Multisensor Information Fusion and Optimization Technique for Wildfire Detection and Prevention in WSN Environment

doi:10.21203/rs.3.rs-920835/v1

Download PDF

Research Article

Multisensor Information Fusion and Optimization Technique for Wildfire Detection and Prevention in WSN Environment

https://doi.org/10.21203/rs.3.rs-920835/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Monitoring natural traits like temperature and humidity are imperative for evaluating climate change. Remote sensing applications in wireless sensor networks face a crucial issue in transmitting the information from sensors in a continuous period to the gateway nodes for processing. Correspondingly, sensor nodes are subject to low memory and processing limitations in concern to wireless sensor networks. Wildfires can be started intentionally or unintentionally. The result, however, is disastrous, causing irreversible damage to the ecosystem. With the advancement of intelligent sensor-based control technologies, data fusion and classification have become an essential component of the environmental monitoring system. To predict the occurrence of fire in forests, a fast and effective intelligent control system using Adaptive Decentralized Kalman Filter is proposed. The Decision tree classification can successfully solve problems on the side of the sink node. Thus the described Decision tree classification determines whether fire detection parameters are in the acceptable range and further utilizes a fuzzy-based optimization to improve the performance of the considered complex fusion environment. The experimental results of the proposed model have a detection rate of 98.3 %, demonstrating that the proposed technique is robust. Thus, providing real-time monitoring of certain environmental variables for continuous situational awareness and instant responsiveness.

decision tree

communication

wildfire

data fusion

wireless sensor networks

Forests help to maintain the earth's ecological balance. Unfortunately, forest fires are only detected after they have spread over a large area. Uncontrolled fires primarily occur in forest areas that have invaded agricultural lands. Wildfires are caused by either intentional or unintentional causes. However, the result is disastrous, causing irreparable damage to the ecosystem [9]. The Australian bushfires began in 2020 and lasted until March 2021, killing nearly 3 billion animals [28]. Mammals, birds, frogs, and reptiles are all included. For example, in the year 2020, the world will be confronted with numerous massive wildfires and forest fires. The Australian bushfires of January 2020 destroyed six million acres of land. Thousands of people were evacuated from Arizona on June 28, 2020, due to a wildfire. On September 8, 2020, California faced more than forty-nine new wildfires that had been burning for several weeks.

Destructive wildfires are becoming an annual event, similar to climate change [28]. As a result, there is an increase in death trolls, resulting in catastrophes that wreak havoc on both humans and the environment. Firefighters are currently on the scene inspecting the remaining fire, but their safety is not guaranteed [4]. Despite increased efforts, fire disasters cannot be effectively predicted. Early detection and monitoring of fires can make immediate evacuation possible [27]. Monitoring potentially dangerous areas and providing early warning of fires will greatly reduce response time, potential damage, and firefighting costs [20]. As a result, it is critical to identify the fire situation and accurately determine the ignition point and real-time dynamics [3].

With the advancement of sensor technologies, multi-sensor information fusion has become an essential component of the environmental monitoring system [25]. The multi-sensor technology detects signals from various sources, analyses the specific parameters, and integrates them as output data. Multi-sensor fusion technology has received considerable attention in applications related to condition-based environment monitoring systems. The data obtained in this manner, however, is redundant, ambiguous, and imprecise. The key aspect of the multi-sensor fusion model is that it omits the redundancy of the data, providing robust and accurate results [19]. The current multi-sensor data fusion algorithm used for fire detection is incapable of effectively removing error data, resulting in false alarms and data omission. The Kalman filter algorithm performs the fusion of a variety of data by removing noise [24]. Therefore, the proposed system is an enhancement of the Kalman filter, fulfilling the objective of the proposed research.

In a wireless sensor network, the sensor nodes are deployed randomly in robust locations. Wireless sensor networks are widely used in environmental surveillance, particularly for low-cost fire detection applications. WSN smart modules can perform data processing, ad hoc communication, remote monitoring and control, and so on. The use of WSN for fire detection has proven to be effective in terms of providing accurate and reliable information [12]. Wi-Fi and ultra-wideband signals are being developed to provide a non-GPS solution [29]. Sensor measurements are inaccurate due to variations in temperature, pressure, noise, and so on. In addition, in a forest environment, sensors may stop working due to a lack of energy [5]. Such factors are incapable of providing reliable and accurate information [6]. In addition to this issue, WSN sends a large amount of redundant data to the sink. The above-mentioned issues can be avoided by employing the data fusion technique [17].

