Project Report
32118 Wireless Communications
Project Report
(Limited to 3000 to 4500 words and 8 figures plus tables, exclusive of Reference and Appendix.)
Executive Summary
Team Name: Group6
Name of Product/Solution
Advancing Bushfire Detection with Sensor and Drone Technology
Please give a brief description of your Product/Solution
Our solution integrates drones, sensors, and AI algorithms to revolutionise bushfire detection in Australia. Our system enables rapid detection of potential fire activity by deploying drones equipped with sensors and cameras for real-time monitoring of remote areas, coupled with strategically placed sensor networks to monitor environmental conditions. Advanced AI algorithms analyse data from sensors and drones to determine the presence of a bushfire while leveraging existing cellular networks and satellite communication technology to ensure seamless data transmission for swift response.
Team Members (Name) University ID Percentage of Contribution to this Project (%) Electronic Signature
Takuto Itami
24595326
Chunkit Chan
12844357
Akshay Bhumkar
14332555
1. Aims and Background
1.1 Originality and Importance (4 marks) 237
Bushfires significantly threaten communities and ecosystems, particularly in regions like Sydney, NSW, and across Australia. The devastating 2019-2020 bushfire season underscored the urgent need for improved detection and response systems. According to the NSW State of the Environment report (2021), this calamitous period witnessed unprecedented destruction, with over 5 million hectares of land consumed, 26 lives lost, and 2,476 homes destroyed across the state. Beyond the human toll, these infernos inflicted irreparable damage to natural habitats and wildlife.
Given the escalating risks exacerbated by climate change, predicting and mitigating future bushfire events is paramount. Enhancing the accuracy and speed of detection mechanisms is a critical step towards proactive intervention and minimizing the devastation caused by such disasters.
Our proposed solution leverages a synergistic blend of cutting-edge technologies, including drones, Internet of Things (IoT), and Artificial Intelligence (AI). While each of these components exists independently, their integration into a unified bushfire detection system represents a novel approach to tackling this pressing challenge. Moreover, our system incorporates advanced wireless communication technologies, utilizing both cellular networks and satellite communication, to ensure seamless data transmission and coverage across diverse terrains.
By harnessing the collective power of these technologies and implementing novel design configurations, our solution aims to revolutionize bushfire detection, extending the reach and precision of early warning systems. Subsequent sections will delve into the intricate details of our innovative approach, elucidating our integrated system's architecture, functionalities, and potential impact.
1.2 Literature Review (5 marks) 922
Traditional Methods vs. Technological Innovations
Although Diaz (2023) states that most bushfires are currently detected from calls by the public, according to her report and NSW Climate and Energy Action (2024), various technologies such as drones and smart cameras have recently started to be introduced for bushfire detection systems experimentally or in certain areas. Since, this project aims to select the best of technologies and propose a better detecting system by incorporating them into appropriate wireless communication design, in this section, we will first compare the various technologies presented in academic papers and identify their advantages and disadvantages.
Drone-Based Detection Systems
The first solution is a drone-based system proposed by Reddy et al. (2023). It was developed by the authors who saw the problem of frequent bushfires in Australia in 2019, when a lot of damage was caused by the lack of prompt sharing of accurate information. This system uses drones equipped with sensors, camera, and GPS to collect information and notify the public of danger, as the diagram below shows. It is explained that the current mainstream detection method is to use a digital camera to take photos at regular time intervals and comparing them to count pixels containing smoke or flames, or to use optical sensors to detect smoke or flames, and if those are detected, the photos are taken by the digital camera and sent to fire departments with its location. In comparison, the use of drones is superior in terms of being able to respond quickly and accurately and not having to build high towers for the cameras to have a field of view. However, it is reported that this information is sent via Wi-Fi, Bluetooth and mobile phone networks, and there are some limitations in that it can only be used in areas where these communication infrastructures are in place.
Figure 1: The Logical Sequence of the System (Reddy et al., 2023)
Sensor-Based and AI-Enabled Detection Systems
The second solution is to install sensors and cameras and use AI to determine if there is a bushfire, which is suggested by Avazov et al. (2023). This is a wildfire detection and notification system proposed in South Korea. The system detects flames by installing a module with a flammable gas and smoke sensor and a camera on cellphone towers installed in the country. There are three steps in the wildfire detection and notification process, starting with the detection of flames using the sensors mentioned above. When flames are detected by the sensor, the camera is activated and takes a photo, which is then processed by the AI. If a bushfire is detected by the AI, the system sends an email notification to fire departments. Like the first solution, this system is also said to send data via Wi-Fi and mobile phone networks, and it is considered to have limitations in that it can only be used in areas where these communication infrastructures are in place. It is also reported that false alarms may occur if the AI is not sufficiently trained.
