.jpg)
1. Introduction
As the possibility of nuclear armament in neighboring countries continues to grow, it becomes crucial to detect undeclared nuclear activities accurately in hidden facilities or areas. The technical detection of areas suspected of nuclear activity is essential in the search for undeclared nuclear activity in states that express a desire for denuclearization [1]. Nuclear activity detection is primarily based on undeclared areas, referring to those not reported by the targeted states [1]. During the detection process, wide-area environmental sampling (WAES) plays an important role in tracing nuclear activities [2, 3]. WAES is an environmental sample collection process to draw conclusions about the absence of undeclared nuclear activities over a wide area [4]. During the WAES process, which involves collecting and transporting samples, the precision of sample-based analysis can decrease due to insufficient environmental information, such as radiation levels and geographic data. Also, there may be distractions such as intentional sample extortion and tampering by the targeted state. To minimize these problems, it is essential to maintain Continuity of Knowledge (CoK) when monitoring all steps, from sampling, storage and transport to the laboratory [5]. CoK is valuable for ensuring compliance with the designated procedures of WAES, making it possible to trace the state of samples [6]. From this perspective, on-site information such as locations and dose rate may be required to maintain CoK during the WAES process. On-site information can be utilized in various ways to track the status of workers or samples, depending on their characteristics.
For example, location information is utilized to track the movements of workers and samples, enabling rapid analyses of routes and establishing measures if emergencies arise. Dose rate controls working times and establishes appropriate radiation protection measures. This information can also be used to assess extortion or damage to samples by monitoring changes in dose rate. In addition, environmental information such as gyroscope data and brightness levels can be used to determine the sample extortion or tampering.
Extensive work has been done to develop sensor technologies for collecting on-site information. Still, the current technologies generally require individual systems to collect each type of information or function. Therefore, this study developed an ‘environmental and geographical data transfer (EGDT)’ device to integrate and transfer on-site information such as locations, dose rate, temperatures, humidity levels, and brightness levels for the WAES process. Furthermore, based on three cases related to the WAES process, referred to here as; ‘Normal Operation’, ‘Sample Extortion’, and ‘Out of Work Area’, simulations of these cases were carried out to evaluate the feasibility of the EGDT.
2. Materials and Methods
2.1 Design and Configuration of the Environmental and Geographical Data Transfer (EGDT)
2.1.1 Types of sensors in EGDT
GPS, radiation, acceleration, lux, temperature, humidity, and proximity sensors are included in the EGDT devices to collect various on-site data.
The GPS sensor in the EGDT device is the PA1616S model (ASCEN GPS Co., Korea). A position accuracy is up to 3 meters. It enables the real-time measurement and transmission of device location to track routes within its software.
The radiation sensor in the EGDT device is the GDK101 model (FTLAB Co., Korea). It has a pin diode gamma sensor, which measures dose rates in units of μSv·h−1. An uncertainty is ±5% up to 100 μSv·h−1, and the measurement periods can be set to either 1 or 10 minutes in the device. These data are used to assess worker’s safety of and establish measures of radiation protection. The device also includes an alarm function ensuring prompt awareness of workers if threshold values are exceeded.
The acceleration sensor in the EGDT device is the LSM6DSOX model (ST Microelectronics Co., Switzerland), which measures data based on a 3D acceleration coordinate system. An uncertainty is ±0.003%. The acceleration data quantify velocity changes, enabling analyses of the movements of both workers and samples.
The lux sensor in the EGDT device is the BH1721FVCTR model (ROHM Semiconductor Co., Japan), which measures data based on 256 levels of light intensity differentiation. An uncertainty is ±15%. With a lux sensor installed inside the sample storage box, changes in the light intensity can indicate whether the storage box has been opened or closed.
The temperature and humidity sensors in the EGDT device are both the SHT41 model (Sensirion AG Co., Switzerland), which measures data in units of Celsius (°C) for the temperature and percentage (%) for the relative humidity. An uncertainty is ±1.8%. Temperature and humidity data are used to assess worker’s safety and track historical changes in the chemical properties of samples during transportation processes.
