An adaptive activation method and device for a sensor array unit

Abstract

The present invention relates to a method and device for adaptively activating a sensor array unit. The method includes the following steps: S1: Number each node in the sensor array unit and assign initial weights, and set the probability interval for each node according to the weight of the node; S2: Randomly wake up the nodes in the sensor array unit according to the probability intervals of the nodes; S3: Adjust the weight of the node according to the change in the reading corresponding to the awakened node, so as to change the probability interval of the node, and then return to step S2. Compared with the prior art, the present invention designs an adaptive mechanism, reduces the detection range, enables the sensor array to focus on the area with targets for detection, reduces the detection energy consumption, reduces the information storage space, and improves the detection effect.

Description

Technical Field

[0001] The present invention relates to the technical field of sensors, and particularly to a method and device for adaptively activating a sensor array unit. Background Art

[0002] When traditional sensor arrays read data, the array nodes read data in sequence according to the time series. For some array nodes, the information obtained does not represent the target information during transmission, that is, the sensor is activated but does not promptly reflect the target state, resulting in waste of energy consumption.

[0003] Taking a traditional camera as an example, although it can detect a lot of pixels, with a large number of pixels and good effects, according to the array sequential reading method of waking up all nodes in each time series, many pixels cannot provide timely and effective information for detecting the target, resulting in unfocused detection and high energy consumption; and as the scale of the array continues to increase, the problem of ineffective wake-up of nodes will become more serious. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned defects of the existing technology that all nodes are awakened in each time series and there is no ability to adaptively detect significant targets, resulting in low output efficiency of the sensor array. As the scale of the array continues to increase, the problem of ineffective wake-up of nodes will become more serious, and a method and device for adaptively activating a sensor array unit are provided.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for adaptively activating a sensor array unit includes the following steps:

[0007] S1: Number each node in the sensor array unit, assign an initial weight, and set the probability interval of each node according to the weight of the node;

[0008] S2: Randomly wake up the nodes in the sensor array unit according to the probability intervals of the nodes;

[0009] S3: Adjust the weight of the node according to the change in the reading corresponding to the awakened node, thereby changing the probability interval of the node, and then return to step S2.

[0010] Further, step S1 further includes defining the maximum value and the minimum value of the weight of each node, and the weight of each node changes within the maximum value and the minimum value corresponding to the node.

[0011] Further, in step S3, the weight of the node is adjusted according to the degree of change in the reading corresponding to the awakened node.

[0012] Further, in step S3, the adjustment expression of the node weight is as follows:

[0013] weight i,t = weight i,t-1 + Δweight i,t

[0014] In the formula, weight i,t is the weight of node i at time t, weight i,t-1 is the weight of node i at time t - 1, and Δweight i,t is the weight adjustment amount of node i at time t.

[0015] Further, the calculation expression of the weight adjustment amount Δweight i,t of node i at time t is as follows:

[0016]

[0017] In the formula, Δread i,t is the difference between the reading of node i at time t and the previous reading of this node, and avg t is the average value of the reading differences when the array is woken up the previous n times in the sensor array unit, and std t is the standard deviation of the reading differences when the array is woken up the previous n times in the sensor array unit.

[0018] Further, the value of n is 25.

[0019] Further, if Δweight i,t is less than 0, the value of Δweight i,t is updated at a smaller rate, and in step S3, the adjustment expression of the node weight is changed to:

[0020] weight i,t = weight i,t-1 + 0.1×Δweight i,t

[0021] Further, in step S1, the initial weights of all nodes are equal.

[0022] Further, in step S2, one node in the sensor array unit is randomly woken up each time.

[0023] The present invention also provides a sensor array unit adaptive activation device, including a memory and a processor. The memory stores a computer program, and the processor calls the computer program to execute the steps of the method described above.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] (1) The present invention adopts a brain-inspired method and proposes an adaptive activation method for sensor array units based on the Monte Carlo method. For example, a frog pays much more attention to objects with frequent movement in vision than to stationary objects. This adaptive activation method also uses the Monte Carlo method to determine the node positions where moving objects are frequently detected, thereby increasing the probability of waking up nodes with high detection heat, reducing the probability of waking up nodes where targets are not frequently detected, and making the utilization of the array more efficient.

[0026] (2) Compared with the detection method of activating all array nodes each time, the method of the present invention has the advantage of low energy consumption.

[0027] (3) In terms of information preservation, since the adaptive detection method wakes up only one node each time, compared with waking up all detection nodes each time, it has the advantages of less information preservation and less memory requirement.

