Pushing 802.11n technology to new heights
The industry is moving toward the emerging 802.11ac standard.
The performance and capabilities of 802.11n technology have kept pace with market expectations, and now more and more multichannel video programming distributors (MVPDs) are finding the technology adequate – and superior in some instances – for transferring video in home networks.
In most cases, 802.11n technology actually offers superior video-handling performance, coverage and cost-effectiveness to wired options including MoCA, HPNA and powerline communications (PLC).
The 802.11n technology also enables industry-standard Wi-Fi to serve as an ultrareliable, carrier-class, whole-home IPTV networking backbone for complementary single-room wireless technologies, including emerging short-range 60 GHz solutions from companies in the WirelessHD Consortium and Wireless Gigabit Alliance (WiGig).
Meanwhile, the industry continues to further improve 802.11n performance. Consumers increasingly want their wireless connection quality to match that of wired techniques. For those that have to meet those expectations, that means that 802.11n solutions must be capable of delivering up to four simultaneous high-definition video streams at more than 100 Mbps data rates – anywhere in the home – with near-zero error rates.
There are various approaches to achieve this. Some manufacturers attempt to improve performance by increasing output power settings. The industry is also exploring new standards. In the near term, the optimal approach requires solutions based on a 4 x 4 multiple input, multiple output (MIMO) antenna architecture that can support four unequally modulated spatial streams while optimizing performance, reach and reliability through the use of low-density parity check (LDPC) coding and dynamic beamforming.
Beamforming is particularly important for optimizing wireless connections between transmitters and receivers – by estimating the channel between the transmitter and the receiver and telling the transmitter how to pre-compensate on a tone-by-tone basis to optimize SNR of the received signal. If this is done dynamically, the MIMO receiver and transmitter can work together to estimate the adverse effects of any objects that would block or deflect the beam and mitigate and/or preempt those affects by redirecting the beams from each of the transmitting antennas.
Figure 1 illustrates the results of investigations into long-term packet error rate (PER) averages for various MIMO configurations.
A 4 x 4 MIMO architecture combined with other 802.11n enhancements has been proven to deliver superior overall performance as compared with solutions that rely on increased power output.
Some 802.11n reference designs are customized so that their output power setting can be raised from the typical legal and standards-approved +23 dBm (200 mW) to an above-normal +30 dBm (1,000 mW). While an additional 7 dB (800 mW) of power will likely increase range and throughput for most 802.11n designs, operating at such high power levels can introduce a variety of challenging issues.
Tests show that, at every power level, a 4 x 4 MIMO reference design will have significantly longer range than a 3 x 3 design, as shown in Figure 2. At standards-compliant power output settings of +23 dBm (blue arrow), the 3 x 3 reference design has a range of 60 feet. This compares to 100 feet for the 4 x 4 reference design. As the output power is raised, range increases for both the 3 x 3 reference design and the 4 x 4 reference design.
While it’s an interesting lab experiment, the +30 dBm setting is legal in only a subset of the available channels in the primary regulatory domains. It also requires more expensive power amplifiers, increases generated heat and power consumption, and reduces device battery life while causing more interference in adjacent channels. Most product vendors reject this power setting for their products.
Today’s 802.11n standard can provide enough throughput for almost all of today's consumer needs. The industry is also moving toward the emerging 802.11ac standard that will support the transition from fixed to wireless links while accommodating higher throughput requirements. Work begun by the IEEE task group TGac in November 2008 is still in the draft stage, with final approval targeted for December 2013. Once approved, it will become an official 802.11 amendment tounder the name 802.11ac. The main goal is to significantly increase throughput within the basic service set (BSS). The defined target rates are a maximum Multi-Station (Multi-STA) throughput of at least 1 Gbps and a maximum single link throughput of at least 500 Mbps.
As the industry begins deploying 802.11ac technology, it's important to note that the standard delivers higher data rates that consumers will like, but also introduces deployment and QoS challenges that are likely to be of concern to service providers, primarily related to signal interference.
While existing 802.11 technologies operate in the 2.4 GHz band, the 5 GHz band or both, 802.11ac operates strictly in the 5 GHz band but supports backwards compatibility with other 802.11 technologies operating in the same band (most notably 802.11n). To achieve its goals, 802.11ac relies on a number of Media Access Control (MAC) and Physical Layer (PHY) improvements. The PHY improvements include:
- Increased bandwidth per channel
- An increased number of spatial streams
- Higher-order modulation – 256 quadrature amplitude modulation (QAM)
- Multi-user multiple input, multiple output (MU-MIMO)
And 802.11ac also supports advanced digital communication concepts that were first introduced in 802.11n, including space division multiplexing, LDPC coding, shortened guard interval (short GI), space-time block coding (STBC) and explicit-feedback transmit beamforming. The MAC layer also includes many of the improvements that were first introduced with 802.11n. One notable enhancement is the larger maximum size of aggregate MAC protocol data units (MPDUs).
