Capacity of multiple-antenna fading channels: spatial fading correlation, double scattering, and keyhole

The capacity of multiple-input multiple-output (MIMO) wireless channels is limited by both the spatial fading correlation and rank deficiency of the channel. While spatial fading correlation reduces the diversity gains, rank deficiency due to double scattering or keyhole effects decreases the spatial multiplexing gains of multiple-antenna channels. In this paper, taking into account realistic propagation environments in the presence of spatial fading correlation, double scattering, and keyhole effects, we analyze the ergodic (or mean) MIMO capacity for an arbitrary finite number of transmit and receive antennas. We assume that the channel is unknown at the transmitter and perfectly known at the receiver so that equal power is allocated to each of the transmit antennas. Using some statistical properties of complex random matrices such as Gaussian matrices, Wishart (1928) matrices, and quadratic forms in the Gaussian matrix, we present a closed-form expression for the ergodic capacity of independent Rayleigh-fading MIMO channels and a tight upper bound for spatially correlated/double scattering MIMO channels. We also derive a closed-form capacity formula for keyhole MIMO channels. This analytic formula explicitly shows that the use of multiple antennas in keyhole channels only offers the diversity advantage, but provides no spatial multiplexing gains. Numerical results demonstrate the accuracy of our analytical expressions and the tightness of upper bounds.

[1]  A. Goldsmith,et al.  Capacity of Rayleigh fading channels under different adaptive transmission and diversity-combining techniques , 1999, IEEE Transactions on Vehicular Technology.

[2]  A. Robert Calderbank,et al.  Space-Time Codes for High Data Rate Wireless Communications : Performance criterion and Code Construction , 1998, IEEE Trans. Inf. Theory.

[3]  W. C. Jakes,et al.  Microwave Mobile Communications , 1974 .

[4]  Helmut Bölcskei,et al.  Outdoor MIMO wireless channels: models and performance prediction , 2002, IEEE Trans. Commun..

[5]  Mansoor Shafi,et al.  On a Gaussian approximation to the capacity of wireless MIMO systems , 2002, 2002 IEEE International Conference on Communications. Conference Proceedings. ICC 2002 (Cat. No.02CH37333).

[6]  C. G. Khatri,et al.  Series representations of distributions of quadratic form in the normal vectors and generalised variance , 1971 .

[7]  R. Valenzuela,et al.  Capacities of multi-element transmit and receive antennas: Correlations and keyholes , 2000 .

[8]  B. V. Bronk,et al.  Exponential Ensemble for Random Matrices , 1965 .

[9]  F. R. Gantmakher.  The Theory of Matrices , 1984 .

[10]  Gerard J. Foschini,et al.  Layered space-time architecture for wireless communication in a fading environment when using multi-element antennas , 1996, Bell Labs Technical Journal.

[11]  M. J. Gans,et al.  On Limits of Wireless Communications in a Fading Environment when Using Multiple Antennas , 1998, Wirel. Pers. Commun..

[12]  Hyundong Shin,et al.  Effect of keyholes on the symbol error rate of space-time block codes , 2003, IEEE Communications Letters.

[13]  John M. Cioffi,et al.  Spatio-temporal coding for wireless communication , 1998, IEEE Trans. Commun..

[14]  Shlomo Shamai,et al.  Spectral Efficiency of CDMA with Random Spreading , 1999, IEEE Trans. Inf. Theory.

[15]  W. L. Ferrar,et al.  Introduction to the Theory of Determinants and Matrices , 1959 .

[16]  B. K. Shah.  Distribution Theory of a Positive Definite Quadratic Form with Matrix Argument , 1970 .

[17]  A. James.  Distributions of Matrix Variates and Latent Roots Derived from Normal Samples , 1964 .

[18]  R. Muirhead.  Aspects of Multivariate Statistical Theory , 1982, Wiley Series in Probability and Statistics.

[19]  J. W. Silverstein.  Strong convergence of the empirical distribution of eigenvalues of large dimensional random matrices , 1995 .

