Design and Implementation of Secure System for IoT-Based Animal Health Monitoring with Adaptive Security

In the quiet corners of modern agriculture, a quiet revolution is taking place, driven not by tractors or plows, but by tiny computers attached to the bodies of livestock. Farmers are increasingly turning to the Internet of Things, a network where everyday objects connect to the internet to share information. For cattle, this means wearable sensors that constantly monitor vital signs like heart rate, body temperature, and movement. These devices act as a continuous health checkup, allowing farmers to spot illness days before a cow shows visible symptoms, shifting animal care from a reactive scramble to a proactive strategy. However, this constant stream of sensitive data traveling over wireless networks creates a new vulnerability. Just as a house with an open door is easy to rob, an unprotected data stream is easy for hackers to intercept, alter, or steal. The challenge for engineers has been to build a system that is secure enough to protect this private information but light enough to run on small, battery-powered devices without draining their energy.

A team of researchers from universities across Pakistan has tackled this problem by designing a flexible security system for monitoring animal health. They built a complete platform where sensors on cows send data to a central cloud server, which then analyzes the information to predict potential diseases. What makes their work distinct is that they did not settle for a single, rigid method of protection. Instead, they created a system where the user can choose the level of security needed for each specific device, right from a web dashboard. This approach acknowledges that not every situation requires the same level of defense; sometimes speed is paramount, while other times, protecting against future, unknown threats is more important. The researchers tested four different ways to secure the data: sending it in plain text, wrapping it in a standard digital lock, encrypting the message itself, and using a new, future-proof method designed to withstand attacks from powerful quantum computers that do not yet exist.

The system operates by attaching a small, low-cost computer to a cow, which gathers data from sensors measuring heart rate, temperature, oxygen levels, and movement. This information is packaged into a message and sent over a wireless network to a central hub. In a traditional setup, this message might travel unprotected or with a single type of encryption. In this new design, a farmer or veterinarian can log into a website and decide how that specific cow's data should be protected. They can choose to send it without any encryption for maximum speed, use a standard secure connection known as TLS, apply a lightweight encryption method called AES-128 directly to the message, or sign the message with a complex digital signature based on post-quantum cryptography. This choice is not just a setting; it changes how the device behaves in real-time, reconfiguring the connection instantly to match the selected security mode.

To see how these different choices affected the system, the researchers set up a controlled experiment using virtual computers to simulate the cows and the network. They measured how long it took for a message to travel from the sender to the receiver, how much memory the system used, and how hard the computer processors had to work. When they sent data without any encryption, the system was incredibly fast and used very little power, but the information was completely exposed to anyone listening in. When they used the standard TLS security, the data became much safer, but the system had to work harder, taking slightly longer to send messages and using more computer memory to manage the secure connection. This is similar to putting a heavy, reinforced lock on a door; it keeps intruders out, but it takes more effort to open and close it every time.

The researchers found that the middle option, using AES-128 encryption applied directly to the message, offered a strong balance. It kept the data secret from the central server itself, ensuring that even if the server was compromised, the specific content of the animal's health data remained unreadable to unauthorized eyes. This method was faster than the standard secure connection and used less memory, making it well-suited for the limited resources of small farm devices. The most advanced option, using post-quantum digital signatures, was designed to protect against a future where hackers might have computers powerful enough to break current security codes. While this method required the most processing power and added a bit more time to the transmission, the system handled it successfully. The computer processors were able to generate and verify these complex signatures without crashing or slowing down the entire network, proving that this future-proof security is feasible even on modest hardware.

The study also looked at how long it took to save this data into a database for later analysis. Whether the data was sent in plain text or protected by the most complex quantum-resistant signatures, the time it took to write the information to the storage system remained very short, ranging from about 40 milliseconds for unencrypted data to 70 milliseconds for the most secure version. These fractions of a second are negligible for a real-time monitoring system, meaning that adding strong security does not create a bottleneck that would stop the system from working smoothly. The results showed that the system could adapt to different needs without breaking. For a farmer who needs immediate, high-speed updates on a large herd, the lighter encryption methods work well. For a situation involving highly sensitive data or a need to prepare for future threats, the heavier, quantum-resistant methods are available and functional.

This work demonstrates that it is possible to build a secure, intelligent system for animal health that is not locked into a single way of operating. By allowing the security level to be changed on the fly, the system can grow and adapt as threats change or as the needs of the farm evolve. The researchers confirmed that while no single method is perfect for every scenario, having the choice allows for a practical trade-off between speed and safety. The system successfully integrated machine learning to analyze the incoming data and predict health issues, proving that security and intelligence can work together. As the threat of quantum computing looms on the horizon, this research provides a blueprint for how agricultural technology can stay ahead, ensuring that the data protecting our food supply remains safe, reliable, and ready for whatever challenges come next.