By: Jean-Daniel Wu, Antenna Specialist AE, Future Electronics
The use of multiple antennas based on Multiple-Input Multiple-Output (MIMO) techniques is the key feature of 802.11n/ac, which in theory outperforms 802.11a/b/g by higher throughput. Engineers who are interested in designing Wi-Fi modules are presented with the problems of whether 802.11n/ac shall be implemented and how much performance improvement can be achieved. This paper will give an overview of the evolution of the IEEE 802.11 standard. Then, it discusses the challenges faced in 802.11 wireless channel, namely path loss, shadowing and multipath effect. Finally, MIMO techniques, a key feature in 802.11n/ac, are introduced to overcome the fading effect with higher throughput.
Since the first establishment of wireless LAN or Wi-Fi standard 802.11 by IEEE in 1997, Wi-Fi standard has evolved a lot from 802.11a/b/g to 802.11n/ac; other standards in the family (c–f, h–j) are service amendments and extensions, or corrections to previous specifications. 802.11b was the first widely accepted wireless network- ing standard, followed by 802.11a, 802.11g, 802.11n, and 802.11ac. 802.11b uses the ISM 2.4GHz band which allows for only a maximum throughput of 11Mbps. 802.11a operates around less crowded ISM 5GHz with higher throughput up to 54Mbps. Combining the best of 802.11a and 802.11b, 802.11g supports bandwidth up to 54 Mbps, and it uses the 2.4GHz frequency for greater range. The 802.11n standard is designed to improve 802.11g in the amount of throughput supported by utilizing multiple Wi-Fi signal channels and antennas instead of one. 802.11ac offers backward compatibility to 802.11b/g/n and supports data rate more than 200Mbps. The key parameters and performances of the above 802.11 standards are summarized in Figure 1.
|Max Throughput Mbps||54||11||54||100+||200+|
|Indoor Range (Ft)||115||125||125||230||115|
|Outdoor Range (Ft)||330||460||460||820||330|
|Frequency (GHz)||5||2.4||2.4||2.4 & 5||5|
|Bandwidth (MHz)||20||20||20||20, 40||20, 40, 80, 160 (optional)|
|Max MIMO Spatial Streams||NA||NA||NA||4||8|
Figure 1: Summary of main classes of 802.11 standards
Higher Throughput and Extended Range
802.11a/b/g Wi-Fi modules target applications that require low cost, low throughput, low band- width, and low power consumption. The main advantages of 802.11b are low cost and better signal integrity due to its higher sensitivity at lower throughput. 802.11a solutions are typically more expensive than those using 802.11b due to more costly components working at 5.8GHz. By moving back to 2.4GHz, 802.11g solutions are more vulnerable to interferences caused
by other wireless devices in 2.4GHz band. Due to increased demand for robust solutions with higher throughput, 802.11a/b/g only modules are less popular nowadays in the market.
At the expense of higher cost antenna subsys- tems and multiple paralleled RF signal process- ing blocks to go with their multiple antennas, the highest data rate in 802.11n is 600Mbps while in 802.11a/b/g is 54Mbps. This is an increase of 11 times the data rate and can be achieved by using four antennas (x4), double bandwidth channels of 40MHz instead of 20MHz (x2), and extra 40% improvement by tweaking the OFDM modulation and coding rate. With multiple antennas to form an array, range can be extended by weighing Wi-Fi signals on each transmitting antenna separately to allow a high gain antenna beam. Besides, the antenna array can help null out adjacent interference sources. The leading manufacturer of Wi-Fi products are promoting the most comprehensive solutions in 802.11a/b/g/n. Some of them have even implemented 802.11ac. The typical Wi-Fi solutions benefiting from the MIMO-based multiple antennas techniques are listed in Figure 2.
|Manufacturer||Series Designation||Standard||MIMO Spatial Streams|
|Cypress||BCM4354/4356||802.11a/b/g/n/ac||2 x 2|
|GainSpan||GS2011MIPS, GS2011MIES||802.11b/g/n||2 x 2|
|Laird||50 Series||802.11a/b/g/n||2 x 2|
|Panasonic||PAN9055/9045||802.11b/g.n||2 x 2|
|Redpine Signals||RS9113||802.11a/b/g/n||2 x 2|
|Silex||SX-PCEAC||802.11a/n/ac||2 x 2|
Figure 2: Typical Wi-Fi solutions featuring MIMO functionality
Challenges in 802.11 Wireless Communications
Over a typical 802.11g link, packets are transmit- ted with 17.5dBm of power and the sensitivity of a receiver can be as low as -76dBm for a packet-error-rate (PER) less than 10%. This is a more than billion-fold loss (93.5dB) of power over the 802.11 wireless channels. The weakening of the signal between transmitter and receiver comes from several effects. They are path loss, shadowing and multipath as illustrated in Figure 3.