[20]  Yu. A. Brychkov,et al.  Integrals and series , 1992 .

[21]  Georgios B. Giannakis,et al.  Outage mutual information of space-time MIMO channels , 2004, IEEE Transactions on Information Theory.

[22]  Thomas L. Marzetta,et al.  Space-Time autocoding , 2001, IEEE Trans. Inf. Theory.

[23]  Brian L. Hughes,et al.  Differential space-time modulation , 1999, WCNC. 1999 IEEE Wireless Communications and Networking Conference (Cat. No.99TH8466).

[24]  Chen-Nee Chuah,et al.  Capacity scaling in MIMO Wireless systems under correlated fading , 2002, IEEE Trans. Inf. Theory.

[25]  A. Edelman.  Eigenvalues and condition numbers of random matrices , 1988 .

[26]  Vi︠a︡cheslav Leonidovich Girko,et al.  Theory of random determinants , 1990 .

[27]  Ammar B. Kouki,et al.  New compound upper bound on MIMO channel capacity , 2002, IEEE Communications Letters.

[28]  Alex J. Grant,et al.  Rayleigh Fading Multi-Antenna Channels , 2002, EURASIP J. Adv. Signal Process..

[29]  Aris L. Moustakas,et al.  MIMO capacity through correlated channels in the presence of correlated interferers and noise: a (not so) large N analysis , 2003, IEEE Trans. Inf. Theory.

[30]  A. Rukhin.  Matrix Variate Distributions , 1999, The Multivariate Normal Distribution.

[31]  Angel E. Lozano,et al.  Effect of antenna separation on the capacity of BLAST in correlated channels , 2000, IEEE Communications Letters.

[32]  A. Robert Calderbank,et al.  Space-Time block codes from orthogonal designs , 1999, IEEE Trans. Inf. Theory.

[33]  Ingram Olkin,et al.  The 70th anniversary of the distribution of random matrices: A survey , 2002 .

[34]  Joong Soo Ma,et al.  Mobile Communications , 2003, Lecture Notes in Computer Science.

[35]  C. Khatri.  On Certain Distribution Problems Based on Positive Definite Quadratic Functions in Normal Vectors , 1966 .

[36]  J. Wishart.  THE GENERALISED PRODUCT MOMENT DISTRIBUTION IN SAMPLES FROM A NORMAL MULTIVARIATE POPULATION , 1928 .

[37]  Reinaldo A. Valenzuela,et al.  Keyholes, correlations, and capacities of multielement transmit and receive antennas , 2002, IEEE Trans. Wirel. Commun..

[38]  N. R. Goodman.  Statistical analysis based on a certain multivariate complex Gaussian distribution , 1963 .

[39]  Thomas L. Marzetta,et al.  Unitary space-time modulation for multiple-antenna communications in Rayleigh flat fading , 2000, IEEE Trans. Inf. Theory.

[40]  A. M. Mathai.  Jacobians of matrix transformations and functions of matrix argument , 1997 .

[41]  Emre Telatar,et al.  Capacity of Multi-antenna Gaussian Channels , 1999, Eur. Trans. Telecommun..

[42]  Peter Lancaster,et al.  The theory of matrices , 1969 .

[43]  Shlomo Shamai,et al.  Fading Channels: Information-Theoretic and Communication Aspects , 1998, IEEE Trans. Inf. Theory.

[44]  R. Fisher.  FREQUENCY DISTRIBUTION OF THE VALUES OF THE CORRELATION COEFFIENTS IN SAMPLES FROM AN INDEFINITELY LARGE POPU;ATION , 1915 .

[45]  V. Girko.  A Refinement of the Central Limit Theorem for Random Determinants , 1998 .

[46]  Joseph M. Kahn,et al.  Fading correlation and its effect on the capacity of multielement antenna systems , 2000, IEEE Trans. Commun..

[47]  Peter J. Smith,et al.  A determinant representation for the distribution of quadratic forms in complex normal vectors , 2000 .