The devices in your home form complex networks. As suggested by a study, there are 11 connected devices in US households on average. These connections facilitate our convenience. Under the connections, your smart devices can control the lights and curtains in your house; you can remotely control your smart cameras at home... Your convenience results from the collaboration of AI and IoT.
Before we dive into AIoT, you should first know...
AIoT is a combination of AI and IoT technologies. So, to understand AIoT better, we should first have a review on the two fundamentals- AI and IoT. AI (Artificial Intelligence) AI is a field that combines computer science and datasets to develop problem-solving skills in machines. When people talk about AI, they are often referring to its machine learning function. Machine learning is one of the most critical and powerful sub-fields in AI, which focuses on imitating the way humans learn using data and algorithms. IoT (Internet of Things) As its name suggests, IoT is a network of "things". The IoT technologies enable us to link every object, such as cars, household appliances. Besides, it allows people to obtain information from distant devices or even control them remotely. The technology enables seamless communication and interactions between people, processes and things. In other words, IoT is a technology that joins the physical world to the digital world. Let's head back to AIoT Literally, "AIoT" combines and utilise the two buzzing IT technologies- AI (Artificial Intelligence) and IoT (Internet of Things) to generate benefits. In this technology, AI contributes its machine learning capabilities, while IoT contributes its ability to connect and exchange data. To conclude, IoT is the data collector, and AI is the data analyst. People aim to achieve a goal with AIoT infrastructure: extracting helpful information from tons of data, the same as general AI systems. The AI in AIoT can be implemented in the centre of the IoT system or at the edge of the Internet. The selection of locations serves different purposes. The AI at the centre of the IoT system can receive data from various data points. With the huge amount of available data, predictive analytics can be applied to generate accurate future insights. Moreover, the AI can monitor and apply anomaly detection on connected devices. IoT systems are scaled systems. Multiple streams of device data are sent to the AI on the cloud. The task of monitoring the devices in the entire IoT system is only possible with the assistance of AI. At the centre of the IoT system, AI is powerful and capable of performing various tasks to uncover insights for the entire IoT system. However, in some applications, the dependency of edge devices on cloud computing would cause problems- bandwidth limitation and latency. In the case of autonomous cars, the problem may result in fatal consequences. If the control of autonomous cars depends on cloud computing results, rapid response to emergencies cannot be guaranteed. When a pedestrian appears in front of the autonomous car, the sensor needs to send the captured images to the cloud for processing and wait for the response. If the connection between the autonomous car and the cloud server is lost, the autonomous car will not receive proper instructions. It might result in injury or even death, which is undesired. The indicates the need for advanced edge computing ability. Edge computing can minimise the bandwidth requirement and latency for edge devices. Moreover, with AI, edge devices are granted intelligence for better decision making based on real-time and predicted information. The robust computing power is essential in some IoT systems, such as the mentioned autonomous car system. To add with, AI at the edge unleash the possibility of a smart transportation system in which data would be shared between vehicles and even between vehicles and infrastructures. AIoT would also benefit the development of autonomous stores and many more smart systems in different industries. Ultimately, it would make our city "smarter". Data security and privacy risks In every system involving a large amount of data, data security and privacy are usually the major concerns. The issues are tough to solve and contradictory. Inevitably, people's convenience is built on top of their data given to smart systems. Smart devices, IoT devices and AIoT devices are constantly collecting data from you and your surroundings. With more available data, The systems can learn your preference and serve you better. The initial intention of data collection is pure- to understand and benefit users. Unfortunately, from time to time, there are reports on breaches of data collected by smart devices. Since most applications rely on cloud computing services, the collected data from your devices has to be transmitted to the cloud servers. During the journey, your data is exposed on the Internet. At the time when your data leaves from where it is captured, a security risk appears. People have considered edge AI processing as a solution to data issues. Instead of sending data to the cloud for processing, people plan to complete collection, storage and processing within devices. Still, some important data may need confirmation from the cloud servers, and those would still be sent to the cloud. However, compared to the past approach, significantly fewer data would be exposed on the Internet. It makes edge AI processing a favorable solution to maintain the application's operation quality while adequately coping with the data concerns. References: What Is the Internet of Things (IoT)? | Oracle Hong Kong SAR, PRC What is AIoT? Applying AI to IoT Data (iotforall.com) What is AIoT and why does it matter? (electronicspecifier.com) Security is the future of AIoT | TechRadar
Subscribe to us now so you won't miss the latest MarTech information and trends!
By submitting this, we will only use data provided to us in accordance with our Privacy Policy. Comments are closed.
|