How to Test a Wi-Fi Embedded Medical Device
Hospitals are one of the most compelling, yet difficult RF environments. To ensure reliable functionality, a Wi-Fi radio that is embedded in a medical device must be tested thoroughly.When choosing an embedded Wi-Fi radio, medical device manufacturers should choose radios from manufacturers who test:
- RF Performance
Testing RF Performance
Achieving adequate RF performance in Wi-Fi radios is dependent on multiple factors. Manufacturers must test RF performance to ensure regulatory compliance, determine RF range, and confirm the ability to maintain connectivity. There are a number of government regulatory bodies such as the FCC and ETSI that each have their own requirements. Regulatory certification testing for each one is often expensive and time consuming for medical device manufacturers, which is why it is typically preferable to choose a radio from suppliers who provide as many certifications as possible. Testing the RF range of a wireless radio is typically done in an open air environment to determine the maximum possible range. This is not the best method, rather range testing should be done in a controlled lab environment. In a lab environment it is possible to automate and simulate many of the RF characteristics of a hospital, for instance, materials that absorb or reflect RF transmissions (resulting in multipath interference), interference from other Wi-Fi clients, or other sources of Wi-Fi interference. The best way to do interference testing is also in a lab.
Although testing Wi-Fi interoperability in a real hospital is ideal, it is also next to impossible. It only allows for a glimpse of the functionality of one specific wireless infrastructure from one vendor with one controller software version and APs with one specific firmware version. It is also difficult for medical device providers to do interoperability testing on their own because it would be costly to obtain infrastructure from every possible vendor. It is less difficult to do interoperability testing in a lab setting. Labs can more easily set up common infrastructure configurations with the most popular firmware and software versions to complete testing. One key advantage of a lab is that you can use APs and other Wi-Fi infrastructure gear from different vendors without having to mount the APs in the ceiling.
WPA2-Enterprise is the preferred approach for Wi-Fi security, however, not every medical device supports it. As a result, Wi-Fi networks in some hospitals support WPA2-Personal instead of WPA2-Enterprise. Because a medical device vendor often cannot dictate to a hospital what method of Wi-Fi security must be used in the hospital, it is important to test that the Wi-Fi radio in the medical device supports both WPA2-Enterprise and WPA2-Personal. Tests of WPA2-Enterprise support should be conducted with a broad range of EAP types and a range of popular authentication servers such as RADIUS (Remote Authentication Dial-In User Service), the most widely-used protocol for authentication servers. TACACS+ (Terminal Access Controller Access Control System Plus) is a Cisco-developed product that is also popular. Security tests must ensure that the radio consistently connects to an AP, passes data to the network at expected data rates, receives data from the network at expected data rates, and roams from one AP to another seamlessly. These tests should work regardless of the Wi-Fi infrastructure and the authentication server on the network.
Laird tests for connectivity and roaming in both the 2.4 GHz band and the 5 GHz band. The major difference between radio operation in each of the bands is that the 2.4 GHz band has fewer channels, so scanning for a new AP takes less time, thus roaming is quicker. In addition, the 5 GHz band has dynamic frequency selection (DFS) channels. The use of DFS channels is a mechanism to allow unlicensed devices to use the 5 GHz frequency bands already allocated to radar systems without causing interference to those radars. If an end user device has DFS channels enabled, the scan time increases significantly. Roaming in the 5 GHz band is more challenging because it takes much longer for the radio to roam to find a new AP. A few ways that Laird tests radios for connectivity and roaming is by using automated tests and by doing “walk around” tests. Walk around tests are done with a complete infrastructure network set up and a test engineer physically walks around with the wireless device. One specific walk around test is an edge of coverage test. In this test, the wireless device is walked outside of the network and then moved back into the network to ensure that the radio stays connected to the edge of the network and then reconnects as soon as the radio is back in range. This test simulates operation in a challenging environment. In addition to this test, Laird does a catastrophic roam test. In a typical “smooth” roam test, the client device experiences gradually decreasing signal strength. In contrast, in a catastrophic roam test the client quickly loses connectivity with its associated access point. This test evaluates the time it takes for the Wi-Fi radio to roam and reconnect to an AP. In an ideal situation, the radio will quickly connect to the original AP or a new AP without a break in the data transfer. These and all other roaming and connectivity tests are done with different possible radio settings to ensure that the radio roams and stays connected in all situations in a variety of roam environments. Sections of this blog post were originally published in Laird's "Testing Wi-Fi Functionality in Medical Devices" white paper and were also featured in the June 2014 issue of AGL Small Cell Magazine.