Adoption of the 802.11n standard recently pushed Wi-Fi into the realm of mission-critical business. Leveraging its higher speeds and greater capacity, wireless networks routinely carry voice and video traffic, Web and retail transactions, and now even life-and-death healthcare alerts and diagnostics. Simultaneously, as mobility evolved from “nice to have” to a backbone of business, user expectations for availability and performance quickly rose. To obtain the kind of performance inherent in 802.11ac, new testing approaches are essential.
- 802.11ac Wi-Fi Access Points Ramp Up In The Consumer Market
- Understanding IEEE 802.11ac Wireless
- What's The Difference Between 802.11n And 802.11ac?
As a result of the explosion of mobile devices and digital content leveraging Wi-Fi, we are already stretching the limits of 802.11n. To keep pace with demand, the next generation of the 802.11 standard, also known as IEEE 802.11ac, was developed to finally break the wireless Ethernet gigabit barrier. This technology is poised to meet the growing digital content challenge by delivering speeds that are three times faster, 50% to 100% greater bandwidth capacity, and improved range while maintaining ultra-high end-user quality of experience (QoE).
Developers of chipsets, wireless local-area network (WLAN) infrastructure equipment, and mobile client devices are some of the first groups to face the challenges of bringing 802.11ac to reality. They are tasked not only with bringing solutions to market before their competitors, but also are expected to make their solutions better than the competition’s. These challenges are even more difficult than with 802.11n, because the new IEEE standard requires a fundamental change in the way new devices behave and must be designed and assessed.
A Brief History
All 802.11 revisions to date have focused on increasing transport speeds, which lead to higher traffic delivery rates and ultimately to faster response times experienced by the end user (Fig. 1). 802.11n permitted traffic delivery over multiple spatial streams (called multiple-input multiple-output, or MIMO) as well as packet aggregation.
MIMO delivered marked improvements in physical transport rates, enabling more bits per second to be transmitted than ever before across Wi-Fi. Packet aggregation delivered equally remarkable improvements in transport experience, allowing devices to send more data once they had gained access to the wireless media.
802.11ac made Wi-Fi history when it preserved aggregation techniques, advanced the physical transport rates yet again, and introduced the concept of parallel transport into Wi-Fi through a technique known as multi-user MIMO (MU-MIMO), where multiple client devices are receiving packets concurrently. Now for the first time, directed traffic can be delivered to multiple client devices simultaneously. This has a significant impact on content delivery in any location with multiple users, especially where content is revenue-generating or critical.
To achieve the next level of performance, 802.11ac uses 256-QAM (quadrature amplitude modulation) and 80-MHz bandwidths, making the demands on the radio systems in terms of signal fidelity, sensitivity, and noise greater than ever. However, the reality is that 802.11ac must be compatible with existing legacy and 802.11n client devices.
Soon, it will be common for 802.11ac access points (APs) to understand and connect with multiple 802.11n phones and tablets and 802.11g laptops, all of which support a variety of features including quality of service (QoS), power save, and multicast. New 802.11ac chipsets must excel at delivering exceptional performance to other 802.11ac devices, while at the same time gracefully interoperate with a multiplicity of devices including TV sets, tablets, smart phones, and other consumer electronics.
802.11ac uses several advances to achieve the targeted performance. The new specification addresses the need for performance improvement through three primary initiatives: enhancing the raw bandwidth, enabling multiple flows to use the medium concurrently, and optimizing the individual client channels.
To increase the physical-layer (PHY) transport rate, 802.11ac uses a higher-rate encoding scheme known as 256-QAM, which transmits four times as much data as the 64-QAM used in the 802.11n standard. Signal-to-noise ratios that previously worked for 802.11n are no longer sufficient for the higher speeds in 802.11ac because the difference in detectable signal level is now four times smaller.
To further increase the amount of data transported per second, channel bonding approaches made popular in 802.11n have been taken further to provide 80-MHz and, ultimately, 160-MHz-wide channels. Increasing channel bandwidth allows for more data to be transmitted simultaneously out of the same antenna. Legacy versions of the 802.11 devices commonly used 20-MHz channels. When using 802.11n, users could select between 20-MHz or 40-MHz channel operation.