As the radiated signal spreads out from a trans- mitting antenna over a wider area in the air, its power will spread out over the surface and causes the power to drop as fast as the square of the distance when the signal propagates. This is called path loss. Outdoor obstacles (like tall buildings, trees, and pools) and indoor walls can significantly attenuate the Wi-Fi signal also due to shadowing effect. The most problematic kind of fading for 802.11 is due to multipath. This causes multipath interference including constructive and destructive interference, and phase shifting of the signal. At 2.4GHz and 5GHz, RF signals bounce off ground surfaces and walls when they propagate indoors. These scatterings cause many copies of the signal to travel along many different paths. If the signals arrived at the receiver at the same time and add out-of-phase, this could cause fast signal fading. Because multipath effects depend on the phases of signals, they are strongly frequency selective. Based on this mechanism, 802.11 uses special modulation techniques called OFDM (Orthogonal Frequency Division Multiplexing) in the physical layers, which can help to mitigate the fading effect if only one single antenna is used. In 802.11a, 20MHz channel band is partitioned into 52 subcarriers shown in Figure 4, such that each subcarrier can be thought of as its own narrowband channel. As different subcarriers will experience different fading, it is very likely that some narrowband channels drop out very fast and some of them are less affected, which can then provide consistent good performance in 802.11.
|802.11a OFDM PHY Layer Parameters|
|Pilot Subcarriers||4 (BPSK)|
|Modulation||BPSK, QPSK, 16QAM, 64QAM|
|Coding Rate||1/2, 2/3, 3/4|
Figure 4: 802.11a OFDM physical parameters
MIMO Technology in 802.11
In IEEE 802.11n and 802.11ac, MIMO using multiple antennas has become an essential element of wireless communication standards. Having more than one single antenna provides extra independently faded paths, which can be exploited to improve the reliability of a Wi-Fi link to fades. There is also an increase in antenna array gain from using multiple antennas. These factors combine to enhance the data rates and improve the range.
There are three basic types of MIMO techniques that are supported in 802.11. They are spatial diversity techniques, spatial multiplexing techniques and beamforming techniques. As the beamforming techniques are not standardized how exactly it was to be implemented until 802.11ac, only spatial diversity techniques and spatial multiplexing techniques are more widely used in 802.11. Spatial diversity can improve the signal-to-noise ratio or SNR, which is defined as the ratio of signal power to the noise power. Spatial diversity is best suited for low SNR applications. Spatial multiplexing can provide more spatial streams and is better to use at high SNR.
In Single-Input Single-Output wireless systems, according to Shannon’s Law, the upper bound to the maximum data rate C of a link can be determined by the available bandwidth B and the signal-to-noise ratio β of the link.
C = B log2 (1 + β)
Figure 5 summarizes the maximum data rate for spatial diversity and spatial multiplexing respectively using multiple antennas.
|MIMO Method||Maximum Data Rate (bps)|
|Spatial Diversity (1 x N or N x 1)||B log2 (1 + Nβ)|
|Spatial Diversity (M x N)||B log2 (1 + MNβ)|
|Multiplexing (M x N)||Ns B log2 (1 + β), Ns = min (M, N)|
Figure 5: Maximum throughput for spatial diversity and spatial multiplexing. M x N represents M antennas and N antennas at the transmitter and receiver respectively.
Spatial Diversity Techniques
Spatial diversity uses multiple antennas to increase range and improve link reliability by transmitting or receiving redundant streams of information in parallel along the different spatial paths between transmit and receive antennas. Diversity techniques can be applied at the transmitter and/or the receiver side.