The proximity sensor in the EGDT device is the VL53L3CXV0DH/1 model (ST Microelectronics Co., Switzerland), which measures distance data by detecting objects in front of the sensors based on an optical measurement method. An uncertainty is ±1.3%. With a proximity sensor installed inside the sample storage box, changes in distances can indicate the presence of an object.
2.1.2 Hardware design
During the WAES process, external signal information such as radiation and lux should not be shielded by the attached state of EGDT device. Therefore, sensors are placed on the front of the device to attach to workers and samples. A display is positioned at the top of the front face of the device, and four LEDs are installed at the bottom, as shown in Fig. 1. The LEDs indicate power, Wi-Fi, a GPS connection, and emergency status through toggling or flashing to enhance visibility. Also, the power button and a USB charging port (Type-C) are located at the bottom of the device enabling both power input and firmware uploads.
2.1.3 Firmware design
The firmware for the operation scenario of EGDT was designed as shown in Fig. 2. The firmware consists of three main components: the monitoring display, the menu display, and a detailed submenu display.
The monitoring display shows real-time on-site data as measured by the device. As shown in Fig. 3, real-time measurement data from each sensor and device status, such as the current time, battery level, and server connectivity, are displayed.
On the menu display, users can modify certain device settings. This menu includes an alarm, a data-reset function, and function activation settings, as shown in Fig. 4.
The ‘Alarm Setting’ menu allows users to check and modify the threshold values of the dose rate, which determines the alarm operation. Users can set threshold values for Low (low-risk) and High (high-risk) alarms. Alarms are activated based on these values.
One of the two operating modes can be selected based on intended use when powering on the device.
The normal mode is designed for monitoring workers during the sample collection step, with all functions activated upon entering the monitoring display. Currently, the device’s backup period is fixed at 10 seconds.
The low power mode is designed to monitor samples as they are transported. To minimize power consumption for long-term operation, the display turns off after 30 seconds, and a more extended backup period can be set before starting the operation.
2.1.4 Software design
The data communication of the EGDT device is configured with a TCP/IP-based server system, allowing data measured by each device to be transmitted to a central server via a mesh Wi-Fi setup. Based on this, software was developed for each device’s independent registration and real-time monitoring. The main interface of the developed software program is shown in Fig. 5.
Data collection via mesh Wi-Fi is based on a common IP address shared among the server, Wi-Fi system, and the EGDT device. Therefore, server connectivity is activated by entering the corresponding IP address into the ‘Server Information’ section within the software. Each device is listed by registering it in the ‘Device Information’ section, and the EGDT software receives the measurement data from the devices individually. The received data are displayed on the center of the software interface with graphs for each sensor, along with the movement path from the GPS data shown on a map on the right side.
If the device’s server connectivity is disabled during the data transfer, the software can recover lost data once the server is restored. Furthermore, the EGDT software enables the user to extract stored data in the Excel format while offering remote control of device functions, such as the data reset and alarm functions in an emergency.
2.2 Feasibility Evaluation Methodology of the EGDT Based on WAES Cases
2.2.1 EGDT usage types
There are two main usage types when the EGDT device is used for the actual WAES process.
First, EGDT can be utilized to monitor the safety of workers in real-time and promptly establish countermeasures in emergencies. The normal mode can be selected during the initial operation of the EGDT device, and the device can be attached to the worker’s uniform, as shown in Fig. 6(a). The attached EGDT measures the dose rate for emergency alarms and measures acceleration data to analyze the worker’s movement.
Secondly, EGDT can be utilized to assess sample conditions and analyze the causes of emergencies. The low power mode can be selected during the initial operation of the EGDT device, and the device can be affixed inside the sample storage box, as shown in Fig. 6(b). The attached EGDT measures the sample’s dose rate, temperature, and humidity to analyze their chemical condition and determine the extent of any damage. Additionally, EGDT complies with lux and distance data to analyze more specific situations, such as sample extortion.
2.2.2 Configuration of cases for a feasibility evaluation
To use the EGDT device in an actual WAES process, several cases related with WAES were simulated to evaluate the feasibility of the EGDT. Cases were categorized into normal types and emergency types, as shown in Fig. 7. In detail, normal types include cases of ‘Normal Operation’, and emergency types include cases of intentional ‘Sample Extortion’ and ‘Out of Work Area’ due to high radiation.
