Virtualised wireless networks - Improving user satisfaction and mobile teaching delivery
Henry Batten - Meru Networks
Traditional wireless networks, often predicated on home-based wireless designs, are based upon multiple channels being utilised to provide coverage across a campus environment Challenges of co-channel interference can occur due to the complexity of education establishment building designs, the lack of channel space available, and the dynamic RF nature that the mobility of education environments present.
In 2012, IDC predicts smartphones and tablets will outsell laptops and desktops for the first time ever. These types of devices can’t be cabled, and rely on wireless as the principle means of connectivity. These devices are typically purchased by younger age groups and find their way onto campus networks. Students think nothing of trying to run media rich applications wherever, and whenever they can. Consumer devices, by design, “stick” to access points, as there’s nothing worse than losing connectivity in a home environment. The introduction of hundreds or thousands of “sticky” consumer devices, all making their own roaming decisions, seriously affects overall system bandwidth and speeds.
Education customers suffer from very high densities of users, often congregating in the same spaces, and these users synchronously want to access wireless networks at the start or end of lessons / lectures. In these high density areas, contention over the air leads to less bandwidth being shared by more users. The resulting degraded bandwidth per device leads to lower satisfaction and a poorer user experience with the wireless network, often in areas like libraries where students need the bandwidth.
The introduction of the 11n standard was often touted as the answer to many of the above issues. The standard allowed for increases in bandwidth over the air. However, many education establishments couldn’t take full advantage of the 11n standard due to a number of reasons. Firstly, consumers and education customers purchased “11n” clients unaware that there are vast differences in the potential speed of different 11n chipsets. Secondly, within the 2.4Ghz spectrum, there’s not enough space within the spectrum for traditional wireless networks to operate a 40Mhz “full speed” 11n network. Even running “slower” 11n using 20Mhz wide channels was problematic due to the increased interference propensity of MIMO antenna technology used in 11n systems. Often customers were forced to use 5Ghz for 11n. Unfortunately many consumer based devices don’t support 11n within the 5Ghz spectrum, and deploying a 5Ghz 11n network required more access points.
The above issues for wireless networks in education will be discussed in the context of virtualised wireless networks where the network is in control of the devices. Newer, virtualised wireless networks coordinate information across all access points, where the network has a global view as to which access point can best serve the application and bandwidth needs of each individual client.
“Network in control” virtualised wireless increases bandwidth per device leading to increased user satisfaction with wireless networks