The multi-sensor data fusion technique integrates data from multiple sensor nodes and transmits the combined data to the base station, reducing communication overhead and saving sensor energy. However, to use a WSN to detect a fire in its early stages, existing environmental factors such as temperature, relative humidity, wind speed, gas detection, and so on must be known [22]. The location of the incident and the hotspot can sometimes have an impact on early fire detection systems [13]. The primary objective of the proposed system is to detect the fire as quickly as possible and to identify its exact location, allowing fire units to respond as soon as possible. To predict the occurrence of fire in forests, a fast and effective ADKF-DT-MF (Adaptive Decentralized Kalman Filter with Decision Tree-based Multi-sensor Fusion) fire detection technique is proposed. The following are the primary contributions:

The ADKF-DT-MF technique attempts to remedy this by detecting a forest fire at an early stage, increasing or ensuring the chance of putting it out before it grows out of control or causes significant damage.
The use of multiple sensor sources and the deployment of sensor nodes in areas not visible to the satellite to collect data on temperature and humidity, among other things, increase the possibility of more accurate and early fire detection.
The ADKF-DT-MF technique obtains redundant and reliable data from multiple sensors for processing, monitoring, tracking, detection, and communication purposes to create environmental awareness.
To perform an intricate intelligent decision-making system with fuzzy optimization capabilities.
To prevent data estimation delays on the side of the fusion center, helping to incorporate knowledge using a decision tree tool. If any of the sensors fails, the information on the current state is incomplete. Hence fuzzy optimization here ensures low sensing cost.
When a hazard is detected, the cloud platform and online SMS for receiving alert messages are integrated for communication purposes. As a result, communication overhead and battery consumption are reduced. Further, to verify the effectiveness of the proposed system, it is being compared with the existing ENN technique.
Thus, providing real-time monitoring of certain environmental variables can improve the effectiveness of forest fire prevention, detection, and control by allowing for continuous situational awareness and instant responsiveness.

The rest of the article's section is organized as follows: Sect. 2 provides a review of the literature related to forest fire detection. Section 3 describes in detail the ADKF-DT-MF-based fire detection technique. Section 4 illustrates the experimental results and discussion, and Sect. 5 concludes and discusses its future scope.

In [23] the authors proposed a reasoning method for establishing a fusion model by combining extension theory and a neural network (ENN). The proposed model was then used to develop a fire detection system. The fusion system integrates at various levels and is classified in to four types: i) centralized, ii) decentralized, iii) mixed type, and iv) feedback type. The sensor signal is initially received and processed by the extension neural network. The number of input nodes is determined by the input layer, and the classification result is provided by the output layer. The layers are joined together using a double weight connection. Three input signals are considered to determine the fire: temperature, smoke, and carbon monoxide concentration. The network is trained to understand the model's validity and superiority. Although the system demonstrated rapid convergence in fire detection, there were training errors, and the fire identification rate was only 93.9 %. In [8] conducted a literature review on video flame and smoke-based fire detection algorithms for indoor and outdoor environments. Forest fires cause environmental damage that takes decades to repair. It was previously detected using a point sensor based on heat, smoke, and flame. In addition, there is gas. Particles that reach the sensor body are activated. When it comes to responses, there is a lag. However, the model is appropriate for small rooms but not for large open spaces. Furthermore, fire flames have characteristics such as color and motion, which point sensors do not use to detect fire.

The authors [28] investigated fire sensor multisensor information fusion using a backpropagation neural network to ensure firefighter safety. Robots were used in the study to identify the remaining fires, lowering labor costs. Temperature, infrared signals, smoke, and carbon monoxide were all considered as fusion parameters. The temperature sensors DS18B20, as well as the smoke and carbon monoxide sensors from the MQ-2 and MQ-7 were used. The fire detection algorithm is a cross between Artificial Neural networks and fuzzy algorithms. Although the robots' fire detection assisted the firefighters, fire detection in dense forest areas, as well as fires caused by environmental factors such as high temperatures, continue to be challenging. In [11] the authors sought to summarize the onboard multisensor configuration for ground vehicle detection in off-road conditions. Intelligent ground vehicles are used for both civilian and military purposes, such as rescue and surveillance. When working with condition-based environmental applications, it is critical to understand the different types of sensors and their performance. Range-based, image-based, and hybrid sensors are the three types of sensors used in intelligent vehicles. The multisensor fusion method is the most effective way to perceive information in a complex environment. Sensor fusion methods are classified as probability, classification, and inference-based. Despite the author's description of the different types of sensors and methods, integrating them into emerging technologies remains a challenge.

