Wireless hits data rate barrier

Wireless hits data rate barrier
Broadband wireless access systems are faced with a major challenge: to provide subscribers with high data rates over nonline-of-sight fading channels at wireline reliability. The requirement for high reliability arises from performance expectations set by competing broadband technologies-namely, cable modems and digital subscriber lines, both of which operate over wires.

It is well documented that first-generation, fixed-broadband wireless access (FBWA) technologies require line of sight in order to approach wireline reliability. The line-of-sight (LOS) requirement brings with it a fundamental set of shortcomings that limit deployments to a few carefully selected markets. And these shortcomings must be overcome prior to commencing mass deployments. Some of the barriers to mass deployment encountered with first-generation systems are as follows:

• High customer acquisition costs ($1,200 to $1,600 per subscriber).

• Limited scalability (limited spectrum-reuse capabilities).

• Poor service reliability (requires LOS).

• Inadequate coverage (30 to 40 percent service coverage in some markets).

The first step in creating mass-deployable systems is to move the customer premises equipment (CPE) down off the roof to "under the eaves" for businesses and to bring it indoors for residential subscribers. A self-install indoor CPE dramatically reduces subscriber acquisition costs by making truck rolls obsolete.

By definition, mass deployment means that a system's capacity must scale with the growing customer base. Since a wireless Internet service provider's spectrum is always at a premium, it must be reused over and over in a cellular architecture to keep up with demand. Basically, efficient spectrum reuse requires interference mitigation or suppression. Interference mitigation on both uplink and downlink requires a multiantenna basestation and CPE.

As a direct consequence of moving the CPE off the rooftop, links between the basestation and CPE are no longer line of sight, requiring a system capable of non-LOS operation.

What exactly is non-LOS? The term has been explicitly defined by the six IEEE 802.16a channel models. Each of these models quantifies identifying non-LOS characteristics such as Rician K factor and delay spread. The channel models range from near line of sight (IEEE.16a-1) to pure non-LOS (IEEE.16a- 4-6).

Rician K factor is defined as the ratio of the power in the specular (LOS) to the power in the diffuse (multipath) component. The higher the K factor, the more LOS the link. (A K factor of infinity means the link is LOS, a K factor of zero means the received signal contains no LOS component.) Delay spread results from a signal's multiple reflections arriving with disparate delays due to the different path lengths traveled en route from the transmitter to the receiver-about 5 microseconds for every "extra" mile traveled.

A key advantage to spatial multiplexing: user data rates are doubled, thus increasing link capacity and spectral efficiency. Source: IOSPAN

In LOS links, the received signal power is predictable and its variation around a nominal level-in time and space-is minimal. Thus only minimal margins for wireline availability are necessary. Margins are defined as the extra power necessary to overcome signal fades when they occur.

In near-line-of-sight links, received signal strength starts to vary in space and in time due to multiple signal reflections from buildings, cars, trees and other topographic features that add together in constructive as well as destructive fashion. The amount of received signal strength variation depends on the degree of "nonline of sightedness." An outage occurs when the received signal on an antenna falls below a predetermined level, so extra transmit power is required to guarantee a particular level of availability.

True non-LOS links exhibit much deeper fades. For single-input, single-output channels under non-LOS, the random received-signal level is said to be Rayleigh-distributed. Rayleigh statistics tell us that 10 percent of the time a signal is 10 dB below, or effectively one-tenth of, its nominal level. To guarantee that a 10-dB fade can be overcome, 10 times more transmit power needs to be sent from a transmitter-that is to say, 10-dB margins are required. Rayleigh statistics also tell us that 1 percent of the time a signal is 20 dB below, or 1/100 of, its nominal level.

Typically, the received signal level in fixed wireless links varies slowly with time (exhibits low Doppler). The consequences are that when a signal goes into a fade, it tends to stay there long enough to cause an outage. From this, one can deduce that without spatial diversity, 99 percent availability would translate into requiring 100 times the transmit power. In like manner, 99.9 percent availability would require 30 dB, or 1,000 times the transmit power without adequate diversity.

This much extra power is rarely available; therefore, the key to wireline availability in non-LOS links is spatial diversity. On the receive end, spatial diversity is where a signal is simultaneously sampled and combined at multiple locations in space. On the transmit end, a signal is simultaneously launched out of multiple locations in space.

One can conclude that multipath fades are drastically reduced when signals are aggregated from multiple locations in space (diversity gain) and coherently combined (array gain). Basically, multiple locations in space are much less likely to simultaneously experience a multipath fade, resulting in diversity gain.

"How far apart do these antennas have to be?" is probably a question that comes to mind. The minimum recommended separation is a function of the scattering environment and is thus related to antenna height. Measurements indicate that at a basestation situated at a relatively high elevation (> 30 meters) the requirement is approximately 10 wavelengths. At the CPE, which is situated relatively low (1 to 2 meters) and in the midst of a scattering environment, half a wavelength is sufficient. Diversity is therefore most readily attained at the multiantenna CPE.

See related chart

Having multiple antennas at both the basestation and the CPE has a multiplicative effect on the amount of diversity-a phenomenon referred to as "hyperdiversity." For example, a multiple-input, multiple-output (MIMO) system delivers sixth-order diversity.

Hyperdiversity's reduced multipath-fade probability lowers the margins required for a given availability. Lower margins render better coverage and extend range.

See related chart

Successful communications require an acceptable signal-to-noise ratio. For example, the received signal must be above the noise floor by a sufficient amount such that the information contained in the signal can be discerned by the receiver. This presents a challenge in a non-LOS fading environment because the received signal power fluctuates, yet the noise floor is assumed to be constant. To combat this phenomenon and deliver wireline availability, MIMO employs transmit diversity. Identical user data is sent out of the multiple-transmit antennas. Multipath fading is thus substantially reduced, virtually eliminating the multipath-fade margins.