AI Case Study
Premier Resort manages its indoor wireless location services using machine learning
Premier Resort automates and scales its indoor wireless location services management without disruptions using Mist's AI-based wireless platform. The platform also helps them monitor the state of connectivity as well as provide location based recommendations for customers.
Consumer Goods And Services
Travel And Leisure
"To deliver an enriching experience, the property requires a state of the art wireless network that enables a personalized location-based service using Bluetooth® Low Energy (BLE).
With hundreds of thousands of square feet for meeting space positioned on three different levels, the resort has unique wireless challenges that traditional WLAN and physical beacon-based location systems could not address. To handle the scale, performance, and ease of use required at the Resort, the property management company uses Mist. the Resort is also using
the Mist Bluetooth® LE solution for proximity-based
messaging to guests.
The Mist platform has a unique Proactive Analytics and Correlation Engine (PACE), which provides the foundation for AI data collection and analysis in the Wi-Fi / BLE domain. PACE collects over 100 pre- and post- connection user and location states in near real-time from every wireless device. This state information is sent to the Mist cloud, where AI algorithms are used for real-time analysis. In addition, the Mist environment uses design intent metrics to classify and monitor the wireless user experience using AI. For example, Mist lets you set, monitor, and enforce your own Service Level Expectations (SLE) for various key Wi-Fi metrics such as “Time to Connect”, “Successful Connections”, “Throughput”, “Coverage”, “Capacity”, “Roaming” and “AP uptime”. These are then used to quantify the Wi-Fi performance of clients, Access Points, and entire locations. For example, you can define a throughput SLE of 30 Mbps for all users in your main campus. PACE will tell you exactly what percentage of the time this SLE is being hit, which users are not getting this level of service, and which device types/operating systems/applications are consistently causing problems. In addition, it can predict if this SLE will be achieved in the future based on current conditions.
Mist’s architecture allows for the capacity and performance to aggregate global metadata across customers. Not only is Mist capable of collecting data for insight into a specific client behavior and location information, it can provide insights and analytics across device types, operating systems, applications, and more. This is key for baselining and monitoring trends, and identifying macro issues early so they can be addressed proactively.
For example, client roaming time, hardware radio performance and device throughput can all be analyzed to identify global issues, such as a performance degradation when a new client operating system version is released.
The Mist system is designed so a disaster does not affect Wi-Fi users. All the business critical services are delivered at the edge through the Access Points. In the rare event of a cloud connectivity disruption for the Access Points where the WAN is still functional, all business critical services will continue to be delivered at the edge through the Access Points. Any existing client device already authorized will continue to access applications through Wi-Fi without undergoing any disruption of services. In case of a WAN outage, all local services will continue to function through the wireless network while WAN services are restored. In other words, Mist Access Points at the edge are completely site survivable in the case of a customer WAN outage or a catastrophic cloud outage"
The Mist Cloud monitors utilization of different services and
scales each module up or down dynamically without requiring end user intervention. The Mist cloud offers elastic scale, without a physical cap on the number of Access Points, client devices, or sites (per customer or globally.)
The Mist platform has a unique Proactive Analytics and Correlation Engine (PACE), which provides the foundation for AI data collection and analysis in the Wi-Fi / BLE domain. PACE collects over 100 pre- and post- connection user and location states in near real-time from every wireless device. This state information is sent to the Mist cloud, where AI algorithms are used for real-time analysis.
The Resort is a flagship property for a major hotel brand and is in an ideal location for vacations, business events, and conventions.
The Mist microservice architecture allows multiple different versions of the same microservice to run in parallel, on the same data sources and environment. This allows a new service, or version of an existing service, to run on real data before being put into live service. Once the data accuracy and performance is verified, the individual microservice switches over to the new version on the live system, with minimal downtime and customer impact.