In [15] proposed the "Safe From Fire (SFF)" intelligent self-controlled IoT-based fire extinguisher system. The system receives data from multiple sensors located throughout the victimized area. Fuzzy logic integration monitors the location and severity of fire outbreaks. The system is aided by data fusion algorithms. Text messages and phone calls are used to notify people about the fire's existence. The SFF system detected fires with a 95% accuracy rate. However, the system accepted multiple readings and issued repeated warning alarms. In [21] proposed a mobile autonomous firefighting robot for fire detection in an industrial setting. The robot has two driving wheels, and the motor controls the robot's speed and direction. The robot employs a variety of sensors, including proximity, thermal infrared, and light sensors. The degree of matching the input signal to an ideal target is measured. Despite an improvement in detection mode, performance on suppressed and noisy targets was inferior.

In [14] developed hybrid body sensor network architecture to aid in the delivery of medical services. A body sensor network gathers physiological, psychological, and activity data to provide medical services. This necessitates gathering the users' current state. The fixed devices allow for flexible human-robot interaction. The layers generate sensory data that is used to perform a multisensor fusion. Although multisensor fusion ensures quality improvement in fusion, fusion decision-making remains a challenge. In [10] created a smart home system with multisensor fusion technology. The authors created wearable motion sensing devices for residents' feet and wrists, followed by an intelligent fire detection system for home safety [2]. The intelligent system transmits real-time temperature and CO concentration, as well as alarms and appliance status. The system's feasibility and effectiveness were confirmed. However, in two cross-validations, the results showed only a 92.2% recognition rate.

According to [1], IoT devices generate useful data and enable data analysis, allowing for optimal real-time environmental decisions. IoT is the primary source of big data. The quality of information is critical in decision-making, and data fusion can help achieve this. The following are some of the challenges associated with multisensor data fusion: i) Information from multiple sensors must be decisive, sensible, and precise. ii) Low-power sensor battery replacement is complicated and necessitates the use of energy-efficient sensors. iii) Hiding the semantics of fused information in critical applications. iv) Additionally, to obtain accurate information data, ambiguity, inconsistency, and conflicting nature must be removed. As a result, developing a multisensor fusion technique that addresses the aforementioned issues remains a challenge.

To meet the needs of wireless sensor network-based applications, the multisensor data fusion technique, which combines data streams from multiple sensors, must improve accuracy requirements. In recent years, the emergence of multi-application paradigms in wireless sensor networks has enabled sensors to observe their surroundings [7]. Multi-sensor fusion's primary goal is to improve system performance by combining data from all sensors. To provide better spatial and temporal coverage, multiple sensors can be used. Furthermore, using multiple sensors improves estimation accuracy and fault tolerance [26]. To meet the long-term and high-precision requirements, a multi-sensor fusion framework must continuously provide all-time weather parameters [16]. The integrated fusion method is used for isolating measurement failures and fusing state estimates to combine local estimation results obtained from each sensor. Temperature, humidity, and noise are among the physical attributes monitored by the sensors. Additionally, send them at regular intervals. The fixed gateway node receives the transmitted data streams and processes them further. As data transmission speeds up, so does network complexity. When the sensor batteries run out, they cannot be replaced. As a result, it is necessary to increase battery life. Here the influential ADKF-DT multisensor fusion technique is applied as a probable solution to the challenge of earlier identification of forest fires. The proposed framework of the forest fire detection system is depicted in Fig. 1.

The sensors perform data transmission through a wireless medium and are further processed by the proposed technique. Finally, the data is sent to the IoT cloud server which fulfills the objective of communicating early an alert message on the fire threat. The dynamic environmental measurements such as temperature and humidity in real-time are considered input data. The system detects fire and sends an alert message to the station. Depending on the budget, any number of sensors can be used, here the sensors of type Node MCU (ESP8266) interfaced with temperature and humidity sensors, i.e. DHT11. The ADKF-DT-MF algorithm combines and modifies the original scheme of the Decentralized Kalman Filter and Decision Tree algorithms to suit the proposed research objective.

It is necessary to gain knowledge of the observed environment to identify threats. The sensed data must be classified to solve this problem. As a result, the proposed ADKF-DT-MF (Adaptive Decentralized Kalman Filter with Decision Tree for Multisensor Fusion) system is used on a condition-based environment monitoring system to provide valuable real-time data [18]. The data classification technique determines predefined classes providing current event-based situational information in a network of sensor nodes. The predefined types are characterized by the attributes of the sensor output signals. The decision tree classifies the forest weather parameters into maximum and minimum values of temperature, and humidity. The classification with a set of rules interprets and converts into a standard set of values for environment-based situation refinement performing accurate decision making.