The Normal Operation case refers to storing and transporting collected samples using a vehicle. The detailed process consists of ‘Device Installation and Standby’, ‘Sample Storage’, and ‘Transportation’.
The Sample Extortion case arises during the retrieval of samples in response to intentional extortion during the sample transportation step. The detailed process consists of ‘Normal Operation’, ‘Sample Extortion’, ‘Standby’, ‘Recovery’, and ‘Normal Operation’.
The Out of Work Area case refers to an escape from an area due to unexpected high-radiation alarms during the sample collection step. The detailed process consists of ‘Normal Collection’, ‘High-Radiation Alarm’, and ‘Out of Work Area’ stages.
2.2.3 Inference of events based on data tendency
To analyze cases using various types of data as measured by the EGDT device, unit events were inferred from the data of individual sensors. The data measurement period for each type of sensor is 30 seconds, with transfers of the average values of accumulated data during this period. To enhance the visibility of the measurement results, all types of on-site data (excluding acceleration data) were normalized using the following formula.
The maximum and minimum values are measured in a three-dimensional coordinate system every 30 seconds for acceleration data. In this study, the acceleration sensor only assesses the subject’s movement. Hence, the acceleration data was transformed into displacement values using the following formula.
The meanings of the data tendencies and unit events for each sensor are derived using the normalized and transformed on-site data, as shown in Table 1. Based on the tendencies of the individual sensor data and inferred unit events, complex situations can be inferred by correlating data from multiple sensors.
Table 1
Correlation of sensor data
| Sensor | Data | State | Event for sample | Event for worker |
|---|---|---|---|---|
|
|
||||
| Radiation | High | High dose | ||
| Low | Low dose | |||
|
|
||||
| Acceleration | 0 | No change | ||
| ≠0 | Change | |||
|
|
||||
| Lux | High | Bright | ||
| Low | Dark | |||
|
|
||||
| Temperature/Humidity | Increase | High amount | ||
| Decrease | Low amount | |||
|
|
||||
| Proximity | High | Far | ||
| Low | Close | |||
An acceleration data value reading of 0 indicates that the object is either stationary or moving at a constant velocity. However, because movement at a constant velocity is impossible, a 0 value of the acceleration data is interpreted as the stationary state.
The situation of opening the storage box to load samples can be inferred from the combination of zero acceleration data and increasing lux and dose rates. Light increases as the storage box opens, and the dose increases as samples are loaded near the sensor.
The situation of extorting samples by opening the storage box can be inferred from the combination of a zero value for the acceleration data, increasing lux, and decreasing dose rates. Light increases as the storage box opens, and the dose decreases as samples are extorted near the sensor.
The situation of transporting samples can be inferred from non-zero values of the acceleration data, low lux, and high dose rates. Light is low because the storage box is closed after loading samples, and the dose remains high due to samples near the sensor. Additionally, the acceleration data fluctuates with non-zero values due to the movement caused by transportation.
The situation of worker sampling in a safe area can be inferred from non-zero acceleration data values along with high lux and low dose rates. Due to the low radiological risk in outdoor work areas, the lux level is high, and the dose level is low. Also, the acceleration data fluctuates with non-zero values due to movement during the sampling step.
Distance data from proximity sensors can be used for various purposes depending on the positioning of the EGDT device inside the sample storage box. When the device is positioned on the inner wall of the storage box, the proximity sensor recognizes samples in front of the device. Accordingly, the distance data can be used to infer the presence of samples. On the other hand, when the device is positioned on the bottom surface of the storage box, the proximity sensor recognizes the upper part of the box in front of the device. In this case, the distance data can be used to infer the opening and closing of the storage box. In this study, the EGDT device was positioned on the bottom surface of the sample storage box as shown in Fig. 6(b). Hence, the distance data were secondary to the lux data in inferring the opening and closing of the storage box.
2.2.4 Procedures of the feasibility evaluation based on cases
To evaluate the feasibility of the EGDT based on organized cases, each case was simulated under similar environmental conditions and with similar procedures. Samples were replaced with 137Cs sources of 35 and 83 μCi. EGDT devices were attached inside the simulated sample storage box and on the worker’s uniform, as shown in Fig. 6.
