as follows:

[0144] Δd i =d i -d i-1 i = 2, 3, ..., n

[0145] Where, d i For the i-th data point, d i-1 For the previous data point, Δd i This represents the difference between data points. In this way, the original data sequence is transformed into a difference sequence. Differential encoding reduces data fluctuations and facilitates compression.

[0146] Building upon differential coding, this invention employs a Huffman static dictionary for further compression. Huffman coding is a lossless compression algorithm that achieves data compression by assigning shorter codes to higher-frequency symbols and longer codes to lower-frequency symbols. Huffman coding can allocate shorter binary codes to frequently occurring data based on signal frequency, thereby reducing redundancy and bandwidth requirements.

[0147] The data collected by vibration sensors (high frequency transients) changes rapidly and usually contains a large amount of high frequency information. Because of the large amount of low-energy high frequency parts in the vibration signal, the data volume cannot be effectively reduced by using traditional compression methods. Therefore, the present application uses the discrete cosine transform (DCT) technique for frequency domain conversion, and combines an adaptive coefficient selection method to optimize the data compression effect.

[0148] DCT is a commonly used signal compression technique, especially suitable for processing periodically changing data. By converting time domain data into frequency domain data, DCT can extract the main components of the signal and discard the low-energy high frequency parts, thereby reducing the data volume. For time domain vibration data x n , the formula for DCT transformation is:

[0149]

[0150] where x n is the original vibration data point, and X k is the transformed frequency domain data. By performing DCT transformation on the vibration data, the data can be converted from the time domain to the frequency domain to obtain the frequency components X k .

[0151] After obtaining the frequency domain data X k , the system dynamically adjusts the number of retained DCT coefficients according to the current SSNR (signal-to-noise ratio). Generally, the data in the low frequency part is more representative of the main characteristics of the signal, while the high frequency part contains less energy and contributes less to the overall characteristics of the signal.

[0152] A threshold is set to determine how many frequency domain coefficients to retain according to the current SSNR. If the SSNR is high, the first 8 low frequency coefficients are retained:

[0153] X k = [X0, X1, …, X7] (SSNR > 25 dB)

[0154] If the SSNR is low, the first 12 coefficients are retained:

[0155] X k = [X0, X1, …, X 11 ] (SSNR < 20 dB)

[0156] In this way, the system can dynamically adjust the compression ratio while ensuring data transmission quality, improving data compression efficiency.

[0157] For the retained low frequency DCT coefficients, the present application uses arithmetic coding for further compression. First, estimate the probability distribution of the occurrence of each coefficient. Let P(X k) represents the probability of the kth low-frequency coefficient X k and the probability distribution satisfies:

[0158]

[0159] Next, arithmetic coding divides the entire interval [0,1) into sub-intervals, each representing a symbol (coefficient). Initialization is low = 0, high = 1.

[0160] For each symbol X k (corresponding interval [low k ,high k ]) update the interval:

[0161]

[0162] Continue this operation until all symbols are encoded. Finally, output the midpoint of the interval as the encoding result:

[0163] Encoded output = Binary representation of midpoint of [low, high]

[0164] In this way, arithmetic coding can convert DCT coefficients into a compact binary code, further reducing data storage and transmission volume.

[0165] The entire data acquisition, processing, compression and transmission process constitutes a highly integrated closed-loop control system, each link coordinates through real-time feedback mechanism, ensuring the efficiency and stability of data transmission. In this system, the operation and decision of each step can be adjusted in real time according to the environmental changes and data characteristics, so as to cope with the dynamic changes of working conditions.

[0166] The core of closed-loop control is context-aware modeling. The system can dynamically adjust the processing method of redundant data according to the sensor type and current environmental conditions. Through real-time acquisition of environmental parameters such as signal-to-noise ratio (SSNR), the system adjusts the redundancy standard flexibly, avoids unnecessary data transmission, and improves transmission efficiency and reduces energy consumption. With the change of environment, the calculation of redundancy threshold can respond in time. When the signal is weak, the threshold is relaxed to ensure the stability of data transmission, and vice versa. When the signal is good, the transmission amount of redundant data is reduced, further improving the efficiency.