The decision-making procedure is employed at the fusion center. Here the large data set is divided into multiple classes, organized into a hierarchical tree structure for efficient classification and decision making. Decision-making shall handle sensor defects, missed out observations/decisions improving the system performance. Once the fire is detected in the perceived area the concerned authorities are communicated regarding the exact location and intensity of the fire which helps to take necessary action. Since sensors exhibit limited energy, the algorithm performs functions such as

predefining the active sensors and determining the shortest path for information transmission,

reduces the amount of data sent to the base station,

sensors that are not active in the path go into sleep mode to save energy.

Algorithm - ADKF-DT-MF (Adaptive Decentralized Kalman Filter with Decision Tree Multisensor Fusion)

Begin //Wireless Sensor Network Configuration//

Declaration

n = 100;w = 2*n; h = 2*n; x, y

Calculate the node distance//Base Network

d = sqrt(xSide^2 + ySide^2);

Calculate the distance to previous node

xSide = abs(X2-X1);

ySide = abs(Y2-Y1);

d = sqrt(xSide^2 + ySide^2);

Apply ADKF Filter//To remove noise and perform data fusion

Calculates cost function for known and unknown variables

CostFunction = @(tour) TourLength(tour,model.D);

nVar = model.n; //number of unknown variables

VarSize = (1 nVar); //unknown variables matrix size

Identify the best cost path

BestCost = zeros(MaxIt,1);

BestCost(it) = BestSol.Cost;

Perform Decision Tree Classification accuracy

depth = 5; % tree depth

noc = 1000; % number of candidates at each tree node

Y1 = Y1-1;

error = sum(Y1 ~ = Y);

accuracy = 1-error/length(Y);

Read the Channel ID and Field ID(Temp, Humidity)

Display

maxTempF//Maximum temperature for the past 24 hours

minTempF//Minimum temperature for the past 24 hours

avgHumidity = mean(humidity);

End

The coverage function (FC) for a fire hotspot region is expressed in Eq. (1)

$$FC=f(x,y,t)$$

The sensing coordinates are x and y, and the time is denoted by t. Eq. (2) expresses the equation for a temperature sensor, while Eq. (3) expresses the equation for a fire event.

$$f(\mathop x\nolimits_{i} ;\mathop \gamma \nolimits_{0} ,\alpha )=\frac{1}{{\sqrt {2\prod \alpha } }}\mathop e\nolimits^{{ - {{(\mathop x\nolimits_{i} - \mathop \gamma \nolimits_{0} )}^2}/2{\alpha ^2}}}$$

$$f(\mathop x\nolimits_{i} ;\mathop \gamma \nolimits_{F} ,\alpha )=\frac{1}{{\sqrt {2\prod \alpha } }}\mathop e\nolimits^{{ - {{(\mathop x\nolimits_{i} - \mathop \gamma \nolimits_{F} )}^2}/2{\alpha ^2}}}$$

It is necessary to gain knowledge of the observed environment to identify threats. The amount of energy required for data transmission is determined by the size of the data packet and the distance between the sender and receiver. The global estimation is calculated by the Adaptive Decentralized Kalman Filter at the sink node. The algorithm, which is predefined in terms of parameters and functions, identifies the adjacent and most cost-effective shortest path to the sink node while synthesizing accurate data in information fusion and interpreting a series of continuous-time awareness of climate change.

Regardless of sensor failure, this approach provides accurate data from predefined paths. The fusion operation is carried out by the fusion centre. The optimal estimation is calculated by the signal level fusion stage, which keeps track of the sensor's estimated state and its uncertainty. The inferred parameters could be inaccurate and uncertain observations. The filter evaluates the incoming evidence recursively, yielding estimates of the state. The system measurement is fed into the filter, and the output is the estimated system state. The filter's dynamic model calculates the time between sensor outputs and sequentially incorporates multiple sensor measurements for data fusion. It also effectively handles uncertainty resulting from the covariance calculated while transmitting noisy sensor data observed within the boundary limits to determine the sensor state.

