In OFDM systems the oscillators used at the digital to analog converter (DAC) and analog to digital converter (ADC) to derive sampling instants at the transmitter and receiver will never have exactly the same period. Thus sampling instants slowly shift relative to each other. In this paper we investigate the performance of a frequency domain re-sampler for a MIMO OFDM system which was developed for a proposed IEEE 802.11n TGn proposal. We present analysis and simulation to demonstrate that this technique results in negligible performance loss relative to perfect sample timing. In addition, the complexity of the technique is compared to a competing technique based upon a variable digital delay filter (Farrow filter). It is shown that the frequency domain re-sampler results in a lower complexity design. Also memory requirements for the design are causal and very low.

We give a simple and complete characterization of the optimum transmitter correlation matrix for multiple-input-multiple-output (MIMO) systems, when antenna subset selection is used. It is assumed that the receiver feeds back to the transmitter only the selected set of antennas without providing any kind of rank ordering of the selected antennas. It turns out that the optimal correlation matrix in this case must be a convex combination of an identity matrix and a matrix whose elements are all ones. This reduces the problem of finding the actual optimum to a one-dimensional search in any given scenario.

Time-domain channel estimation and detection techniques are presented for multicarrier signals in a fast and frequency-selective Rayleigh fading channel. As a consequence of the time-varying channel, the orthogonality between subcarriers is destroyed in conventional frequency-domain approaches, resulting in interchannel interference, which increases an irreducible error floor in proportion to the normalized Doppler frequency. An important feature of the proposed technique is the ability to exploit the time-selective channel as a provider of time diversity. This enables us to achieve performance superior to any other structure without increasing bandwidth or incorporating redundancy, in order to reduce the complexity of the estimator, we apply the theory of optimal low rank approximation to a minimum mean squared error channel estimator and present a theoretical calculation of mean squared error and simulations to confirm that the estimator is robust to changes in channel characteristics.

Convergence analyses for the least mean square algorithm with update delay (DLMS) exist, but most of them are based on the unrealistic independence assumption between successive input vectors. We consider the DLMS algorithm with decreasing step size /spl mu/(n)=a/n,a>0 and prove the almost-sure convergence of the algorithm under the mixing input, satisfying of the law of large numbers, and uniformly bounded input assumptions.

In several practical applications of the LMS algorithm, including certain VLSI implementations the coefficient adaptation can be performed only after some fixed delay. The resulting algorithm is known as the delayed LMS (DLMS) algorithm in the literature. We consider a normalized version of this algorithm with a decreasing step size, /spl mu/(n), and discuss the almost sure convergence of the non-homogeneous algorithm, assuming a mixing input condition and the satisfaction of a certain law of large numbers.>

When a delay of even one iteration is introduced in the coefficient update loop, the projection properties of the NLMS (normed least mean square) algorithm are lost, allowing the error vector to increase as well as decrease in any given update. This makes the analysis of the algorithm with delay much more difficult. An exact analysis of the delayed update algorithm, on a sample function basis, is developed. It is shown that for any delay, the gain parameter, can be chosen sufficiently small to guarantee exponential convergence, assuming only that the input satisfies the standard mixing condition.>

Consideration is given to the stability properties of the conventional continuous-time LMS algorithm. The algorithm is investigated for the case of stationary ergodic inputs, and a necessary and sufficient condition for exponential almost-sure convergence is presented. This condition is shown to be less restrictive than the well-known persistency of excitation condition. Also, the authors point out and clarify an apparently common error regarding the connection between persistency of excitation and positive-definite autocorrelation in stationary ergodic vector waveforms.>