[0028] (4) As the scale of the array continues to expand, the time of the traditional detection method will increase linearly. However, since the method of the present invention is based on the way of wake-up heat, the number of wake-up nodes is extremely small compared with the traditional detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic flow chart of an adaptive activation method for a sensor array unit provided in an embodiment of the present invention;

[0030] Figure 2 It is a probability distribution diagram of sensor activation when there is no target provided in an embodiment of the present invention;

[0031] Figure 3 It is a probability distribution diagram of sensor activation when there is a target provided in an embodiment of the present invention;

[0032] Figure 4 It is a schematic diagram of detection results based on a traditional sequential polling method provided in an embodiment of the present invention;

[0033] Figure 5 It is a schematic diagram of detection results based on the adaptive method of the present invention provided in an embodiment of the present invention;

[0034] Figure 6 It is a frequency distribution histogram of no target provided in an embodiment of the present invention;

[0035] Figure 7 It is a frequency distribution histogram of having a target provided in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Components of the embodiments of the present invention described and illustrated herein generally may be arranged and designed in a variety of different configurations.

[0037] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0038] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0039] Embodiment 1

[0040] This embodiment provides a method for adaptively activating a sensor array unit, including the following steps:

[0041] S1: Number each node in the sensor array unit, assign initial weights, and set probability intervals for each node according to the weights of the nodes;

[0042] That is, first define the data structure of a single node, including information such as label, data acquisition value, weight, etc.; preferably, also define the maximum and minimum weights for each node to be awakened to control the awakening probabilities of the nodes with the highest and lowest heat; in addition, define a probability interval to save the awakening probability of each node.

[0043] S2: Randomly awaken the nodes in the sensor array unit according to the probability intervals of the nodes;

[0044] S3: Adjust the weight of the node according to the reading change corresponding to the awakened node, thereby changing the probability interval of the node, and then return to step S2.

[0045] That is, randomly awaken the sensor nodes, determine the magnitude of the awakening probability according to the change of the node readings, so that the nodes with high heat are more likely to be activated, and the nodes with low heat are not easily activated, and concentrate the detection attention on the nodes with high heat.

[0046] Continuously adjust the weights of the nodes in the probability interval according to the awakening heat of the activated sensor nodes, dynamically adjust the weights of each node, and achieve an adaptive effect.

[0047] In step S3, according to the degree of change of the readings corresponding to the awakened nodes, the weights of these nodes are adjusted.

[0048] The specific process is shown in Equation 1 below, which is the formula for the weight adjustment method.

[0049] Where Δweight i,t is the weight of the update probability of the i-th node at time t. The larger this value, the greater the updated weight and the higher the probability of being detected. Δread i,t is the difference between the reading of node i at time t and its previous reading, and avg t is the average value of the reading differences during the first 25 awakenings of the array in the sensor array unit, and std t is the standard deviation of the reading differences during the first 25 awakenings of the array in the sensor array unit. If Δweight i,t < 0, then Δweight i,t = 0.1 × Δweight i,t , reducing the probability of the target node with small reading changes.

[0050]

[0051] As shown in Equation 2, the weight value updated each time is the sum of the weight at time t - 1 and the changed weight value.

[0052] weight i,t = weight i,t-1 + Δweight i,t (2)

[0053] As Figure 1 shown, the specific implementation process of the adaptive activation method of the sensor array unit in this embodiment is as follows:

[0054] 1) Use C language or Python as the programming language, use the STM32F103ZET6 single-chip microcomputer as the controller for reading the sensor array, and use the HCSR04 ultrasonic sensor as the node in the sensor array.

[0055] 2) Set the algorithm parameters: The limits for increasing and decreasing the weight each time, and the maximum and minimum values of the weight of a single node, respectively affect the sensitivity of the sensor array to changes and the probability of a single sensor being activated.

[0056] 3) First, number all the sensor nodes and assign equal initial activity weights, so that the probability of all nodes being activated at the beginning in the array is equally likely.

[0057] 4) The sensors are randomly activated. For sensors with large reading changes, their weights are increased, while for sensors with small reading changes, their weights are decreased, and the read data is saved.

[0058] 5) The activity weights of each sensor are simulated as the probability intervals of each sensor node. The length of the interval represents the weight size. The longer the interval length, the higher the possibility of being selected.

[0059] 6) To ensure that the sum of the wake-up probabilities of the sensor array nodes is equal to 1, the maximum value of the total length of the probability intervals should be set. To ensure that each sensor array has the opportunity to wake up, the minimum value of the length of the probability interval of a single node should be set.

[0060] For the traditional sequential polling method for reading sensor arrays, since all nodes need to be woken up for each polling read, there are often some nodes that are not easy to detect the target, but they will still be woken up with equal probability in the next time series, resulting in high energy consumption. This patent designs an adaptive mechanism to narrow the detection range, enabling the sensor array to focus on the areas with target changes for detection, reducing the detection energy consumption, decreasing the information storage space, and improving the detection effect.

[0061] To verify the actual effect of this method, this embodiment verifies the reliability of this method and compares the characteristics of the adaptive method of this patent with the traditional method for array reading.

[0062] This embodiment uses an ultrasonic sensor array and is controlled by an STM32 series single-chip microcomputer. The following demonstration effects take twenty-five of them as an example.

[0063] When the array does not detect target changes, the weight intervals of each sensor are equal, and each time a node is randomly woken up with equal probability. Figure 2 is the probability of each node being woken up at the beginning.