The proposed filter improves on previous methods because measurements do not need to be inverted, making it suitable for real-time processing. The filter accomplishes this by calculating timeframes between multiple sensor measurements for information fusion. Similarly, the experiment continues to manage uncertainty resulting from the covariance calculated in Eq. (4).

$$\operatorname{cov} =(x1,x2)=E(x1 - \overline {{x1)}} (x2 - \overline {{x2)}}$$

Where x1 and x2 are independent random variables. The sensor state is determined at this point by the transmission of data within the boundary limits. The system performs observation z using Gaussian noise w as expressed in Eq. (5).

$$z(k)=H(k)x(k)+w(k)$$

Where x (k) represents the system state vector and w (k) represents the unknown zero mean. The proposed technique's efficiency is also validated by comparing it to the existing ENN technique [23]. Here it develops a decision-making system for situation refinement in a multisensor fusion environment with optimization capabilities. It prevents data estimation delays on the side of the fusion center, helping to incorporate knowledge using a decision tree tool. If any of the sensors fails, the information on the current state is incomplete. Hence optimization here ensures low sensing cost calculating the best path. Tests on simulated data carried out by applying the fuzzy membership optimization function show improvement in the performance of the proposed intelligent control system. Each node in the tree acts as a test case for some attribute and the edge is the possible answer to the test case. The process is recursive and is repeated for every sub-tree. For the classification and regression tree at each level, the root node is considered for attribute selection. The selection of attributes at the root node on each level depends mainly on: Entropy, Information Gain, and Gini index. These criteria calculate the attribute value, the attribute of high value is considered for Information Gain and placed at the root node. The continuous attribute value is calculated using the Gini index. The entropy for a single attribute is mentioned in Eq. (6)

$$\text{E}\left(\text{S}\right)=\sum _{\text{i}=1}^{\text{c}}-{\text{p}}_{\text{i}}{\text{l}\text{o}\text{g}}_{2}{\text{p}}_{\text{i}}$$

Where S, the current state; ${\text{p}}_{\text{i}}$the probability of event i of state S; c, the class where i node exists. For multiple attributes, the entropy is as in Eq. (7)

$$\text{E}\left(\text{S},\text{X}\right)=\sum _{\text{c}=\text{X}}\text{P}\left(\text{c}\right)\text{E}\left(\text{c}\right)$$

Here X, the selected attribute. Information gain is given in Eq. (8) the statistical property separates the attribute for their target classification. An increase in information gain decreases entropy.

$$\text{I}\text{n}\text{f}\text{o}\text{r}\text{m}\text{a}\text{t}\text{i}\text{o}\text{n} \text{G}\text{a}\text{i}\text{n}\left(\text{S}, \text{X}\right)=\text{E}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y}\left(\text{S}\right)-\text{E}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y} (\text{S},\text{X})$$

i.e. substituting before, dataset before the split, and subset j after split for k number of subsets the information gain modified as shown in Eq. (9)

$$\text{I}\text{n}\text{f}\text{o}\text{r}\text{m}\text{a}\text{t}\text{i}\text{o}\text{n} \text{G}\text{a}\text{i}\text{n}=\text{E}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y}\left(\text{b}\text{e}\text{f}\text{o}\text{r}\text{e}\right)-\sum _{\text{j}=1}^{\text{k}}\text{E}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y}(\text{j}, \text{a}\text{f}\text{t}\text{e}\text{r})$$

The cost function corresponds to the Gini index, which evaluates the splits in the dataset favoring large partitions. Whereas information gain favors small partitions with distinct values. The Gini is computed by summing the probability${\text{p}}_{\text{i}}$ of node i being chosen times the probability as shown in Eq. (10), a mistake in categorizing that node.

$$\sum _{k\ne i}{p}_{k}=1-{p}_{i}$$

The Gini index reaches a minimum zero value when all cases in the node belong to the same target category. The Gain Ratio is calculated as in Eq. (11)

$$\text{G}\text{a}\text{i}\text{n} \text{R}\text{a}\text{t}\text{i}\text{o}=\frac{\text{I}\text{n}\text{f}\text{o}\text{r}\text{m}\text{a}\text{t}\text{i}\text{o}\text{n} \text{G}\text{a}\text{i}\text{n}}{\text{S}\text{p}\text{l}\text{i}\text{t} \text{I}\text{n}\text{f}\text{o}}$$