In the Normal Operation case, the evaluation was conducted by opening the sample storage box inside the vehicle, loading the sources, closing the storage box, and moving a certain distance.
In the Sample Extortion case, extortion was simulated by stopping the vehicle while transporting the source-containing storage box and removing the sources from the storage box. After a moment, the sources were reloaded into the storage box, and the vehicle was driven to simulate a recovery after the extortion event.
In the Sample Extortion case, extortion was simulated by stopping the vehicle while transporting the source-containing storage box and removing the sources from the storage box. After a moment, the sources were reloaded into the storage box, and the vehicle was driven to simulate a recovery after the extortion event.
In the Out of Work Area case, the worker moved to the sources to simulate a high-radiation environment. Then, the worker quickly moved away from the source to simulate an escape from the site.
The EGDT device collected the data during the three simulations, and the data trends were analyzed to evaluate the performance of the EGDT device during the simulations.
3. Results and Discussion
A feasibility evaluation of the EGDT was conducted based on the three cases. Temperature and humidity data were not included in the evaluation, because these data are used to track the sample’s chemical behavior during transportation.
3.1 Analysis of Results Based on the Normal Operation Case
The Normal Operation case was simulated by the procedure described in section 2.2.4. EGDT devices was attached inside the sample storage box during this process. Fig. 8 shows the collected data in the simulation of Normal Operation case. All data except for the acceleration is normalized.
Fig. 8
The monitoring of lux, distance, dose rate and acceleration using EGDT in assessing ‘Normal Operation’ case. (a) A closed storage box without any samples in a stationary state, (b) An opened storage box containing samples, (c) A closed storage box containing samples in transit.
Based on the inference of events in section 2.2.3, Fig. 8(a) suggests a closed sample storage box due to low lux and distance values. Also, the dose rate and acceleration values which converge to zero suggests that the storage box is stationary without any samples inside. Therefore, Fig. 8(a) is inferred that the closed sample storage box is waiting for samples to be loaded.
Fig. 8(b) suggests the opening process of the sample storage box by increased lux and distance data. Additionally, the gradual increase in the dose rate suggests the event of loading samples into the storage box. During this step, the dose rate increases after the lux and distance increase. This tendency indicates that the storage box was first opened, then loaded with samples. Also, the acceleration shows non-zero values before the dose rate increases. It is the movement to open the storage box before sample loading. The distance shows a rapid decrease and increase before the dose rate increases. This phenomenon is inferred that the worker’s body or equipment passes the EGDT devices during the sample loading process. In the final segment shown in Fig. 8(b), there is a sharp decrease in the lux and distance, but the dose rate tends to remain high. This suggests the closing process of the storage box after sample loading.
Fig. 8(c) suggests a closed storage box containing samples by the low lux and distance, and the high dose rate. Also, movement of storage box can be inferred from the continuous changes of the acceleration. Compared to the acceleration data in Fig. 8(b), Fig. 8(c) shows larger variations. This indicates a greater change in the speed, suggesting sample transportation rather than the opening of the storage box.
3.2 Analysis of Results Based on the Sample Extortion Case
The Sample Extortion case was also simulated according to the procedures described in section 2.2.4. EGDT devices was attached inside the sample storage box during this process. Fig. 9 shows the collected data in the simulation of Sample Extortion case. All data except for the acceleration is normalized.
Fig. 9
The monitoring of lux, distance, dose rate and acceleration using EGDT in assessing ‘Sample Extortion’ case. (a) A closed storage box containing samples in transit, (b) An opened storage box where samples are being extorted, (c) An opened storage box without any samples in a stationary state, (d) A closed storage box containing recovered samples in a stationary state, (e) A closed storage box containing samples in transit.
Based on the inference of events in section 2.2.3, Fig. 9(a) shows low lux and distance, a high dose rate, and continuous fluctuations in the acceleration. This pattern suggests a process similar to the sample transportation step, as shown in Fig. 8(c).
