Ultra Low Power 802.11n Wi-Fi – Wireless Connectivity for “The Internet of Things”
By N. Venkatesh, Redpine Signals
The IEEE 802.11n standard was proposed through a ‘High Throughput Study Group’ effort at the IEEE. Although the major focus of the standard has been on deriving high user throughputs in a WLAN environment using multi-stream MIMO techniques, the standard’s benefits are not limited to devices with multiple antennas. There are a plethora of devices that require ultra-low power connectivity. Computers, electronic devices and machines all talk to each other exchanging volumes of information in a variety of scenarios. More than just communicate with each other, these devices form a vast, IP-based network – ‘the Internet of Things.’ In this article, we look at how these low power battery operated devices can benefit from 11n and why new designs will use 11n rather than 11g.
The 802.11n standard, at first glance, defines ways of obtaining high data rates through doubling of the channel bandwidth from 20 MHz to an optional 40 MHz, and through the use of two, three or four stream MIMO. This directly indicates data rates up to eight times of legacy rates. But the standard also defines ways of increasing on-air data rate, as well as user throughput, outside of these hardware intensive and processing power intensive techniques.
Physical Layer Rate Enhancement
At the physical layer, 11g or 11a OFDM uses 48 subcarriers every symbol to carry data. 801.11n increases this to 52. Each symbol lasts 4 microseconds. At the highest data rate, legacy or 11n OFDM both use 64-QAM – using 6 bits to modulate each subcarrier. In addition 11n defines an error coding rate of 5/6 as opposed to a highest coding rate of ¾ in 11a/g.
The 11a/g OFDM transmission uses guard intervals between successive symbols transmitted. This interval of 800 ns duration helps protect a receiver from multipath effects. The 11n protocol provides an option to reduce the guard interval to 400 ns. This reduces the symbol time to 3.6 microseconds from 4 microseconds and results in an increased data rate of 72 Mbps. However, this limits the scenarios in which error-free reception can take place.
MAC Layer Enhancements
The goal of 11n has been to increase user throughput – the actual data rate seen by an application such as file transfer or video streaming. The WLAN protocol includes various overhead that reduces user throughput. 801.11n has introduced several ways in which this overhead is reduced. Much of this effort has addressed the MAC layer, for it is here that overhead is more significant.
The MAC layer protocol defines inter-frame spaces and the need to acknowledge every frame transmitted, that in particular result in appreciable overhead. The overhead can even be longer than the entire data frame. In addition, contention for the air and collisions also reduces the maximum effective throughput of 802.11. 802.11n addresses these issues by making changes in the MAC layer to improve on the inefficiencies imposed by this fixed overhead and by contention losses.
Aggregation and Block-Ack
A standard WLAN set of packet exchanges follows a timeline similar to what is shown below.
802.11n helps reduce this overhead by ‘aggregation’ – putting multiple data frames into a single transmission. Aggregation can be of two types – of MAC Service Data Units (MSDU) or of Message Protocol Data Units (MPDU). There are some differences and unique features of each, but broadly they enhance throughput by increasing the ratio of data to overhead as shown below.
MSDU and MPDU aggregation differ in the constitution of the individual data frames – and therefore of the nature of the Acknowledgement. MPDU uses a ‘Block Acknowledgement’ method where the individual constituents of the aggregated data frame are separately acknowledged. In the case of MSDU, the recipient sends a single ACK in response to the aggregated frame. The MPDU method uses more overhead, but is more robust in most wireless environments where there is a less than minimal possibility of packet loss.
Reduced Interframe Space
The 802.11e extension for quality of service added the ability for a single transmitter to send a burst of frames during a single, timed transmit opportunity. During the transmit opportunity, the sender does not need to perform any random backoff between transmissions, separating its frames by the smallest allowable interframe space, the short interframe space (SIFS).
802.11n improves on this mechanism, reducing the overhead between frames, by specifying an even smaller interframe space, called the reduced interframe space (RIFS). RIFS cuts down further on the dead time between frames, increasing the amount of time in the transmit opportunity that is occupied by sending frames. However, RIFS can be used only in greenfield deployments—that is, only deployments where there are no legacy 802.11a, b, or g devices in the area. Additionally, in such greenfield environments, the 11n standard has defined the use of a shorter Greenfield preamble that further reduces PHY overhead.
