On the optimality and performance of transmit and receive space diversity in MIMO channels

Abstract

The issue of channel state information at the transmitter is investigated using MIMO channel measurements and by deriving expressions for ergodic and outage capacity in a Rayleigh fading channel. Expressions for bit error rates in Rayleigh fading channels are also presented for orthogonal space time block codes and for beamforming where the bit error rates in the beamforming case follow from the distribution of the largest eigenvalue to Wishart matrices. We demonstrate by measurements that Rayleigh fading is a valid assumption in non line of sight channels, although a Nakagami-m distribution showed to be a more appropriate distribution model in both line of sight and NLOS environments. It was also demonstrated that channel state information at the transmitter is less useful in high-SNR scenarios but is more useful in line of sight channels compared to non line of sight environments.

Statistical Analysis of MIMO Beamforming With Co-Channel Unequal-Power MIMO Interferers Under Path-Loss and Rayleigh Fading

“…In the case of STBC, the numerator follows a chi-square distribution with degree , which corresponds to the diversity gain that the desired received signal can obtain. For an ( -Tx, -Rx) MIMO channel, MBF can also achieve the maximum spatial diversity gain of order [24]- [26]. Hence, for an ( -Tx, -Rx) antenna configuration, both STBC and MBF are equivalent to a receive diversity system with receive antennas.…”

“…Because the power of every decoupled channel is proportional to its corresponding eigenvalue in the Wishart matrix, we define the percentage of energy achieved by the channel with gain as (26) where are the nonzero eigenvalues of the Wishart matrix . Using (1), (19), and (23), the denominator in (26) is (27) where denotes the trace of a matrix.…”

Performance analysis of the maximum-SNR design in Rayleigh fading MIMO channels

“…(14) for CYM = 1/2 and PM = 1 is equivalent to the BPSK closed-form presented in [8], [4]. The analytical form of the polynomial cpn(z) was not provided in [SI; only the rational polynomial was given for some ( n~, n~) .…”

“…In other respects, the possibility to (coherently combine more signals at the receiver leads to the maximum combining gain. For a given diversity order, the best final performance Figure 4 plots SEP for a (8,8) system with different modulations and constellation sizes. It shows that the gap between the curves at moderate and high SNR is close to the modulation gain (16).…”

“…However, GM remains a valid comparison factor even for a large system (e.g. (8,8)), which has the exact SEP close to its asymptote at very high SNR.…”