[0167] Compression transmission strategy is also a key link in closed-loop control. The system will select the appropriate compression strategy to optimize data transmission according to real-time data characteristics and environmental conditions. When the redundancy is high, a stronger compression method is used; in adverse environmental conditions, a lightweight compression is selected, that is:

[0168] If SSNR > 25 dB, the first eight DCT coefficients are retained:

[0169] X k = [X0, X1, …, X7] (SSNR > 25 dB)

[0170] If SSNR < 20 dB, the first twelve DCT coefficients are retained:

[0171] X k = [X0, X1, …, X 11 ] (SSNR < 20 dB).

[0172] For the data of the low-redundancy layer, during data transmission, the overall energy efficiency of the system is further optimized by adjusting the transmission frequency, bandwidth, and power.

[0173] To ensure the stability and accuracy of data transmission, the system dynamically adjusts the transmission frequency to adapt to changes in data volume and signal conditions, thereby avoiding unnecessary bandwidth waste and prolonging the service life of sensor devices.

[0174] The detection data of the passive sensor is D = [d1, d2, …, d n ], which is divided into multiple levels according to the degree of redundancy, and each level has different threshold values for data redundancy; the redundancy R k of the kth level is:

[0175]

[0176] where n k represents the number of data points in the kth level, d i is the original data point collected, is the prediction or expected value of the current level, and the redundancy R k indicates the degree of data redundancy in this level.

[0177] The data is pre-processed, and only data with actual significance is transmitted, reducing the transmission burden of invalid data. This filtering mechanism can automatically optimize according to the changes in real-time data flow and historical trends, ensuring the efficiency of data transmission.

[0178] Finally, through the feedback mechanism of the closed-loop control system, the system can respond in real-time to changes in signal quality and environmental conditions, and flexibly adjust various operations, thereby maximizing the efficiency of data transmission and ensuring stable operation of the system even in complex environments. The invention effectively guarantees the efficiency and low-power characteristics of the system during long-term operation, further improving the overall architecture of the passive sensor data transmission optimization system based on context awareness and CDT-LC dynamic redundancy threshold calculation.

[0179] Millisecond level dynamic acquisition of environmental parameters, suitable for monitoring mechanical impact with passive vibration sensors and monitoring instantaneous temperature with passive temperature sensors. Fixed sampling windows of 1 ms, 10 ms, 20 ms, and up to 1000 ms can be used, and data continuity can be ensured through timestamp alignment.

[0180] Second level steady-state acquisition of environmental parameters, suitable for monitoring temperature and humidity and slowly changing parameters such as normal operation of equipment. Intervals of 1 s, 2 s, 3 s, 10 s, and 100 s can be used for periodic sampling.

[0181] In the data compression transmission of the passive temperature sensor, the binary code is linked into the periodic transmission buffer through the low-frequency channel, and the channel occupancy rate is reduced to below 15%.

[0182] It can be understood that the above specific description of the present application is only used to illustrate the present application and is not limited to the technical solutions described in the embodiments of the present application. Those skilled in the art should understand that the present application can still be modified or replaced equivalently to achieve the same technical effect; as long as the use needs are met, it is within the protection scope of the present application.

Claims

1. A passive sensor data transmission optimization method, characterized in that... Includes the following steps: Step 1: Monitor the environmental parameters of the passive sensor, using these parameters as contextual information; Step 2: Calculate the dynamic redundancy threshold based on context information; Step 3: Compress and transmit the data collected by the sensor; Step 4: After the data is compressed and transmitted, obtain feedback information on transmission quality and energy utilization efficiency, and return the feedback information to Step 2 to form a closed-loop control. Step 2 includes the relationship between environmental parameters and redundancy threshold, hierarchical coding and redundancy calculation, and dynamic adjustment of redundancy threshold. Environmental parameters determine the setting of redundancy threshold, and redundancy calculation affects the redundancy threshold adjustment strategy. By calculating data redundancy, the redundancy threshold is adjusted according to the deviation between the redundancy and the current redundancy threshold. The updated redundancy threshold guides the data compression and transmission strategy. The relationship between the environmental parameters and the redundancy threshold is as follows: The initial redundancy threshold is set to T0. The relationship between the dynamic redundancy threshold T and the environmental parameter vector X is as follows: T = T0 × f(X) Where f(X) is a mapping function, representing how the environmental parameter vector X affects the redundancy threshold T; α SSNR α T and α H The influence weights of SSNR, temperature, and humidity on the redundancy threshold are controlled separately. ref T ref and H ref For reference only; SSNR ref Select the signal-to-noise ratio value measured in an environment with no signal interference or low background noise; T ref : This is the temperature value measured under standard experimental conditions, which is the ideal temperature value set during sensor testing; H ref Select a standard humidity value measured under laboratory conditions; T env Ambient temperature parameter; H env : This refers to the ambient humidity parameter.

