An elegant alternative to homegrown test environments for wireless systems.
By Fanny Mlinarsky, Azimuth Systems
As Wi-Fi technology matures, wireless LANs are moving from the relatively tolerant SOHO market to the demanding enterprise a market where high network performance is needed to support mission-critical applications, a large number of network users, and a diversity of network elements. Enterprise IT managers need accurate performance data on wireless systems to ensure the interoperability, functionality, and performance of the wireless infrastructure.
Testing of wireless access points, clients, and networks is critical to developing WLANs hardy enough for enterprise adoption. Yet WLAN testing is often conducted in unstable environments where it is impossible to ensure the repeatability and reliability of test outcomes. The result is a testing process that is costly and unreliable.
Complexities of WLAN Testing
Wireless networks are more complex than Ethernet-based networks due to the instability of the physical layer (air) and the mobility of the end user. WLAN system developers and test laboratories go to great lengths and expense to test WLAN networks and devices, often fruitlessly, due to the difficulties in reproducing the wireless environment in the test lab.
A wireless testing environment should allow for mobility of the devices under test while at the same time eliminating random interference. This is difficult since interference is generated from a wide variety of RF-emitting sources, such as 2.4 GHz phones, microwave ovens, radar, adjacent channels, and objects in motion.
Current Testing Methodologies
To reproduce real-world wireless networking conditions, system developers have relied largely on the development of homegrown solutions such as large screen rooms to isolate devices under test from extraneous RF interference. To test large-scale random mobility and roaming, some system designers rent empty office buildings or even open outdoor spaces such as football fields, and perform tests using mobile carts piled with laptops and other client devices.
Such approaches present several problems that demonstrate the near impossibility of acquiring repeatable results in situations where RF cannot be controlled. First, isolating the network under test from extraneous RF noise while simultaneously subjecting it to real world network conditions such as mobility and roaming involves conflicting requirements, because mobility and roaming are difficult to test in the physical limitations of an isolated room.
As seen in Figure 1, tests performed in open air environments also lack the repeatability and interference containment required to yield accurate results and measurements. And finally, these manual techniques lack scalability and automation, which are required to maintain a cost-effective testing solution.
Testing in a Cabled RF Environment
Testing WLANs in a controlled RF testing environment is preferable to testing in the open air because it:
Enables real-world wireless networking conditions to be emulated in a repeatable manner.
Provides the highest levels of repeatability and efficiency.
Eliminates correlation difficulties between different organizations and geographically dispersed sites.
Minimizes time spent retesting due to inconsistent results.
Eliminates the need for space consuming and costly RF screen rooms.
Allows engineers to work side by side without overlapping interference.
Eliminates the need to design, build, and maintain homegrown, non-standard test beds.
Removes the RF complexity from the test setup.
Minimizes problems in the field by allowing for efficient and simple generation of real world scenarios such as roaming, rate fall-back, and hidden stations.
A new generation of off-the-shelf solutions for validating WLANs replicates the network in a controlled, cabled environment by stabilizing the RF connection, isolating the devices under test in a shielded chamber, and interconnecting devices via a network of RF attenuators and combiners. By incorporating RF isolation into the testing environment, these platforms emulate real-world wireless networking conditions in a repeatable manner and provide the highest levels of repeatability and efficiency. They eliminate open air interference, costly constraints, and the often uncomfortable working conditions of a homegrown testing environment.
Such platforms have programmable, controlled RF test beds and allow the user to configure an entire WLAN network on a bench top chassis that is designed for complete RF isolation and control. A software engine provides automation of configuration and analysis, and the user can thoroughly evaluate WLAN equipment and networks under varying conditions and traffic patterns and precisely analyze the results. The clients and access points that comprise the test network are actual clients and access points of the user's choice.
Such standard test systems will eventually replace homegrown solutions such as shielded rooms or rented office buildings that are unable to provide repeatable test outcomes.
By providing wireless system designers with repeatable and consistent test results, reduced time-to-market for new products, and decreased cost of test, these new products will hasten the enterprise adoption of WLANs.
Glossary of Acronyms
SOHO - Small Office, Home Office
WLAN - Wireless Local Area Network