Skip to main content
eScholarship
Open Access Publications from the University of California

Link adaptation schemes for MIMO wireless systems

Abstract

Multiple input multiple output (MIMO) systems can significantly increase channel capacity, especially in power, bandwidth, or complexity limited systems. Link adaptation techniques, where signal transmission parameters such as the modulation and coding rate are dynamically adapted to the changing channel states, also are powerful tools for increasing the system spectral efficiency. Hence, adaptive MIMO systems are emerging as one of the key techniques for easing the bottleneck of traffic capacity in future Internet-intensive wireless networks. We first examine the traditional role of multiple antennas in a CDMA uplink, combating fading and reducing multiple access interference. Specifically, we assume the fades of the antennas in the receive array are correlated, which reduces the diversity gain against fading without affecting the array's capability for interference suppression. Assuming perfect channel knowledge available at the transmitter, Maximal Ratio Transmission (MRT) is employed to weight the transmitted signal optimally in terms of combating signal fading. At the receiver, adaptive beamforming reception is adopted to both suppress MAI and combat the fading. We evaluate the antenna array performance with joint fading reduction and MAI suppression. Among numerous space-time coding techniques, the Vertical Bell-Laboratories Layered Space- Time (V-BLAST) scheme has been adopted for 4G systems since it exhibits the best tradeoff between performance and complexity. Here we present a practical implementation of a V-BLAST type system, in which the MIMO open-loop capacity can be closely approached by using adaptive modulation with appropriate channel codes and optimum successive detection (OSD). First, the constellation is selected based on the instantaneous capacity of each channel realization. Then the density evolution technique is employed to determine the maximal achievable rate of an LDPC code for each transmit antenna for each channel realization, at a given SNR. If the fading process is non- ergodic, the outage capacity corresponding to a given outage probability is used to measure the channel performance. As an example, we design the LDPC codes for an adaptively modulated 2 x 2 V-BLAST system to approach its outage capacity for a given outage probability. Since the constellation size and channel code rate have to match each channel realization, codes of different rates and block lengths are required for different transmissions. Flexible-length rate-compatible punctured irregular repeat -accumulate (IRA) codes are introduced to accomplish this goal. We propose a two-step shortening and puncturing process to obtain codes of different rates and block lengths from one underlying IRA mother code while satisfying constraint imposed by the chosen modulation alphabet, fixed frame length (in symbols) and the target code rate. A key advantage of this approach is that the optimality of the degree distribution is maintained in the shortening process. Further, good performance of codes with different rates is guaranteed by optimizing the shortening and puncturing distributions. The shortening step has to preserve the code rate and information node degree distribution while reducing the mother code to the target block length. Higher rate codes can be obtained by puncturing the shortened mother code according to the optimal puncturing distributions of the information bits and the parity bits

Main Content
Current View