Over a given period, real-time environmental datasets for variable temperatures and humidity are observed. The simulation setup was created in Matlab 2018 (a) on a built-in system with an Intel (R) Core (TM) i7-4500U CPU and RAM PC running Windows 10. Furthermore, data communication is carried out via the ThingSpeak IoT platform channels. The sensors were placed at distinct locations so that a look could be kept on the entire forest territory to distinguish the start of disturbing temperature and humidity. The proposed Adaptive Decentralized Kalman Filter (ADKF) is used for data fusion. After fusion, fire detection occurs, and the decision tree, i.e. ADKF-DT-MF, classifies for accurate prediction. The assumption here is that the sensor nodes directionally transmit the temperature and humidity change variations to the sink. Due to the environmental complexity, temperature and humidity are phenomenal features whose characteristics can be effectively used for several applications. The relative certainty of the measurements and the current state estimate are significant to accomplish a fastidious performance. Each sensor node has local estimation transmitted to the nearby node in the predefined path. The sensor node carries its local estimation and the global estimation value transmitted throughout the path. The measurements need not be inverted, appropriate for real-time processing. Decision tree related to rule induction as shown in Table 1 presenting the various range of tables such as low, medium, and high. Each path from the root of the decision tree to one of its leaves is transformed into a rule.

Table 1

Rule list
Rule No.	Condition Values (Low-1, Medium-2, High-3)
1	1	1	1	1	1
2	1	2	1	1	1
3	1	3	2	1	1
4	2	1	1	1	1
5	2	2	2	1	1
6	2	3	3	1	1
7	3	1	2	1	1
8	3	2	3	1	1
9	3	3	3	1	1

The classification starts from the root of the tree comparing the value of the root attribute with the record's attribute. The target variable type is continuous the decision tree here is a continuous variable decision tree. The major challenge is to forecasts the climate change threats in the forest environment and improves the prediction accuracy The decision tree of classification and regression technique was implemented to conquer the accuracy problem. The decision tree of 5 level depth consisting of 1000 candidates at each tree node is modeled. The temperature and humidity changes are observed on an hourly basis. Each recursion happens every 24 hours. The stages are classified into three levels low (1), medium (2), and high (3). The fuzzy decision model considers the membership function as the degree to which the input belongs. The value varies between 0 and 1. The decision tree algorithm is simple to understand and gives a suitable decision tree information discovery. Here the input attributes are set as temperature (low, medium, high), humidity (low, medium, high) with a membership function of between 0 and 1 as the target variable. The decision tree compares the information gain and splits tree based on the three features. The split with the highest information gain will be considered for the first split and the process will continue until the information gain value is zero. The objective is to develop a decision-making system for situation refinement in a multisensor fusion environment. Hence, there is a need to provide optimization capabilities to prevent data estimation delays on the side of the fusion center, helping to incorporate knowledge using decision tree tools. If any of the sensors fails, the information on the current state is incomplete. The environment condition is based on the information obtained from the sensors. The fuzzy-based decision tree model is used to adjust the signaling devices, respectively, and optimizes the system. Thus minimizes the generalization error. The optimization is based on maximum temperature, priority, and distance as depicted in Table 2. The variable value is pure and the level of impurity reaches the value 0. Figure 2 shows the optimized view of the decision tree used to form a rule as shown in Table 2 in the phase time calculation of the proposed fuzzy model.

Table 2

Fuzzy Optimization Constraints
Condition
1	1	1
0	0	1
1	0	0
0	1	0

The proposed intelligent control system consists of two parts: sensor nodes and the ThingSpeak IoT Cloud server. A sensor node is made up of temperature and humidity sensors, as well as a controller board and a power supply. The data from the sensor is collected by the controller and sent to the IoT Cloud service. Where they are kept in blocks known as channels. Each channel saves the current temperature, humidity, and time of update. The data stored in the training data channel is used to perform the training. Due to the limitations of ThingSpeak's free account, which only allows the creation of a maximum of four channels per account, the fused data and predicted data are stored on the same channel. The React App monitors the fire status in the channel, and if a fire is detected, it will trigger the ThingHTTP App, where the SMS gateways APIs are pre-configured during implementation. When the API request reaches the SMS gateway server, the SMS will be sent to the numbers specified in the received request. Finally, the results are sent to the appropriate authorities for immediate action. Figure 3, 4, 5, and 6 depict the input data sensing by Sensor Node-1 and Sensor Node-2. Here temperature and humidity data are considered as input for the system. The graph depicts the temperature and humidity change curves before preprocessing. It has been discovered that the collected data is inferred by external factors.

The data from the sensor node is fused using the ADKF algorithm, and the visualization is done on a separate channel on ThingSpeak. Figure 7 (a) and (b) depict a combined graph visualization of data from Sensor Node 1 and Sensor Node 2, as well as the fused version of both sensor data.