Sensor-Based Warning Systems for Smart Homes
The last solution is to use sensors to record temperature, humidity, wind strength and direction and share them with the host using Zigbee, which is proposed by Sun et al. (2010). This system is more of a bushfire warning system than a detection system. The system deploys a number of sensors that measure temperature and humidity, called slaves, as shown in the figure below. The information recorded by the slaves is shared at regular intervals via Zigbee to the master, which records the temperature and humidity as well as wind strength, its direction, and the location using GPS. When the recorded values exceed pre-defined thresholds, e.g. high temperatures or low humidity, the time interval at which the record is shared is shortened, allowing the user of the system to get more accurate information and be more alert. The information gathered using Zigbee is also sent to the control room together with GPS location information, but no mention is made of the wireless communication technology used for this long-range communication. The system is good in that it is low-cost compared to other systems and can be easily implemented on an individual basis, but Zigbee's range is limited to a maximum of 90 meters, making it unsuitable for the system covering forested areas across the entire country, which is our aim in this project.
Figure 2: Application of Sensor-Based Warning Systems for Smart Homes (Sun et al., 2010)
Integration and Hybrid Solutions
In conclusion, bushfire warning and detection systems using technology, such as sensors, cameras and drones, already exist, even if it is only experimental or proposed, which are not pervasive across the whole of the country. However, many of the wireless communication technologies used to share the measured data can only be used where there is a wireless communications infrastructure such as cellular networks and Wi-Fi. Using such wireless communication technology to share the information would be the most cost-effective way, if we are considering a system to save only human lives, because it is only necessary to detect bushfires in areas where people live in large numbers, where there is usually no need to worry about wireless communication infrastructure. However, there are environments and ecosystems that need to be protected in areas that are not inhabited and do not have such infrastructure, for example mountainous areas, so it may be necessary to consider a system that combines multiple wireless communications to create a system that can cover such areas as well.
2 Technical details (10 marks)
In this comprehensive exploration of our proposed bushfire detection solution, we delve into the intricate technical aspects, elucidating the methodology for detection and the wireless communication technologies crucial for data transmission.
2.1 Bushfire Detection Methodology
Our innovative approach amalgamates cutting-edge technologies such as drone-based systems, sensors, and AI algorithms to discern and counter bushfire threats across Australia efficiently. The detection process unfolds with meticulous precision:
Detection Mechanism:
Sensor Deployment: A network of sensors is strategically deployed across bushfire-prone areas, meticulously calibrated to monitor crucial environmental parameters. These parameters include temperature, humidity, and smoke levels, serving as early indicators of potential bushfire activity.
Sensor Alert System: Upon detecting anomalous environmental readings indicative of a potential bushfire, such as a sudden spike in temperature or the presence of smoke particles, the sensors trigger the activation of nearby cameras. This activation initiates a process of visual confirmation to validate the occurrence of fire activity.
Visual Confirmation: Cameras stationed at strategically located base stations promptly capture high-resolution images of the surrounding terrain. These images are directed towards the direction indicated by the triggered sensors, providing tangible visual evidence of potential fire activity. This visual confirmation serves as a crucial validation step in the detection process.
Data Transmission: Wireless sensors transmit collected data to a central gateway, which serves as an intermediary hub. This gateway aggregates the sensor data and then relays it to an internet-based database through secure and reliable network connections. Once the data reaches the database, it undergoes comprehensive management and analysis processes. These processes involve organising the data, running diagnostic checks, and applying analytical algorithms to extract valuable insights. This robust transmission and analysis framework ensures that the information is both accessible and actionable for emergency response teams and decision-makers, enabling timely and informed responses to potential bushfire threats.
AI Analysis: The captured images are swiftly transmitted to an AI-driven analysis system via high-speed 4G/5G networks for rigorous examination. Leveraging sophisticated machine learning algorithms, the AI system meticulously scrutinises the images to ascertain whether the detected anomaly indeed constitutes a genuine bushfire or a false alarm. This automated analysis streamlines decision-making and ensures swift and accurate responses to potential threats.
Human Intervention: In scenarios where the AI system encounters uncertainty in discerning the presence of a bushfire, the images are routed to human operators for further evaluation and decision-making. Human intervention adds an additional layer of validation, ensuring the accuracy and reliability of the detection process.
Drone Deployment:
According to Figure 3, the 3G/4G/5G coverage in Australia is very limited due to Australia's vast geography and relatively low population density present challenges for telecommunications infrastructure deployment. So, we will use satellite communication for data transmission for our drone.