[0064] Figure 2 What is shown is the activation probability of 25 sensor nodes without being activated, that is, when no target to be detected is found within the detectable range, it can be found that the probabilities are exactly the same.

[0065] When the randomly selected node detects target changes, the weight interval of this node changes. As the detection time continues to increase, the wake-up probability of the nodes in the area with a large change in the detected target distance will increase due to the increase in the probability interval, while the wake-up probability of the nodes in the area with a small change in the detected target distance will decrease conversely. As Figure 3 shown.

[0066] Figure 3It shows that the target to be detected appears among 25 sensor nodes, and some nodes are repeatedly activated, increasing the probability of activation of nodes with high activity, making their attention focus on nodes with more changes in the detected target rather than nodes that often fail to detect the target, thereby improving the utilization efficiency of array transmission.

[0067] The traditional sequential polling detection method has a good effect, and the images detected at each detection time interval have little difference. The non-black area is the information obtained by nodes with targets, and the black area is the information obtained by nodes that fail to detect the target. Although the target is not detected in these black areas each time, each node will still be awakened during the next detection process, increasing the waste of energy consumption. As Figure 4 shown.

[0068] Figure 4 It is a sensor array data acquisition method based on the traditional serial time series. Data is acquired once within a time series and then combined and output. It can be observed that the detection effect is good, and the difference in each detection is not large, which is suitable for some small arrays. When the sensor array becomes larger and larger, the problems of high energy consumption and slow detection speed will become more and more obvious.

[0069] Figure 5 It is a sensor array data acquisition method based on the adaptive algorithm. Each time, only one sensor node is randomly activated, and the probability of activation of each node is not equal, depending on the frequency of data change of the node during activation. It can be observed that different from Figure 4 this, since the nodes are not awakened sequentially but randomly, Figure 5 it takes a certain amount of time to detect the complete target. However, as the number of array nodes increases, the detection method used in the present invention will greatly improve the detection efficiency of moving objects, only focusing on the hot spot area, thereby reducing the detection time and improving the detection efficiency.

[0070] Using the adaptive algorithm of this patent, the statistics of the frequency of 25 sensor wake-up times are carried out 2000 times. As Figure 6 shown.

[0071] Figure 6 It represents the statistics of the detection frequency in the case of no target to be detected. It can be observed that the frequencies of activation of each node are not very different. When a target appears in the array, the activation frequency changes, as Figure 7 shown.

[0072] Figure 7 It represents the statistics of the detection frequency in the case where an object can be detected at some nodes. It can be observed that the number of times the sensor nodes are activated at the nodes where there is object movement is significantly more than that of ordinary nodes, achieving the purpose of adaptation.

[0073] This embodiment also provides a sensor array unit adaptive activation device, including a memory and a processor. The memory stores a computer program, and the processor calls the computer program to execute the steps of the above-mentioned sensor array unit adaptive activation method.

[0074] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. An adaptive activation method for a sensor array unit, characterized in that, It includes the following steps: S1: Number each node in the sensor array unit, assign initial weights, and set the probability intervals of each node according to the weights of the nodes; S2: Randomly wake up the nodes in the sensor array unit according to the probability intervals of each node; S3: Adjust the weight of the node according to the change in the reading corresponding to the awakened node, thereby changing the probability interval of the node, and then return to step S2; In step S3, the adjustment expression of the node weight is: weight i,t = weight i,t-1 + Δweight i,t where weight i,t is the weight of node i at time t, weight i,t-1 is the weight of node i at time t-1, and Δweight i,t is the weight adjustment amount of node i at time t; The weight adjustment amount Δweight of the node i at time t i,t is calculated by the following expression: where, Δread i,t is the difference between the reading of node i at time t and the previous reading of this node, avg t is the average value of the reading differences when the array in the sensor array unit is awakened the previous n times, std t is the standard deviation of the reading differences when the array in the sensor array unit is awakened the previous n times.

2. The adaptive activation method of a sensor array unit according to claim 1, wherein Step S1 further includes defining the maximum and minimum values of the weights of each node, and the weights of each node change within the maximum and minimum values corresponding to that node.

3. A method for adaptively activating a sensor array unit according to claim 1, wherein In step S3, adjust the weight of the node according to the degree of change in the reading corresponding to the awakened node.

4. A method for adaptively activating a sensor array unit according to claim 1, characterized in that, The value of n is 25.

5. A method for adaptively activating a sensor array unit according to claim 1, characterized in that If Δweight i,t is less than 0, then Δweight i,t is updated at a smaller rate. In step S3, the adjustment expression for the node weight is changed to: weight i,t = weight i,1-1 + 0.1 × Δweight i,t .

6. A method for adaptively activating a sensor array unit according to claim 1, characterized in that, In step S1, the initial weights of each node are equal.

7. A method for adaptively activating a sensor array unit according to claim 1, characterized in that, In step S2, randomly wake up one node in the sensor array unit each time.

8. An adaptive activation device for a sensor array unit, characterized in that, It includes a memory and a processor. The memory stores a computer program, and the processor calls the computer program to execute the steps of the method according to any one of claims 1 to 7.

Images