2. The passive sensor data transmission optimization method according to claim 1, characterized in that... The environmental parameters in step 1 include signal-to-noise ratio, multipath fading coefficient, ambient temperature, and ambient humidity; the signal-to-noise ratio is the power ratio between the received signal and the ambient noise when the passive sensor sends data to the base station or receiver; the multipath fading coefficient is the degree of fading of the received signal due to the multipath propagation characteristics in the channel where the passive sensor is located.

3. The passive sensor data transmission optimization method according to claim 2, characterized in that... The passive sensor converts the 2.4GHz microwave energy emitted by the base station into DC power through a microwave rectifier circuit. The relationship between the conversion efficiency η and SSNR is as follows: in, For received power; is the transmit power; k is a constant coefficient that comprehensively reflects the conversion efficiency of the rectifier circuit, antenna coupling loss, and device characteristics. The environmental parameter vector is: X=[SSNR,T env ,F env ,vibration freq_peak ,vibration band_energy ] T env For ambient temperature parameters, H en v represents the ambient humidity parameter. freq_peak The peak frequency of vibration. band_energy The vibration bandwidth energy; the environmental parameter vector serves as contextual information.

4. The passive sensor data transmission optimization method according to claim 1, characterized in that... The layered coding and redundancy calculation section: The detection data of the passive sensor is D = [d1, d2, ..., d n The data is divided into multiple levels based on its redundancy level, with each level having a different threshold for redundancy; the redundancy level R of the k-th level is... k for: Where, n k d represents the number of data points at level k. i The raw data points collected, The redundancy R is the predicted or expected value at the current level. k This indicates the degree of data redundancy at this level.

5. The passive sensor data transmission optimization method according to claim 1, characterized in that... The part about dynamically adjusting the redundancy threshold: Redundancy threshold T k The adjustment rules are expressed by the following formula: Among them, T k base This is the basic redundancy threshold for this level; α k This is the adjustment coefficient at this level, controlling the sensitivity of the redundancy threshold to environmental changes; T ref and H ref It is the set standard value; α T and α H It is an adjustment coefficient that controls the degree to which changes in temperature and humidity affect the threshold.

6. The passive sensor data transmission optimization method according to claim 5, characterized in that... When SSNR < 20dB, T k base Price increased by 20% to 30%; When SSNR > 25dB, T k base A decrease of 10% to 15%.

7. The passive sensor data transmission optimization method according to claim 1, characterized in that... In step 3, the data compression transmission of the passive temperature sensor is as follows: The original temperature data sequence is D = [d1, d2, ..., d n The differential encoding process is as follows: Δd i =d i -d i-1 ,i=2,3,…,n Where, d i For the i-th data point, d i-1 For the previous data point, Δd i The difference between data points; the original data sequence is transformed into a difference sequence; Based on differential coding, Huffman static dictionary is used for further compression. The frequency of the data points after differential coding is counted, and the data points are converted into binary code through frequency mapping to complete the data compression. The compressed data is transmitted through the communication link between the passive temperature sensor and the receiving end.

8. The passive sensor data transmission optimization method according to claim 1, characterized in that... In step 3, the data compression transmission of the passive vibration sensor is as follows: Time-domain vibration data x n The formula for DCT transformation is: Among them, X k The transformed frequency domain data; P(X k ) represents the k-th low-frequency coefficient X k The probability of represents the frequency of the k-th coefficient appearing among all frequency domain coefficients; P(X k The distribution satisfies: Arithmetic encoding divides the entire interval [0,1) into subintervals, each subinterval representing a coefficient. Initially, low = 0, high = 1. For each coefficient X k The corresponding interval is [low] k high k Update range: Continue this operation until all coefficients are encoded; the final output interval's midpoint is the encoded result. Encodedoutput=Binaryrepresentationofmidpointof[low,high].

Images