MAS approach optimizes WiMAX networks

Next generation wireless networks are being developed to meet the growing demand for wireless broadband services over a wide coverage area. Improved spectral efficiency relative to current wireless deployments is a key performance requirement to meet the required economics of wireless broadband networks. Specifically, the use of two or more antennas per sector along with multi-antenna signal processing (MAS) is a key enabler to meet the spectral efficiency targets. Integrating MAS into WiMAX base station platforms can significantly reduce time-to-market and performance risks.

MAS technology is an inherent feature of next generation wireless standards such as WiMAX. The standards include features that support MAS, leaving room for equipment vendors and operators to innovate and improve on the baseline approaches. Building on top of baseline approaches that utilize the time and frequency dimensions, MAS processing includes a multitude of techniques that exploit the "spatial" dimension of the wireless signal. The approaches that are most relevant to WiMAX fall into four categories — receive diversity; transmit diversity; beamforming and interference cancellation.

Receive diversity Receive diversity is a basic MAS technique that is employed in both user and base station equipment. Desired signals received at two or more antennas are combined to improve the signal-to-noise ratio (SNR). As the antennas are separated in space, the probability that the signal to the antennas will be subjected to a simultaneous fade is reduced. Most common among the algorithms used is MRC, or maximal ratio combining, which is optimal for independent white noise channels. Improved MAS techniques can exploit diversity present in multipath fading and interference-limited channels.

Transmit diversity Transmit diversity approaches use specific coding techniques that encode the base station transmit signals to introduce additional diversity in the downlink (base station to mobile direction). Space-Time Coding (STC) is an example of such an approach. The WiMAX profile specifies two forms of STC to provide downlink diversity. MIMO-A requires two antennas at the base station and one antenna at the user device. MIMO-B specifies two antennas at the base station and TWO antennas at the user device, making it possible to spatially multiplex two data streams and achieve a theoretical doubling of data rate vs. MIMO-A. Improved MAS approaches extend MIMO to any number of base station and mobile antennas. The practical data rate that can be achieved depends on real channel conditions.

Beamforming Beamforming using multiple antennas at the base station can significantly improve uplink and downlink performance by weighting the received or transmitted signals to match the propagation characteristics of the terrain. This is typically done at the base station on a "per user" basis - to send/receive more signal to/from each mobile's specific location. More antennas provide greater benefits, up to a point of diminishing returns.

In the uplink, weights are applied to the signals received from multiple antennas prior to combining them, resulting in an enhancement of the base station's SNR. The SNR increase allows a higher modulation class to be transmitted by the user device, resulting in more bits transferred in a given channel allocation.

Transmit beamforming is typically used in the downlink. Weights are applied to the transmitted signal on each antenna such that the corresponding signals arrive co-phased at the mobile receiver, enhancing the mobile's receiver SNR. More transmit energy is focused on the target device, which indirectly reduces interference caused to other users. Beamforming can be combined with MIMO-A and MIMO-B diversity schemes or applied to single stream transmission.

Interference cancellation Related to beamforming, weights can be applied to locate signal "nulls" in cell edge regions to directly minimize interference to/from users in neighboring cells. Simultaneously maximizing the desired signal through beamforming and minimizing interference through nulling provides the greatest improvement in the signal-to-interference plus noise ratio (SINR).

Different algorithms can be used for beamforming weight computation. One technique relies on explicit estimation of the signals' direction of arrival (DoA). These approaches may perform well in line-of-sight (LOS) scenarios, but do not perform well in other propagation conditions (non-line-of-sight, or rich multipath). They also impose strict requirements on antenna array geometry and calibration precision. Improved MAS techniques can be used at the base station which perform well across a wider variety of channel conditions and have no antenna geometry restrictions.

In order to provide optimal WiMAX base station performance across different propagation scenarios and a wide variety of antenna configurations, the MAS software should reside within the physical layer of a base station. Thus, it can dynamically and individually select from a suite of proven weighting algorithms to maximize system performance by analyzing channel conditions for each user. Because the use of a single MAS approach, such as MIMO, will not be effective under all conditions, dynamic switching of MAS options is required to consistently reap the benefits of multiple antenna deployments.