Sustaining Higher Rates Over a Larger Area
One of the drawbacks of wireless transmission is that the quality of data transfer deteriorates over distance or change in the environment. The mechanism adopted by 802.11 to counter this is to use one of a number of available data rates with the general rule that a lower data rate is more robustly delivered over a greater distance. In an indoor environment, though, it is not distance alone that degrades a signal. The more significant factor here is multipath – brought about by reflection from walls, ceiling, furniture, and other obstacles. 802.11n provides techniques to help counter the problem of multipath. In general, receivers can counter multipath on their own by employing multiple antennas to obtain diversity, but the 11n standard has made it possible to have this diversity in single-antenna handheld units.
One of the techniques is the use of Space Time Block Codes (STBC), which uses two (or more) antennas at the transmitter that send out modified repetitions of the same symbol in order to provide a choice of two versions of the transmitted signal at the receiver. STBC can provide over 3 dB of gain at the receiver.
Another facility provided by the 11n standard is beamforming. Beamforming uses multiple antennas at a transmitter to create a strongly aligned signal, with higher signal strength, at a particular receiver. The beamformer requires knowledge of the characteristics of the signal path from each of its antennas to the antenna of the receiver. This is computed during an initial sounding exchange. The gain proffered by beamforming can vary and can even exceed 8 to 10 dB, depending on the number of antennas employed at the transmitted (the access point).
Developers of wireless devices further reduce operational power consumption through innovative means. For example, functionality is partitioned between hardware and software with the criterion of picking the method with the lower power consumption. Individual functional blocks may have multiple modes of operation – offering a choice of low power or high performance for selection based on profile of operation. Fine grained power control methods ensure maximum possible time spent in sleep states.
Benefits of Higher Throughputs and Greater Coverage
The benefits of higher throughput could be obvious – mainly a better user experience of connectivity where higher data rates are desired. However, for the apparently low throughput needs of handheld devices, significant benefits come in the form of lower latency, better QoS, and most importantly lower power consumption.
In addition, for the network service provider, 11n with its higher throughputs overall, and in particular higher throughputs over a wider area mean a greater density of users and the ability to provide each user with a higher user throughput.
The alternative way to look at low power operation is through energy efficiency. Battery life of portable devices is what governs the user experience – and 802.11n provides for higher energy efficiency. For example, let us consider a case where a user with a handheld device moves around in an office, and where he has the need to download data or files into his device via the access point positioned in the office. A plan of the office with the sample locations is shown below.
At each location the useable data rate of the device depends on the quality of the signal at that point. 11n, being capable of robust data transfer through techniques such as STBC and beamforming, would provide a higher useable data rate for a given condition. In this example we use STBC to provide the robustness. In addition to a higher useable data rate, we saw that 11n also helps reduce overhead to provide a greater efficiency of data transfer. These benefits together provide the ability to transfer data into an 11n handheld at greater energy efficiency than into an 11g device. Energy is saved mainly because the handheld device can enter into a low power standby state as soon as transfer is done – and this happens much sooner in an 11n environment.
The figure below summarizes the energy savings in this example.
Sustaining the 802.11n Advantage
Today, a majority of embedded devices that integrate WLAN still use the legacy 802.11b or 802.11g standards. There is no perceived need to have access to high data rates. We have seen, however, that 802.11n provides for enhanced battery life of devices; and with easy to integrate modules available from vendors, such as Redpine Signals, 11n is the preferred method for new designs. There is another, significant, benefit accruing from the use of 802.11n – the enhancement of the overall capacity of the network, enabling greater device concentrations and a more satisfactory user experience. However, this benefit of 802.11n can be fully realized only when all nodes on the wireless network are capable of communicating using 802.11n methods or are compatible with 11n. The presence of legacy 802.11a/b/g nodes in a network forces the other 802.11n nodes to resort to the use of RTS/CTS protection mechanisms to preserve network integrity, thereby reducing overall network capacity by 30 percent or more.
Wireless LAN based on the IEEE 802.11n standard will thus be a primary means of connectivity in low power devices. Redpine Signals has worked on abstracting the complexity of wireless integration so that handling the wireless interface becomes the same as handling any traditional I/O, enabling pervasiveness of 11n in a multitude of embedded systems.
N. Venkatesh is the vice president of advanced technologies at Redpine Signals, and has over 24 years of engineering and management experience in wireless system design, chip design, telecommunications, optical networking and avionics. He holds a Masters Degree in Electrical Engineering from the Indian Institute of Technology, Madras, India.