These wider channel bandwidths and the need for proper channel separation mean 802.11ac can only be used in the 5.0-GHz band, where more non-interfering bandwidth is available. Note that dual-band APs will still be produced, but the 2.4-GHz band will be limited to 802.11bgn and will not be able to be configured for 802.11ac.
The use of multiple spatial streams is also a key factor in the 802.11ac equation. The 802.11n standard accommodates for up to four spatial streams to achieve a maximum of 600 Mbits/s of performance, whereas the 802.11ac standard allows for up eight spatial streams.
Finally, packet aggregation is present in 802.11n as well, but is worth mentioning because it is often the single biggest performance multiplier on a per-transmission basis. With aggregation, once a high-performance device obtains its transmit opportunity, the transmitter strings multiple frames together and transmits them in succession without having to reacquire the medium.
To support increased traffic, 802.11ac has moved away from only allowing one 802.11 device to transmit at a time. MU-MIMO allows an AP to transmit data to multiple client devices on the same channel at the same time by directing some of the spatial streams to one client and other spatial streams to a second client.
This is much like driving down a four-lane freeway. Traffic is only significantly delayed when all of the lanes are blocked, so overall throughput can be much higher. There are permutations to this basic principle, but MU-MIMO is critical to performance improvements in environments with high client counts.
The final major performance boost comes from technologies that optimize the communications when speaking to a specific client. The first major enhancement is a concept known as transmit beamforming (TxBF).
With TxBF, the AP communicates with the client devices to determine the types of impairment that are present in the environment. Then the AP “precodes” the transmitted frame with the inverse of the impairment so that when the medium transmits and transforms the next frame, the client receives it as a clean frame. Since no two clients are in the same location, TxBF needs to be applied on a client-by-client basis and constantly updated to reflect the changing environment.
Providing superior 802.11ac requires a rethinking of the usual approach to testing (Fig. 2). Traditionally, the RF section is verified using one set of equipment, and then the upper layer functions are tested using a second set of tools. Previous approaches used a digitized data record approach for both generation and analysis, creating or capturing what are known as I/Q files. Equipment was typically adapted from the general-purpose RF domain.
The new generation of testing should increase the complexity and precision of new designs, as well as adopt a more holistic approach to benchmarking performance. As such, two major fundamental shifts must occur in the approach to testing and assessing performance throughout development.
First, real-time, line-rate traffic generation must replace the use of I/Q waveforms. Traditionally, the RF piece has been verified using one set of equipment while upper-layer functions were assessed using a second set of tools. Vector signal analyzers (VSAs) and I/Q waveforms were employed, but this approach requires extensive expertise. Often, creating thorough validation and performance tests proved both cost- and time-prohibitive.
I/Q waveforms physically cannot be used to generate realistic, continuous flows of frames, which limit assessments to small, insufficient samplings. Verifying performance at 802.11ac’s higher rates along with legacy versions of Wi-Fi requires transmitting and receiving hundreds of frame definitions varied by modulation rate, frame length, bandwidth, frequency, channel model, and power levels.
For transmit functions, the limited samplings captured with the traditional I/Q approach fail to adequately address power supply and clock interference issues as well as the decoupling problems that impact the fidelity of transmitted signals. Receivers must also be subjected to continuous frames at line rate with real-life variation from frame to frame. Unable to subject receivers to realistic flows, the I/Q approach does not enable the real-time analysis of every frame now needed.
Second, layers 1-7 must now be addressed within a single test system. The overall leap in complexity also requires tighter coordination and control between the various layers of the protocol stack to quickly pinpoint performance issues to a specific function of the hardware. 802.11ac’s higher speeds also require greater signal-to-noise ratios because the difference in detectable signal levels is now significantly smaller.
At layer 1, 256-QAM and the wider bandwidths used by 802.11ac pose daunting new challenges to radio designers. New designs must deliver performance advances in virtually every dimension that impacts digital modulation, including phase noise performance, noise floor, and modulation accuracy.