Figure 3: 3G/4G/5G coverage in Australia
Satellite Communication: In regions devoid of cellular network coverage or base stations, drones equipped with advanced sensors, cameras, and GPS modules are dispatched to gather real-time data on potential bushfires. These drones serve as agile and versatile aerial platforms, capable of swiftly maneuvering across vast and inaccessible terrain.
Immediate Data Transmission: Drones eschew the need to return to base stations for data transmission by leveraging satellite communication technology. Captured images and GPS coordinates are transmitted directly to Low Earth Orbit (LEO) satellites, ensuring rapid dissemination of critical information. This immediate data transmission enables swift coordination of emergency response efforts and facilitates more efficient allocation of resources for bushfire containment and management.
Enhanced Response Time: By circumventing the reliance on ground-based infrastructure, drones facilitate expedited response times and enable more judicious allocation of resources for bushfire containment and management. Their agile and nimble nature allows for swift deployment to remote and inaccessible areas, enhancing the effectiveness of response efforts.
Data Analysis and Response:
AI Decision Making: A sophisticated AI system meticulously analyses data emanating from sensors and drones to ascertain the presence of a bushfire. Leveraging advanced machine learning algorithms, the AI system can accurately discern patterns and anomalies indicative of fire activity. In instances of uncertainty, human operators intervene, leveraging their expertise to make well-informed decisions based on visual evidence and additional contextual information.
2.2 Wireless Communication Technology
Efficient wireless communication technologies serve as the backbone of our bushfire detection solution, ensuring seamless data transmission across the system:
Sensor Communication:
Cellular Networks: Sensors leverage existing 4G/5G cellular communication infrastructure for data transmission. Triggered sensors seamlessly communicate with nearby cameras and transmit captured images to the central control centre via cellular networks. This utilisation of cellular networks ensures reliable and high-speed data transmission, facilitating swift analysis and response to potential threats.
Narrowband IoT (NB-IoT): NB-IoT is an efficient, low-power wide-area (LPWA) technology that supports long battery life and wide coverage areas, making it ideal for connecting sensors in remote locations. It ensures reliable communication with low data rates, which is sufficient for transmitting environmental readings.
Drone Communication:
LEO Satellite Communication: Drones leverage satellite communication technology to transmit data directly to LEO satellites. This eliminates the need for ground-based infrastructure and facilitates seamless data transmission from remote or inaccessible regions. By bypassing terrestrial limitations, satellite communication enables real-time monitoring of vast and remote areas, enhancing the system's coverage and response capabilities.
By synergising sensor networks, drones, and AI-driven analysis with efficient wireless communication technologies, our solution endeavours to redefine bushfire detection and response paradigms. With a comprehensive understanding of the technical intricacies involved, we aim to bolster the resilience of communities and ecosystems against the ominous spectre of wildfires.
Wireless sensor with a central gateway to send data to the internet(database)
Zigbee20-250 kb/s10-20 m
NB-IoT200kb/s1-10km
Wifi1.3 Gb/s50-70 m
LoRaWAN1kb/s10 km
3 Simulation Results (6 marks)
[Whether the Matlab simulation results provide sufficient support for the performance/correctness of the proposed wireless systems/networks?]
Assuming that cellular communications and Wi-Fi have been determined to work without problems in previous experiments and reports, can satellite communications from the drone be used to ensure sufficient speed and data volume? (raining, latency, etc.) ????????????????
Key (what result we want?) photo(HD?)?video? distance, power.
4 Benefits (3 marks) 416
In light of the challenges posed by traditional bushfire detection systems and the limitations of existing technological solutions, our proposed bushfire detection system offers significant benefits that could mitigate the devastating impact of bushfires in Australia. Here's an exploration of the benefits:
4.1 Reduction in Bushfire Damage
Traditional bushfire detection systems have often failed to provide timely and accurate information, leading to delays in response and increased damage to life and property. Conversely, our integrated system, leveraging drones, sensors, and AI, enables early detection and rapid response, potentially minimizing the extent of bushfire damage across a larger area of Australia. By swiftly identifying and confirming bushfire incidents, our system empowers authorities to take proactive measures to protect lives, property, and the environment, such as evacuations and firefighting efforts.
4.2 Enhanced Coverage and Response
One of the key advantages of our system is its ability to extend coverage to areas that were previously underserved by traditional detection methods. By leveraging drones equipped with sensors and cameras, we can overcome the limitations imposed by inadequate wireless communication infrastructure. This enables us to monitor remote and inaccessible regions, providing early warning and facilitating rapid response even in areas lacking ground-based infrastructure.