Transmit Diversity Techniques
As shown in Figure 6(a), two transmit antennas are sending signals to one receive antenna. This is a 2 x 1 system. The antenna at the receiver receives 2 copies of signals that are modified by a gain of hij due to the Wi-Fi channels. hij is a complex number which can represents both signal attenuation and phase shift. With transmit diversity in 802.11, the Wi-Fi transmitter with multiple antennas, like the access point, is capable of transmitting signals that have been modulated with identical information to one receiver. To improve the reliability of the link, the transmitter simply switches and selects the single best transmit antenna that can send the signal to the receiver with highest SNR. The other transmit diversity technique with better performance is called pre-coding. This is a pre- processing technique. Instead of selecting only one best antenna, the transmitter encodes and weighs all the signals across transmitting antennas in such a way that the received signals can add up constructively at the receiver. Of course the transmitter must know the channel state information (CSI) beforehand in order to select the best single transmit antenna and perform the pre-coding.
Consider the pre-coding process on a 2 x 1 diversity arrangement illustrated in Figure 7, assuming the channel state information represented by matrix He has been estimated and known. If the estimated matrix He is very close to real matrix H and the transmitted signals S are pre-coded with estimated channel gains, the received signals at the receiver can be easily detected and recovered in Gaussian noise environment and the signal processing work at the receiver can be significantly reduced. It is important to point out that the discussion above doesn’t lose the generalization of the pre-coding method when the transmitter has more than two antennas.
There is another transmit diversity method called space-time codes which is also very popular. In comparison to the pre-coding method, space-time codes are less effective in performance when more than two antennas are used at the transmitter which is beyond the scope of this paper.
Receive Diversity Techniques
Figure 6(b) is a 1 x 2 receive diversity system. Similar to transmit diversity systems, to improve the robustness of a Wi-Fi link, the simplest way of utilizing the multiple receiver antennas is just to use the antenna with the highest SNR. One disadvantage of this method is that it wastes the power of the received signal by the unused antenna. The other better method is called maximal-ratio combining (MRC). The principle is to use and combine the two received signals constructively. As is shown in Figure 8, the two received signals, A∠Θ1 and B∠Θ2 are processed at receiver and scaled in amplitude and delayed in phase until they are added in phase.
In this way, both of the received signals are well utilized and not wasted at the cost of increased hardware complexity and power consumption at the receiver. Here two receive antennas are used for illustration purposes; this also applies to the case when the receiver has more than two antennas.
Spatial Multiplexing Techniques
Consider the spatial multiplexing arrangement in Figure 6(c). Spatial multiplexing takes advantage of two independent spatial paths to transmit and receive two independent streams of information at the same time over the same frequencies. The two receiving antennas decode the signal and combine them constructively. Comparing to diversity techniques simply boost the SNR, spatial multiplexing can increase the maximum data rate significantly according to Shannon’s Law.
If channel state information is not known beforehand, the simplest spatial multiplexing method is to transmit spatial streams with equal power and modulate each stream at the same rate. In this case, the maximum data rate is not optimal. If the transmitter knows the channel state information and can track the channel variations, maximum data rates can be improved by allocating most of the transmit power to the best Wi-Fi channels at the cost of weakening the rest. In theory, this method can approaches the maximum data rates by selecting modulation and coding rates separately for each spatial stream. In practice, 802.11 uses unequal modulation rate but uniform coding rate across streams; and it is very complex and expensive to maintain and allocate different power level across different transmit antennas. This will definitely limit the extent to which pre-coded MIMO can be optimized.
Number of Antennas: the More the Better?
Using multiple antennas in 802.11 can help boost the data rate and increase the distance of the Wi-Fi link. Over a wide range of SNR values, the data rate always improves with an increase in the number of antennas according to Shannon’s Law. So, is it always better to use as many antennas as possible in Wi-Fi design?
The answer is no. The great performance improvement on throughput and range is always achieved at the cost of more advanced RF antenna systems, more complex signal process- ing units, and more powerful microcontroller units. Larger board areas, faster RF switches, additional phase shifters, and more costly power amplifiers all contribute significantly to the extra costs in implementing Wi-Fi MIMO systems. Power consumption will be much higher than single antenna Wi-Fi modules due to much longer “On” time when the MCU must stay awake. With that being said, it is always a tradeoff between performance and cost. So,
you can hardly see Wi-Fi modules designed with more than four antennas. Most of the time, Wi-Fi MIMO modules implemented with 2 x 2 and 3 x 3 configuration are more commonly supported by commercial Wi-Fi modules for space, cost and power reasons.