Figure 8 shows the graphical status of fire prediction. When the fire is predicted an alert message is sent to the authorized authority. The volume of data sent is reduced. As a result, the proposed system's anti-inference capability for fire detection has been enhanced. Furthermore, the ADKF eliminates noise data, increasing detection accuracy. The proposed technique is compared to the existing technique [23] to demonstrate its efficiency. The comparison results in terms of detection rate are shown in Fig. 9. Furthermore, the proposed model has a detection rate of 98.3 %, demonstrating that the proposed technique is robust. During rainy seasons, the moisture content of the air is high, indicating a lower risk of fire. Due to a lack of rain during the summer, temperatures rise, the air becomes drier, and the likelihood of a fire increases. However, the sensors' measurements can be inaccurate at times because the sensor is faulty or sends malicious data. Due to a lack of energy, some sensors do not function properly. As a result, reliable or accurate information about the fire event is impossible to obtain. The ADKF-DT multi-sensor fusion technique provides precise and reliable information about a fire event.

Although fire has always been a natural and beneficial part of many ecosystems, climate change, and other human-caused factors are fundamentally changing the frequency, intensity, and patterns of wildfires, as well as their behavior. The proposed classification of fusion data using the ADKF-Decision tree and intelligent fuzzy membership optimized control system for forest fire detection detects wildfires as early as possible before they spread across a large area, thereby preventing poaching. As a result, the experiment was a success in predicting fire, with a detection rate accuracy of 98.3 %. For technical feasibility, only minimum sensor nodes were tested. In addition, the study on connectivity to the IoT cloud server could be considered as future research scope.

Compliance with Ethical Standards

The authors herewith ensure objectivity and transparency in research and have accepted principles of ethical and professional conduct. Also, the research has not involved human participants or animals.

Conflicts of interest

All the authors have no conflict of interest.

Funding

The authors did not receive support from any organization for the submitted work.

Financial interests

The authors declare they have no financial interests.

data Availability

Data sharing does not apply to this article as no datasets were generated or analyzed during the current study.

Code availability

Can be available on request.

Authors' contributions

All the authors equally contributed to preparing the manuscript.