4.3 Cost-Effective Implementation
While the immediate practical application of our system may pose challenges, particularly concerning AI reliability and drone availability, the feasibility of our approach is rooted in cost-effective wireless communication mechanisms. By leveraging existing cellular networks and strategically supplementing coverage with satellite communication, we minimize the need for costly infrastructure upgrades. Furthermore, we can reduce installation costs and expedite deployment by integrating modular components with existing structures such as cell phone towers.
4.4 Future Potential and Adaptability
Despite current challenges, we believe that our system holds significant promise for future implementation. As AI technologies continue to advance and become more reliable, the incidence of false alarms can be mitigated, enhancing the overall effectiveness of our detection system. Similarly, increasing the availability and affordability of drones will facilitate broader adoption and implementation of our solution. With ongoing technological advancements and refinement, our system stands poised to become a critical tool in the fight against bushfires, offering adaptability and scalability to address evolving threats.
In conclusion, while there are challenges to overcome, the benefits of our proposed bushfire detection system are substantial. By harnessing the power of technology and cost-effective wireless communication mechanisms, we can enhance early detection, improve response times, and ultimately reduce the impact of bushfires on communities and ecosystems across Australia.
5 Conclusions 267
In conclusion, the development of effective fire detection systems using sensors and drones represents a critical advancement in addressing the escalating threat of bushfires, particularly in regions like Sydney, NSW, and across Australia. Through the integration of innovative technologies such as drones, the Internet of Things (IoT), and Artificial Intelligence (AI), our proposed solution offers a promising approach to early detection and rapid response to bushfire incidents.
The literature review highlighted the limitations of traditional detection methods and the potential of emerging technologies to overcome these challenges. By leveraging sensor networks, drones equipped with advanced sensors and cameras, and AI algorithms for data analysis, our system enables timely detection of bushfires, thereby facilitating proactive intervention and minimising the devastating impact on communities and ecosystems.
Furthermore, the utilisation of wireless communication technologies, including cellular networks and satellite communication, enhances the system's coverage and response capabilities, extending its reach to remote and inaccessible areas. While challenges remain, such as ensuring the reliability of AI algorithms and addressing the cost implications of drone deployment, the feasibility of our solution is underpinned by its cost-effective wireless communication mechanisms and the potential for future advancements in technology.
In summary, our proposed fire detection system stands as a promising solution for mitigating the impact of bushfires in Australia. By combining cutting-edge technologies and cost-effective wireless communication solutions, we aim to reduce bushfire damage, enhance coverage and response times, and ultimately safeguard lives, property, and the environment. With continued research, development, and collaboration, we are optimistic about the potential of our system to make a significant difference in bushfire management and resilience efforts nationwide.
6 Reference
Avazov, K., Hyun, A. E., Sami S, A. A., Khaitov, A., Abdusalomov, A. B., & Cho, Y. I. (2023). Forest Fire Detection and Notification Method Based on AI and IoT Approaches. Future Internet, 15(2), 61-. https://doi.org/10.3390/fi15020061
Diaz, A. (2023, January 17). Fighting fires from space: how satellites and other tech could prevent catastrophic bushfires. ANU reporter. https://reporter.anu.edu.au/all-stories/fighting-fires-from-space-how-satellites-and-other-tech-could-prevent-catastrophic-bushfires
NSW Climate and Energy Action. (2024, January 16). Investing in technology for rapid bushfire detection, response and conservation in Kosciuszko National Park. NSW Climate and Energy Action. https://www.energy.nsw.gov.au/nsw-plans-and-progress/government-strategies-and-frameworks/taking-action-climate-change/fire-management
NSW State of Environment. (2021). Fire. NSW State of Environment. https://www.soe.epa.nsw.gov.au/all-themes/land/fire
Reddy, V. S., Prasad, V. K., Wang, J., & Reddy, K. T. V. (2023). Smart Cities Implementation: Australian Bushfire Disaster Detection System Using IoT Innovation. In Soft Computing and Signal Processing (Vol. 313, pp. 421437). Springer. https://doi.org/10.1007/978-981-19-8669-7_37
Sun, Y., Yuan, Q., & Al-Sarawi, S. (2010). A bushfire monitoring and detection system for smart homes using ZigBee technology. International Journal of Computer Aided Engineering and Technology, 2(23), 234249. https://doi.org/10.1504/IJCAET.2010.030547
7 Appendix
Additional detail such as Software codes, pictures of additional implementation results, etc.