Simply optimizing layer 1 performance is insufficient to ensure a high-performance solution. More stateful than previous iterations, 802.11ac introduces unprecedented complexity at the media access control (MAC) layer. Being able to see both the RF and MAC results at the same time greatly improves problem identification and resolution of layer 1 problems.
Along with addressing layers 1-7, integrated 802.11ac test solutions should enable testers to progress from functional and performance testing to system-level testing using a mix of TCP-based (Transmission Control Protocol) traffic and real application traffic flows. This critical testing equips developers to tune features and algorithms to further optimize application performance.
The Right Questions
Designers should ask a series of questions that address high-level considerations for ensuring the selection of an optimal 802.11ac test system so performance assessments, and ultimately performance itself, are not gated by the limits of the test system.
For example, was the system architected for 802.11ac? The demands of 802.11ac are such that purpose-built test systems now prove essential both to achieving adequate test coverage and minimizing development costs and cycles. From a capability standpoint, the following questions should be asked:
• Does the solution provide integrated level 1-7 testing?
• Can it generate realistic, line-rate traffic and impairments?
• Does it capture and deliver detailed information for each frame in real time?
• Is channel emulation addressed?
• Can the solution be used throughout design, development, and regression testing?
• Can complex testing be automated for repeatability and to minimize effort?
If the solution meets these first-line criteria, additional questions regarding the test systems’ traffic generation capabilities can be asked to evaluate its strength in each critical area.
Next, do the systems’ traffic generation capabilities meet 802.11ac demands? To determine this, the following questions should be asked:
• Can the system generate line-rate traffic for a wide variety of frame conditions?
• Can 802.11ac and legacy traffic be generated in various combinations representative of the actual diversity, rate, and complexity that devices will experience in the field?
• Can any desired frame and sequence of frames be generated quickly and easily, without limitation to adequately test receiver performance?
• Can testers enable multiple features on a single client to ensure that the basic functions work properly (aggregation, power save, IPv4 and IPv6, QoS, etc.)?
• Can the system deliver 1.7 Gbits/s of traffic to test maximum performance at all frame sizes?
• Can performance be benchmarked with multiple clients under both real and ideal conditions, at full scale?
• What if any specialized expertise is required to configure and run tests?
• Will the system continue to keep up with evolving 802.11 standards?
If the test solution meets the demands for traffic generation, more questions can be used to determine if it can capture and analyze the traffic. First, does the system accurately capture traffic and provide actionable insight? To find out, use the following questions:
• Does the solution enable testing of each layer individually to allow visibility into metrics from both the RF layer and the upper layers at the same time?
• Can the system eliminate stray RF effects so performance degradation can be sourced to RF design issues, upper-layer protocol issues, or impairments in the RF spectrum?
• Are RF measurements made at line rate for every frame transmitted? If so, can the system determine each frame’s RF characteristics as well as frame-level performance?
• Do capture capabilities support 802.11ac as well as 802.11n and legacy 802.11a/b/g traffic?
• Does the solution see capture files of problematic sequences with a variety of PHY and MAC layer statistics?
• Does the tester offer frame loss measurement, a critical component for supporting multiplay services?
• How many spatial streams does the traffic capture device support? A MIMO test solution has to support four spatial streams to accommodate 4x4 situations.
Breakthrough Testing For Breakthrough Technology
802.11ac holds the promise of gigabit+ performance that will enable much broader adoption in key target markets such as enterprise, residential video, and carrier hot spots. To realize the performance and density promise, chip and hardware developers must navigate some significant technical challenges and can use the questions and information outlined above to choose an 802.11ac solution that will help deliver the experience consumers expect today.
Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicable.
Joe Zeto serves as a director of product marketing for Ixia. He has more than 17 years of experience in wireless and IP networking, both from the engineering and marketing sides. He also has extensive knowledge and a global prospective of the networking market and the test and measurement industry. Prior to joining Ixia, he was director of product marketing at Spirent Communications running the Enterprise Switching, Storage Networking, and Wireless Infrastructure product lines. He can be reached at [email protected]