Another big 802.11ac difference is that it goes beyond 20 MHz and 40 MHz channels to also support 80 MHz channels. Optionally, the use of contiguous 160 MHz channels or non-contiguous 80 + 80 MHz channels is also allowed. While doubling channel bandwidth to 80 MHz is an efficient way to increase performance, performance is generally not just a function of the PHY rate. It also is affected by interference from other networks in close proximity, as well as various implications of the selected MIMO architecture.
Interference is a particularly challenging issue. With only four or five available 80 MHz channels, it is harder for an 80 MHz system to avoid interference from neighboring networks (which could be either 80 MHz networks or 20/40 MHz networks).
Figures 3a and 3b illustrate a possible difference in interference scenarios between an 80 MHz system and a 40 MHz system. Figure 3a shows an 80 MHz system occupying four 20 MHz channels, while one of the 20 MHz channels is also used by a legacy 20 MHz system. In this scenario, the 80 MHz system has no way to avoid the occupied channel and must share access to the medium with the 20 MHz system. If access is equally shared between the two systems, the capacity of the 80 MHz system is halved. Note that the 80 MHz system cannot fall back to 40 MHz transmission in this case since the overlap happens in the primary 40 MHz channel.
In comparison, Figure 3b shows how a 40 MHz system can avoid the occupied 20 MHz channel-by-channel selection. In this scenario, the 40 MHz system has full unshared access to the medium. A single-stream 40 MHz system would have the same capacity as the single-stream 80 MHz system shown in scenario 3a. If the same 40 MHz system were to support two streams, however, its capacity would be double that of the single-stream 80 MHz system shown in scenario 3a.
Other 80 MHz system issues arise based on this MIMO architecture configuration. These systems can provide the same performance as a 40 MHz system by using a lower number of antennas. However, reducing the number of antennas eliminates diversity and reduces the robustness of the transmission.
This is problematic for high-quality video content, which requires more than just increasing the maximum ideal PHY rate. MVPDs desire sustained data rates of at least 120 Mbps in order to deliver each video stream to CE devices. They also need near-perfect PER data transfer performance to provide enough quality and reliability for the full range of entertainment applications, such as watching a full HD sporting event on a 72-inch screen or playing multi-player online role-playing games that require low-latency performance using bidirectional controllers.
To ensure stable video delivery, the number of antennas should be higher than the number of spatial streams. Diversity is a critical part of stable data delivery with quality of service. Therefore, even 80 MHz systems will have to be built using multiple antennas if they are going to be used in applications that require reliable multistream video transmission. This narrows the cost and power advantage between a (single-stream) 80 MHz bandwidth system and a (two-stream) 40 MHz system. Also, the bandwidth increase of 802.11ac is a concern where there are limited bandwidth resources. Frequency is a scarce resource that must be used as efficiently as possible.
Exploiting channel diversity by using a higher number of spatial streams allows more efficient spectrum use than simply doubling the bandwidth of the transmission. Channel and antenna diversity, therefore, remain important requirements, even for systems that are capable of wider bandwidth.
It is believed that a 4 x 4 system with a maximum number of spatial streams and MU-MIMO will be required, at a minimum, in order for 802.11ac to fully realize its potential, not to mention spacedivision multiplexing, LDPC, STBC, beamforming and multiple streams, as well as a variety of other PHY, MAC and coexistence enhancements.
Such a system would provide higher bandwidth in sparsely populated networks while providing QoS, good performance and coexistence in denser network environments.
QoS is one of the most important requirements for service deployment. The first major carrier to deploy whole-home broadband wireless video networking to its subscribers was Swisscom, which collaborated with Quantenna and Netgear in 2010 to transport multiple HD video streams from residential gateways to multiple set-top boxes throughout the home.
According to Swisscom, it was vital to be able to guarantee the reach, performance and reliability that its subscribers needed for an excellent experience. These capabilities also ensure that operators can reduce customer support requirements and minimize truck rolls and overall deployment costs. Other benefits include the opportunity to support new operator service models, such as centralized, multichannel DVRs; remote DVR/STB maintenance and management; digital rights management; and content/privacy security.
The industry continues to extend 802.11n technology. Today, the latest 802.11n solutions deliver full HDTV quality with 1080p and higher video resolution, all the time, anywhere in the home. For most current applications, the narrower 40 MHz channels of 802.11n, combined with beamforming and other video-handling enhancements, provide the optimal solution and will likely be the service providers’ choice because of superior reach and reliability. In the future, 802.11ac offers a higher-throughput wireless extension, similar to Gigabit Ethernet, for those that truly have a need for speed.