In Fig. 9(b), the acceleration value which converges to zero suggests that the storage box has stopped. Furthermore, opening the storage box can be inferred from the rapid increase in the lux and distance, and the removal of samples can be inferred from the decrease in the dose rate. Based on the events derived in this way, two situations can be considered. The first is where the worker directly removes the samples during transportation, and the second is where the samples are extorted by a third party. However, it is unlikely that the worker would remove the samples during transportation without additional reporting. Therefore, this can be inferred as a sample extortion by a third party. There are several decreases and increases in the distance, similar to the sample loading process in Fig. 8(b). This tendency suggests the body or equipment of extorter passing in front of the device during the process.
Fig. 9(c) suggests an opened sample storage box due to the high lux and distance. Furthermore, as the dose rate and acceleration converge to zero, it can be inferred that the storage box is stationary and does not contain any samples. This situation is derived to consider data both before and after this step. Accordingly, Fig. 9(c) can be inferred as the process of stopping transport and considering a countermeasure after the recognition of sample extortion.
Fig. 9(d) suggests the event of sample loading due to increasing dose rate. The zero value of the acceleration indicates that the storage box is stationary. Furthermore, the sharp decrease in the lux and distance can infer the process that the storage box is being closed. For this situation as well, it is required to consider data both before and after this step. Accordingly, Fig. 9(d) can be inferred as the retrieval of extorted sample and its storage back into the storage box.
Based on the similar trend to the normal transportation process in Fig. 8(a), Fig. 9(e) can be inferred as the process of re-transporting the recovered sample.
3.3 Analysis of Results Based on the Out of Work Area Case
The Out of Work Area case was also simulated according to the procedures described in section 2.2.4. EGDT device was attached to a worker during this process. Fig. 10 shows the collected data in the simulation of Out of Work Area case. In this case, distance and lux were excluded due to their unsuitability for monitoring workers. All data except for the acceleration is normalized.
Fig. 10
The monitoring of dose rate and acceleration using EGDT in assessing ‘Out of Work Area’ case. (a) Movement for sample collection without high radiation, (b) High radiation generation and recognition, (c) Movement to areas without high radiation.
In Fig. 10(a), the low dose rate suggests a safe environment, and the non-zero values of acceleration suggest the movement of workers. These data can be inferred as the worker’s general sampling process in a low-radiation area.
Fig. 10(b) shows a sharp increase in the dose rate. This tendency suggests an exposure to a high-radiation environment. Furthermore, the zero value of the acceleration indicates that there is no movement of workers. These data suggests that workers are exposed to high radiation environment, and they stop to sampling process to make countermeasures.
Fig. 10(c) shows a gradual decrease in dose rate. This tendency suggests a transition to a low-radiation environment. Also, this section shows relatively more variation in acceleration than the normal sample collection process in Fig. 10(a). It indicates that the movements during this section are faster than those during the sampling. Consequently, the worker’s rapid departure from the high-radiation environment can be inferred from this tendency in the dose rate and the acceleration.
4. Conclusion
In this study, EGDT was developed to track the status of samples and workers for WAES in undeclared areas. The EGDT’s feasibility was evaluated by organizing and simulating cases related to the WAES process. The developed EGDT device can measure various on-site information using its GPS, radiation, acceleration, brightness, temperature, humidity, and proximity sensors. Also, the functions of real-time monitoring and emergency alarm for high-radiation were developed as firmware features. The measurement data from each device can be compiled into central control software, enabling a comprehensive analysis. To evaluate the feasibility of EGDT, several events that can be inferred by combining each sensor data had been initially derived. Next, three cases related to WAES were classified and outlined.
Cases were simulated to evaluate the EGDT, and the measured on-site data showed significant tendencies to infer each case. Unit actions were inferred from data from various sensors, and detailed situations and overall cases were inferred from a complex analysis of the unit actions. For fluctuations in distance and acceleration found in certain section, their factors were analyzed by comparing data tendencies both before and after each section.
In addition to the data analysis methods used in this study, the GPS and other technologies can be used to increase the inference accuracy. Also, the target of the proximity sensors can be changed by the position of the EGDT device within the sample storage box. It allows the user to diversify the interpretation of distance data.
These results can enhance the safety of WAES process in undeclared areas. Furthermore, it is anticipated to help secure the completeness of nuclear activity detection.
