Enterprise WLAN vendors have long had the ability to turn access points (APs) into Wi-Fi monitoring sensors, and recent announcements by the major vendors have renewed interest into whether this makes “sense” (no pun intended). Certainly as Wi-Fi transitions from a nice-to-have to a must-have infrastructure asset, WLAN vendors see the market need to create the best built-in solutions possible. Since 7signal works closely with all these vendors we welcome any technology that helps network operators and IT departments. But don’t be fooled – there will always be wide gaps in capabilities between independent, purpose-build Wi-Fi sensors and APs turned into Wi-Fi monitoring stations, and the cost difference is negligible when all factors are considered. This is especially true when 24/7 end-user quality of experience is at stake.
Here are 5 Factors to Consider:
APs generally use omnidirectional antennas tilted slightly downward towards the clients. Therefore, when they are turned into a sensor, either temporarily or permanently, they will not see other APs as well as a purpose built sensor with directional antennas and beam steering, which helps suppress energy from the areas it is not scanning. This Wi-Fi monitoring technology typically translates into the ability to cover 6-8 access points.
Recently, vendors have introduced software defined radios into their APs with the thinking that this radio could be useful to measure performance. While you probably want to do more useful things with this radio like support more clients, the task of Wi-Fi monitoring with one radio while supporting clients on the same frequencies is quite daunting. The antennas have to be specially designed to minimize interference and prevent the noise floor from raising up to such a level that it adversely affects automatic gain control and roaming algorithms. This greatly limits Wi-Fi monitoring effectiveness by the AP.
There are two ways to turn an AP into a Wi-Fi monitoring sensor. The first is to permanently change it into a sensor, but then you have the coverage problem discussed above. The second is to temporarily stop using the AP (or software defined radio in an AP) to communicate with clients. This could save travel time and costs for troubleshooting rather than sending someone out to the site with a laptop. However, when you operate in a reactive mode like this, by the time you look at the problem it may have gone away. The real value in Wi-Fi monitoring is to be testing all the time, in several parts of the network, so you can identify trends, see the issues before they become critical problems and take remedial actions before people complain.
Most farmers don’t like the idea of the fox guarding the hen-house. The same holds true for application performance and network performance monitoring, which is why APM/NPM is a $5B market today. We’ve seen cases where AP vendors have gamed network class of service priorities in order to improve throughput measurements. And beyond the obvious value, independent Wi-Fi monitoring provides many not so obvious benefits. For example, many of our customers like to see what happens when they upgrade parts of their network. Most times it has a positive effect, which they like to confirm through measurements. However, sometimes it has a negative effect, which they want to understand and correct. If you are upgrading your APs and Wi-Fi monitoring sensors with essentially the same software, then you are likely to lose this capability.
Ultimately the goal of Wi-Fi monitoring is to improve the quality of the end-user experience. Although an AP turned into a sensor could theoretically provide the same software capabilities as a purpose-built sensor, the end-user experience is just not the focus of WLAN vendor. At 7signal we have focused extensively on the entire end-to-end process of attaching, authenticating, portal navigation, IP address retrieval and DNS resolution. Our broad range of active tests include TCP and HTTP throughput testing, Ethernet/WAN testing, website testing, and voice quality measurements. Our Wi-Fi monitoring sensors are high performance end-user clients, benchmarking network capabilities 24x7. Plus, we are watching all the activity between other clients and access points to see what their experience is. Furthermore, our software provides prescriptive steps to address factors which may impact Wi-Fi performance. Add to that our mobile app for iOS / Android devices to crowdsource actual client performance data, and you can build a comprehensive enterprise-wide view of end-user quality of experience.
Ah, but the cost! It’s hard to ignore the free price of Wi-Fi monitoring when it is built into an AP. But if you want a proactive solution versus reactive, you’ll want to dedicate APs to be used all the time as sensors... and then you have coverage problems because of the antenna design. This means you’ll need many more AP-sensors than purpose-built Wi-Fi monitoring sensors. Also, with this market taking off, the price of purpose-built sensors have come down. Add to that the other elements available like small wall-mount sensors that require no Ethernet cabling plus mobile apps for iOS / Android solutions, and the equation starts to favor the independent, proactive solution.
My father-in-law liked to say, “If you are waiting on me, you’re backing up!” Unfortunately, the same holds true for Wi-Fi monitoring. The WLAN vendors make great APs and know how to provide the connectivity required for complex networks. But if you are waiting on them to provide a best-in-class solution to provide the critical visibility to identify and resolve network and client issues before people complain, then I’m afraid you’re “backing up” too!
Want even more technical details from our founder and CTO, Veli-Pekka Ketonen?