Edge AI helps solve security issues in the future development of AIoT

Categories:NEWS
Author:apuqi
Origin:Comprehensive network
Time of issue:2020-03-10
Views:304

(Summary description)The proliferation of smartphones and mobile devices has greatly benefited our society. The existence of these smart devices has enriched and simplified our lives. The convenience brought by smart devices and the Internet of Things (IoT) and artificial intelligence (AIoT) has been deeply embedded in our minds, so that we now take it for granted.





Edge AI helps solve security issues in the future development of AIoT

(Summary description)The proliferation of smartphones and mobile devices has greatly benefited our society. The existence of these smart devices has enriched and simplified our lives. The convenience brought by smart devices and the Internet of Things (IoT) and artificial intelligence (AIoT) has been deeply embedded in our minds, so that we now take it for granted.

Categories:NEWS
Author:apuqi
Origin:Comprehensive network
Time of issue:2020-03-10
Views:304

The proliferation of smartphones and mobile devices has greatly benefited our society. The existence of these smart devices has enriched and simplified our lives. The convenience brought by smart devices and the Internet of Things (IoT) and artificial intelligence (AIoT) has been deeply embedded in our minds, so that we now take it for granted.

We believe in these smart connected devices and share a lot of private data during use. We store schedules, contact information and photos on our phones, allowing smart devices to control smart bulbs and smart air conditioners in the house, smart network security cameras monitor our property and valuables, and smart speakers know what we want to hear next First song.

Smart devices can bring us convenience and security. According to Gartner estimates, more than 22 billion IoT devices are currently used globally, and by 2025, 41.6 billion IoT devices will be connected to the Internet. A recent Deloitte study found that in the United States, the average household has 11 connected devices, including 7 smart screens for viewing content. Once 5G is widely available, the number of smart devices in American homes is expected to increase rapidly, and 62% of consumers say they will replace their Internet with mobile broadband.

Fully connected smart homes will soon become as popular as today's smart phones, but a society that realizes full connectivity still faces an important issue: concerns about privacy and security.

Cloud service security issues

With the rapid increase in the expected number of IoT and AIoT devices, we need to address the various security issues that arise. Although we have improved our lives through smart devices and IoT and AI technologies, we have also seen countless reports of leaks of family privacy data. Once the device is damaged, the door is opened for the criminals, allowing attackers to easily access or control various smart networked devices in our home, such as smart light bulbs, network security cameras, smart speakers, and even smart locks that guard the door.

Artificial intelligence (AI) in smart devices allows devices to understand our schedule, recognize our faces, recommend movies to us, and more. Most of these devices only use cloud connections to run AI applications and access the databases they need to run, because the cloud provides an easy way to access the data and processing power needed.

For many years, cloud services have been the first choice for consumers and enterprises because of the obvious advantages of access convenience provided by the cloud. The recent frequent security and privacy violations have caused many users to question whether cloud storage is still the only solution, and privacy violations are only the cost of using these devices. Therefore, there are many problems with using devices that can only work with cloud connectivity.

At present, security is still the number one problem in cloud computing, and there are a large number of public cloud vulnerabilities. From personal information, data certificates to intellectual property rights, once data is leaked, there are hidden dangers and risks of loss.

In cloud computing, the problem of data ownership is a problem that many companies and consumers are trying to solve. The data and encryption key bits are stored in the third-party service provider. Therefore, if an unexpected situation occurs and a downtime occurs, the user may not be able to access the data. If this happens, it may mean that the facial recognition camera that authorizes the user to enter the home will not automatically open the door, etc. The management will have other auxiliary countermeasures, but will cause inconvenience to the user.

Distributed denial of service (DDoS) attacks are already very common and highly destructive. Some services commonly used in our lives may be under the control of attackers at any time. It is estimated that by 2025 there will be 41.6 billion IoT devices online worldwide, and DDoS attacks will likely double.

Currently, the amount of data that needs to be transferred to the cloud for processing is very large, and will only continue to grow as our dependence on cloud services increases. Our bandwidth requirements will also increase exponentially, as well as those that need to be stored and calculated in the cloud. Content and cost will also increase. In the end, these high costs are ultimately paid by consumers.

Terminal equipment edge AI

Edge AI processing on terminal devices is a way to increase cloud service applications. It supports the local processing and analysis of data on the device, and it can work normally without transferring the data to the cloud. Therefore, there is no need to continuously send large amounts of data to servers run by large technology companies or government agencies and store them in the cloud.

Instead, most of the data will be calculated on the device, allowing most useless data to be deleted, and only important data that needs to be confirmed will be saved and sent to the cloud for further processing. This is currently the safest and most private way for many business applications, while also significantly reducing the costs associated with higher bandwidth requirements and cloud storage.

In the past, chips soon became obsolete, but the chips used in terminal device technology are reconfigurable, which means that they can adapt to the upgrade and development of AI technology and adapt to future AI models. Essentially, by decomposing the AI model into basic building blocks, it can be easily reconfigured to take advantage of next-generation software enhancements. Therefore, if the AI technology is significantly updated, the chip can be updated in the same device without replacing it all.

If the terminal equipment with edge computing is fully adopted, it has the ability to create the most secure and private network environment to connect every AI smart device and service in the user's life. Devices will communicate with each other in a private network instead of communicating back and forth between terminal devices and the cloud. This will help to achieve seamless, secure and private applications of all artificial intelligence, thereby improving people's smart life experience.

Hybrid solution

However, if you completely abandon the cloud and only rely on the edge processing power on the device, it will be no different from the mistakes made by developers in the past. This is not a direct choice between cloud services and smart edge terminals. On the contrary, we should devote ourselves to the collaborative work of cloud services and smart terminals.

The security and privacy of smart terminal devices, coupled with powerful storage and data analysis and processing capabilities in the cloud, will shape the next generation of smart devices and the Internet of Things. This hybrid solution will enable AI to achieve its maximum potential. A fully connected society supported by advanced terminal device chips and powerful cloud technology will be closer and more secure than ever. Through the correct technical implementation and deployment, we can take advantage of the technology to enrich our lives and move towards a new AIoT era.