Healthcare facilities today are as connected to the wireless ecosystem as any large technology company. Every department from medical services to patient registration to financial services depends on the WiFi in some manner or another. In larger facilities, there can be as many as tens of thousands of wirelessly connected devices. These devices support mission critical processes and must work efficiently at all times. This, in turn, requires the wireless ecosystem to be consistent and reliable. Therefore, any issue that degrades network performance can have serious consequences.
To best protect the network, healthcare facilities need to use real-time and historical analytics. The importance of real-time analytics might be more-or-less self-explanatory: Knowledge of real-time network behavior allows IT departments to quickly resolve issues before they affect end users or morph into colossal headaches. But why do we need historical analytics?
Learn more about how to navigate the cybersecurity challenges that medtech manufacturers, hospitals and healthcare organizations face at the Medical Device Cybersecurity Conference | September 22–23, 2020 | A Virtual EventWhat Historical Analytics Provide that Real-Time Analytics Can’t
A wireless network’s health consists of more than its real-time performance. Trends over time in client distribution, airtime and client utilization, and infrastructure performance reveal the overall health of the network and enable IT to identify what upgrades or updates the network needs in order to continue providing optimal service.
Even if a network is performing optimally day-to-day, it is still undergoing changes that can have a long-term effect on network service. These changes might not be apparent with real-time analytics, but will be easily identified with graphs and charts depicting historical analytics. While it’s never possible to predict the future with complete accuracy, facilities armed with this information can attack budget and network capacity planning even more strategically. The planning process becomes a little less of a guessing game with administrators supported in making well-guided assumptions about how the network will change in the future based on past analytics.
By improving network performance, historical analytics help improve clinical care.
Setting Up a Wireless Network
To truly benefit from historical analytics, a healthcare facility needs two things:
- Analytics from every wirelessly connected device
- Strong network performance
If network performance is consistently intermittent,this is a problem in and of itself. Resolving this issue is necessary to safeguard patient health. Once network performance is consistently reliable, all real-time processes dependent on the WiFi are protected, and a facility can begin collecting historical analytics.
Perform a Site Survey. If a facility is setting up its wireless network for the first time, or if the network hasn’t been performing optimally and it’s been several years since it was designed and installed, it is helpful to have a site survey. Engineers will perform a survey to review a facility’s layout and network needs in order to determine the WiFi design that will provide the best coverage, capacity, and quality of service. Without a site survey, facilities can experience issues such as coverage holes and debilitating interference.
Any significant changes in or around a healthcare facility, such as construction of a new building on a medical campus or renovations of an existing building, can affect a facility’s network. These changes might make it necessary to schedule another site survey. If facilities are using an AI-based analytics solution to monitor the network as discussed below, this solution can provide the data necessary to optimize the network without a survey.
How to Gather Historical Analytics
Once the wireless ecosystem is established, it’s time to begin gathering analytics. It is recommended that analytics are collected with the use of an AI-based analytics solution. This is both because analytics should be gathered 24/7 and AI-systems never tire, and because there will be massive amounts of data packets to capture and analyze. This would be a full-time job for multiple IT employees. Using an AI-based analytics solution means that those employees can focus on other mission critical responsibilities that can’t be performed by AI-technology.
Solutions with the following capabilities will provide the analytics needed for a complete review of network performance, as well as optimize the network in real-time so that service is always on-point and data is never lost.
Historical Analytics. Be advised that not all analytics solutions automatically save historical data. Look for a solution that does so or be sure to periodically save the data yourself for compilation and review at a later date.
Constant Monitoring. The entire network ecosystem needs to be monitored at all times—even during off-business hours. This means more than monitoring wireless devices. A network operates in the radio frequency (RF) environment, and this environment includes any nearby networks that are sharing the airspace, as well as non-wireless devices like microwaves that can cause interference. This entire environment must be monitored to gain the truest understanding of network health.
Additionally, this environment is dynamic, which means that there is always a chance that a change will occur that affects a facility’s network performance. Without 24/7 analytics, there can’t be complete visibility into how the network is currently performing, and how that behavior has changed over time.
For true visibility, the solution must use multiple radios. There should be one radio each for the 2.4GHz and 5GHz spectrums, as well as an additional radio to run network tests. Without multiple radios a solution must switch from spectrum to spectrum, or from spectrum to testing, which results in missing data and lost analytics.
Scheduled Network Tests. The key to optimizing network performance is proactive behavior. Historical analytics enable a facility to proactively identify network trends and plan accordingly, and network tests support an IT department in proactively identifying issues and resolving them before end users are affected. As network tests run over time, consistently gathering metrics on network health, the results are added to other historical analytics for a complete network picture. Tests can run as frequently as every hour, and should run at least once a month.
Automatic Alerts. If the analytics solution identifies any change in network behavior, it should automatically alert IT for the quickest resolution possible. Ideally, it will identify the root cause of the change and provide suggested recommendations for resolution. This practice supports a strong and consistent WiFi network, as it enables IT teams to respond to issues proactively rather than having to wait for end-users to alert them to difficulties.
Time-to-Value. Consider the time-to-value offered by the solution. Facilities need at least a month’s worth of data in order to start identifying historical trends. This means that the analytics solution used should start working as quickly as possible. Additionally, the solution shouldn’t stop working if the facility changes network vendors (i.e., it should be vendor agnostic). There’s nothing worse than having to search for a new solution and start from scratch simply because all access points were switched out.
Bonus: Graphs, Charts. While not necessary, if you’re looking for a little bonus, look for a solution that automatically creates graphs and charts that depict trends over time. If these visualizations are easily exported to share with colleagues, all the better.
Using Past Data to Support Future Connections
Network optimization isn’t only about the here-and-now, but also about looking forward over the next three to five years. What will a facility be asked to support during that time? What changes might it see in the number of patients, or in wireless device utilization? The best and most strategic answer will be found with the use of historical analytics. These analytics give a facility the power to look back and observe how data such as client distribution, airtime and client utilization, and infrastructure health have changed over the last few months or year. Armed with this knowledge, the facility can make the best decision possible about how to invest in its wireless network to support optimum health and performance moving forward.