Performance Of Minimum Mean Square Error Research Articles

MMSE analysis of certain large isometric random precoded systems

Linear precoding consists in multiplying by an N/spl times/K matrix a K-dimensional vector obtained by serial-to-parallel conversion of a symbol sequence to be transmitted. In this paper, new tools, borrowed from the so-called free probability theory, are introduced for the purpose of analyzing the performance of minimum mean-square error (MMSE) receivers for certain large random isometric precoded systems on fading channels. The isometric condition represents the case of precoding matrices with orthonormal columns. It is shown in this contribution that the signal-to-interference-plus-noise ratio (SINR) at the equalizer output converges almost surely to a deterministic value depending on the probability distribution of the channel coefficients when N/spl rarr/+/spl infin/ and K/N/spl rarr//spl alpha//spl les/1. These asymptotic results are used to analyze the impact of orthogonal spreading as well as to optimally balance the redundancy introduced between linear precoding versus classical convolutional coding, while preserving a simple MMSE equalization scheme at the receiver.

Performance of MMSE Receiver Based CDMA System with Higher OrderModulation Formats in a Fading Channel

This paper studies the effect of using higher order modulation formats on the performance of minimum mean-squared error (MMSE) receiver based direct-sequence (DS) code-division multiple access (CDMA) systems at different loading levels in additive white Gaussian noise (AWGN) and slow fading channels. The performance of BPSK, QPSK, and 16QAM modulation formats are compared and analytical and simulation results are presented in terms of the bit error rates (BER) for these different modulation formats. A comparison of the rejection of the near-far effects for each modulation scheme is also presented. The main contribution of this paper is in showing that user capacity may be increased by using higher order modulation schemes to cause the MMSE receiver to operate away from the interference limiting region. In particular it is shown that under high loading levels, 16QAM outperforms QPSK and BPSK for identical bandwidth and information rate, while at moderate loading levels, QPSK represents the best option. A combination of pilot symbol assisted modulation (PSAM) and linear prediction are used to estimate the fading process. A general structure of the MMSE receiver capable of demodulating a wide range of digital modulation formats in this type of environment is presented.

Linear MMSE multiuser receivers: MAI conditional weak convergence and network capacity

We explore the performance of minimum mean-square error (MMSE) multiuser receivers in wireless systems where the signatures are modeled as random and take values in complex space. First we study the conditional distribution of the output multiple-access interference (MAI) of the MMSE receiver. By appealing to the notion of conditional weak convergence, we find that the conditional distribution of the output MAI, given the received signatures and received powers, converges in probability to a proper complex Gaussian distribution that does not depend on the signatures. This result indicates that, in a large system, the output interference of the MMSE receiver is approximately Gaussian with high probability, and that systems with MMSE receivers are robust to the randomness of the signatures. Building on the Gaussianity of the output interference, we then take the quality of service (QoS) requirements as meeting the signal-to-interference ratio (SIR) constraints and identify the network capacity of single-class systems with random spreading. The network capacity is expressed uniquely in terms of the SIR requirements and received power distributions. Compared to the network capacity corresponding to the optimal signature allocation, we conclude that at the cost of transmission power, the gap between the network capacity corresponding to optimal signatures and that corresponding to random signatures can be made arbitrarily small. Therefore, from the viewpoint of the network capacity, systems with MMSE receivers are robust to the randomness of signatures.

Recursive filtering and smoothing for reciprocal Gaussian processes with Dirichlet boundary conditions

The minimum mean square error (MMSE) estimation problem for a discrete-index reciprocal Gaussian process impaired by additive white Gaussian noise is completely solved in the general case of noisily observed Dirichlet random boundary conditions. Finite sets of recursive equations are obtained for the computation of the filtered sequence and of the fixed-point, fixed-interval, and fixed-lag smoothed sequences. Recursive expressions are also given for the MMSE performance of the presented estimators. Simulation results confirm the validity of the performance analysis.

Adaptive filtering using filter banks

This paper derives a theory for adaptive filters which operate on filter bank outputs, here called filter bank adaptive filters (FBAFs). It is shown how the FBAFs are a generalization of transform domain adaptive filters and adaptive filters based on structural subband decompositions. The minimum mean-square error performance and convergence properties of FBAFs are determined as a function of filter bank used. A parametrization for a class of FIR perfect reconstruction filter banks is derived which is used to design FBAF's having optimal error performance given prior knowledge of the application. Simulations are performed to illustrate the derived theory and demonstrate the improved error performance of the FBAFs relative to the LMS algorithm, when prior knowledge is incorporated.

Reduced-complexity spatial and temporal processing of underwater acoustic communication signals

Multichannel processing of high-speed underwater acoustic communication signals requires computationally intensive receiver algorithms. The size of adaptive filters, determined by the extent of ocean multipath, increases with signaling rate and limits system performance through large noise enhancement and increased sensitivity of computationally efficient algorithms to numerical errors. To overcome these limitations, reduction in receiver complexity is achieved by exploiting the relationship between optimal diversity combining and beamforming. Under relatively simple conditions, two adaptive receivers, one based on diversity combining which does not rely on any spatial signal distribution, and the other based on optimal beamforming, are shown to achieve the same performance. The beamforming approach, however, leads to a receiver of lower complexity. Carrying these observations over to a general case of broadband transmission through an unknown channel, a fully adaptive receiver is developed which incorporates a multi-input multi-output, many-to-few combiner, and a reduced-complexity multichannel equalizer. Receiver operations are optimized jointly to ensure minimum mean-squared error performance of data detection. Results of processing experimental shallow water data demonstrate the capability to fully exploit spatial diversity of underwater multipath while keeping the complexity at an acceptable level.

Quadratic power spectrum estimation with orthogonal frequency division multiple windows

Studies a new quadratic power spectral density estimator implemented with orthogonal frequency division multiple windows. The windows are constructed from a single window shifted in frequency. The orthogonality constraint ensures that the estimator meets a minimum mean-squared error performance criterion. Solutions for the window are obtained numerically. The new estimator has good frequency selectivity characteristics, and its statistical properties are comparable to the best multiple window estimators available.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A More Generalized Wiener Filtering Technique

Wiener filtering is a classical technique of signal estimation. It was initially applied to one-dimension continuous signals. Pratt extends it to time sampled signals. In this paper, we start from a more generalized Wiener filtering model to obtain optimum design of the system with minimum mean-square error. The analysis shows that in plottina optimum design more generalized Wiener processing produces the same minimum mean-square error performance as Wiener filtering. The more generalized Wiener filtering processing is also a system model in which the discrete signals, assumed to be composed of signal and noise components, pass through the filtering. The paper gives deduction and discussion on the optimum filtering design formula of the system.