Mehmood., A. F., Katib., R., Albogami, N. N., & Albeshri, A., I, &. (2017). Data fusion and IoT for smart ubiquitous environments: A survey. IEEE Access, 59533–59554
Chou, P. H., Hsu, Y., L, Lee, W., L, Kuo, Y., C, C. C., C, Cheng, Y. S., & Lee, H. H(2017).. Development of a smart home system based on multi-sensor data fusion technology, In 2017 international conference on applied system innovation (ICASI), IEEE, 690–693.
Gupta, C. V., & Singh, R., M. K, &. (2018). A review on forest fire detection techniques: a decadal perspective. Networks, 4, 12
Crowley, M. A., Cardille, J., A, White., J. C., & Wulder, M. A. (2019). Multi-sensor, multi-scale, Bayesian data synthesis for mapping within-year wildfire progression. Remote sensing letters, 10(3), 302–311
de Farias, C., M, P. L., Fortino, G., & Guerrieri, A. (2019). A multi-sensor data fusion technique using data correlations among multiple applications. Future Generation Computer Systems, 92, 109–118
El Abbassi, M., A, Jilbab, A., & Bourouhou, A. (2016). Detection model based on multi-sensor data for early fire prevention, In 2016 International Conference on Electrical and Information Technologies (ICEIT), IEEE, 214–218.
Fengbo, W., Xitong, L., & Huike, Z. (2018). Design and Development of Forest Fire Monitoring Terminal, In 2018 International Conference on Sensor Networks and Signal Processing (SNSP), IEEE, 40–44.
Singh. A, G. A., Kumar., A., Kumar, A., & Kapoor, K. (2020). Video flame and smoke based fire detection algorithms: a literature review. Fire technology, 56(5), 1943–1980
Gong., F., Li, C., Gong, W., Li., X., Yuan. X, Ma, & Song, T. Y, & (2019). A real-time fire detection method from video with multifeature fusion, Computational intelligence and neuroscience, 2019.
Hsu, Y., L, Chou, P. H., Chang, H., C, Lin, S., L, Yang, S., C, Su, H. Y., & Kuo, Y. C. (2017). Design and implementation of a smart home system using multisensor data fusion technology. Sensors, 17(7), 1631
W, H. J., Zheng, B. Y., Wang, C., Zhao, C., H, Hou, X., L … Xu, Z. Q, & (2020). A survey on multi-sensor fusion-based obstacle detection for intelligent ground vehicles in off-road environments, Frontiers of Information Technology & Electronic Engineering, 21, 675–692.
Javadi, S. H., & Mohammadi, A. (2019). Fire detection by fusing correlated measurements. Journal of Ambient Intelligence and Humanized Computing, 10(4), 1443–1451
Liang, Y. H., & Tian, W. M. (2016). Multi-sensor fusion approach for fire alarm using BP neural network, In 2016 International Conference on Intelligent Networking and Collaborative Systems (INCoS), IEEE, 99–102.
Lin., K., Li., Y., Zhou, S. J., & Zhang, Q., D, &. (2020). Multi-sensor fusion for body sensor network in medical human–robot interaction scenario. Information Fusion, 57, 15–26
Mobin, M. I., Abid-Ar-Rafi, M., & Hasan, M. R. Islam. M. N, & (2016). An intelligent fire detection and mitigation system safe from fire (SFF), International Journal of Computer Applications, 133(6), 1–7.
Rachman, F., & Hendrantoro, G. Z, & (2020). A Fire Detection System Using Multi-Sensor Networks Based on Fuzzy Logic in Indoor Scenarios, In 2020 8th International Conference on Information and Communication Technology (ICoICT), IEEE, 1–6.
Rakib, T., & Sarkar, M. R. (2016). Design and fabrication of an autonomous firefighting robot with multisensor fire detection using PID controller, In 2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), IEEE, 909–914.
Reyana, A., & Vijayalakshmi, P. (2021). Multisensor Data Fusion Technique For Environmental Awareness In Wireless Sensor Networks, European Journal of Molecular & Clinical Medicine, 7(08), 2021.
Seiti, H., & Hafezalkotob, A. (2018). Developing pessimistic–optimistic risk-based methods for multi-sensor fusion: An interval-valued evidence theory approach. Applied soft computing, 72, 609–623
A, S. R., Ofoli, A., R, Krakani, S. N., & Fiawoo, S. Y. (2016). Hardware design and web-based communication modules of a real-time multisensor fire detection and notification system using fuzzy logic. IEEE Transactions on Industry Applications, 53(1), 559–566
ur Rehman, Necsulescu, A., & Sasiadek, J., D. S, &. (2015). Robotic based fire detection in smart manufacturing facilities. IFAC-PapersOnLine, 48(3), 1640–1645
Verma, N., & Singh, D. (2020). Analysis of cost-effective sensors: Data Fusion approach used for Forest Fire Application, Materials today: proceedings, 24, 2283–2289.
Wang, T. C., Xie, Y., Z, & Yan, H. (2016). Research of multi sensor information fusion technology based on extension neural network”. Mathematical Modelling of Engineering Problems, 3(3), 129–134
Hu. J, W. T., Ma, T., & Song, J. (2020). Forest fire detection system based on Fuzzy Kalman filter, In 2020 International Conference on Urban Engineering and Management Science (ICUEMS), IEEE, 630–633.
Yang, C., Mohammadi., A., & Chen, Q. W. (2016). Multi-sensor fusion with interaction multiple model and chi-square test tolerant filter. Sensors, 16(11), 1835
Yooyativong, Y. A., & Chaisricharoen, R. T, & (2017). The wifi multi-sensor network for fire detection mechanism using fuzzy logic with IFTTT process based on cloud, In 2017 14th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), IEEE, 785–789.
Mpimpoudis., Z. E., Anagnostopoulos, A., C, Sekkas, O., & Hadjiefthymiades, S. (2011). Multisensor data fusion for fire detection. Information Fusion, 12(3), 150–159
Ye, Z. J., & Li, K., Z, &. (2020). Multi-sensor information fusion detection system for fire robot through backpropagation neural network. Plos one, 15(7), e0236482
Wang, Z. H., Z, Q. S., Shen., Y., Zhang., L., Tang, K., & Fortino, G. (2018). Heading drift reduction for foot-mounted inertial navigation system via multi-sensor fusion and dual-gait analysis. IEEE Sensors Journal, 19(19), 8514–8521

Download PDF

Editorial decision: Major revisions
07 Oct, 2022
Reviewers agreed at journal
24 Jul, 2022
Reviewers invited by journal
22 Jul, 2022
Editor assigned by journal
20 Sep, 2021
First submitted to journal
18 Sep, 2021

You are reading this latest preprint version

Multisensor Information Fusion and Optimization Technique for Wildfire Detection and Prevention in WSN Environment

Status:

Version 1

Abstract

Figures

1. Introduction

2. Related Works

3. Adkf-dt Multisensor Fustion Technique

4. Performance Evaluation

5. Conclusions

Declarations

References

Status:

Version 1