More

    AI Key to Efficient Management of an Ohio School System Wireless Net

    North Canton Schools (Ohio) are using AI and machine learning to improve wireless services and make life much easier for those running the network.

    The district consists of 11 buildings and serves about 4,400 students. Wireless is ubiquitous: Students, faculty and staff connect and it is used for devices such as projectors. It’s a fractious environment, says Jon Strong, the co-founder and managing partner of Technology Engineering Group (TEC), which manages the network. “Schools are the most brutal environment for wireless,” he told IT Business Edge. “It has the most variables and is the most uncontrolled.”

    There are two deeply related elements to what the stakeholders – the school system, TEG and the vendor, Mist – are doing to address the challenge. The first focuses on the overall approach to managing the devices and diagnosing problems. The second involves AI, machine learning and natural language processing.

    The school system had relied on Cisco Meraki equipment, some of which remains. In the initial architecture, which is common in WLANs, the system assessed problems and collected management data from the core looking “downward” toward individual users. The problem is that the device itself is not part of the equation. The closest point to the device that is being assessed is its access point (AP). Information generated in this top-down approach, Strong said, “is never deep enough, never real-time enough, never complete enough.”

    The system is using 300 Mist AP 41 802.11ac access points and the company’s Assuance Cloud service. Mist’s approach is to work in the other direction — from the device to the core of the network. If, for instance, a student on a Chromebook is sitting behind a filing cabinet, a traditional top-down approach wouldn’t readily identify the obstruction as the problem or even know that there is a device looking for connectivity.

    The Mist approach would, because the Chromebook would connect to the Mist AP via Bluetooth Low Energy and provide the key data. “For our customers who are leveraging BLE, we are using the employees as proxy testers on the Wi-Fi side, and there is a mobile device component to the data element,” wrote Mist CTO Bob Friday in response to questions from IT Business Edge. “On the data side, we take the data from the Access Point, the data from the client, and the virtual assistant injects data from these points.”

    The system also manages the network more generally. Running such a chaotic environment is difficult. Bandwidth is limited and there is a complex interplay between channel use (channelization) and power allocations in adjacent sectors. “No mere mortal can hunt and peck all those scenarios,” Strong said. “It’s impossible.”

    The AI provides system management and machine learning that enables the system to adapt over time based on changing conditions. “Over time, the system will create a baseline of your environment and make recommendations of what is needed,” Strong said. “The first two most tangible will be the channelization and power settings of the AP.”

    The machine learning element harnesses that profile and the data is amassed and used to limit issues in the future. Machine learning, Strong said, “is using the AI results over time to do something, to move forward.”

    AI is the key element of Mist’s platform. “AI is not an ‘element’ per se, rather, it’s the concept of building a system that does something on par with a human,” Friday wrote. “To build that system, the elements needed are deeply integrated — these include Natural Language Processing (NLP), a variety of machine learning algorithms and the data set, including user information from the edge.”

    Mist also uses Marvis, a virtual assistant, to make it easier for technicians to use the platform. In the future, techs will be able to verbally ask the system about current levels and other details. At this point, however, the platform relies on written input from the tech. “It’s very enabling,” Strong said. “As an end user, I can troubleshoot the site on what is going on right now.”

    Carl Weinschenk covers telecom for IT Business Edge. He writes about wireless technology, disaster recovery/business continuity, cellular services, the Internet of Things, machine-to-machine communications and other emerging technologies and platforms. He also covers net neutrality and related regulatory issues. Weinschenk has written about the phone companies, cable operators and related companies for decades and is senior editor of Broadband Technology Report. He can be reached at cweinsch@optonline.net and via twitter at @DailyMusicBrk.

    Carl Weinschenk
    Carl Weinschenk
    Carl Weinschenk Carl Weinschenk Carl Weinschenk is a long-time IT and telecom journalist. His coverage areas include the IoT, artificial intelligence, artificial intelligence, drones, 3D printing LTE and 5G, SDN, NFV, net neutrality, municipal broadband, unified communications and business continuity/disaster recovery. Weinschenk has written about wireless and phone companies, cable operators and their vendor ecosystems. He also has written about alternative energy and runs a website, The Daily Music Break, as a hobby.

    Get the Free Newsletter!

    Subscribe to Daily Tech Insider for top news, trends, and analysis.

    Latest Articles