Search

In this paper, we present a cross evaluation of different a priori SNR estimation methods as well as different time-frequency analysis processing using a subjective listening test. The noisy signal is corrupted by different types of background noise i.e. babble and pink, and varying levels of input SNR (0 dB and 10 dB). The signals are processed using Short Time Fourier Transform (STFT) or Critical Band (CB) processing. After estimating the clean speech signal, it was presented to 10 participants for evaluation using a subjective listening test according to (ITU-TP.835) methodology. The results demonstrate that the participants preferred the speech signal processed using CB for low SNR levels and non-stationary background noise, which means that critical band based frequency scale is more useful in adverse noisy conditions.

Broadband beamformers are known to be sensitive not only to mismatches in microphone characteristics but also in the acoustic environment. In this work, we propose a robust broadband beamformer design using weighted least square optimization technique, which takes into account the room reverberation as well as the mismatches in microphone gain and phase characteristics. The robust design procedure optimize the mean performance of the beamformer with the room impulse response estimated using the image source method. Design examples demonstrate performance improvement in terms of significant reduction in error sensitivity for the robust design formulation.

In adaptive feedback cancellation (AFC) methods, the step-size plays an important role in controlling the convergence speed of an adaptive filter in the feedback canceller path. The selection of this step-size provides a compromise between a low steady-state error and a fast convergence rate. The use of a variable step-size (VSS) is a potential solution to achieve both fast convergence and low steady-state error. In this paper, we propose a frequency-domain AFC method which integrates an improved practical VSS (IPVSS) algorithm into a partitioned-block frequency-domain (PBFD) implementation of the prediction error method (PEM) for hearing aid applications. The proposed method derives benefit from the IPVSS algorithm, e.g., a better compromise solution between convergence and steady-state error, and from the PBFD-PEM, e.g., a low numerical complexity and an improved convergence. The proposed method is evaluated for different types of speech incoming signals as well as for a sudden change of the acoustic feedback path. Simulation results show that the proposed method provides a significant performance improvement compared to the PBFD affine combination approach as well as the PBFD using either the upper or the lower step-sizes which are utilised as boundaries in the IPVSS algorithm.

The affine projection algorithm (APA) is commonly used for adaptive filtering in acoustic echo cancellation (AEC) due to its higher convergence and tracking rate compared to the conventional normalized least mean squares (NLMS) algorithm, especially for spectrally colored incoming signals. However, its application to adaptive feedback control (AFC) in hearing aids (HAs) is still not common because of the inherent correlation between the loudspeaker and incoming signals as well as the increase in computational complexity. In this paper, we investigate a way to employ a low projection order APA in conjunction with an improved practical variable step-size (IPVSS) for the prediction error method (PEM) based adaptive feedback control in HAs. The proposed approach is evaluated for a speech incoming signal and for a sudden change of the acoustic feedback path. The experimental results show that the proposed approach yields much better performance than a system employing either the upper or the lower fixed step-size used in the IPVSS as well as the PEM using the IPVSS for the NLMS algorithm (PEM- IPVSS-NLMS). By choosing a small projection order for the APA there is only a slight increase in the computational complexity. Moreover, a small amount of frequency shifting (FS) is integrated to further improve the system's performance.

In this paper, a modified a priori SNR estimation method with application in speech enhancement is presented. The decision directed (DD) approach for the a priori SNR estimation has been the most popular method due to its ability to eliminate musical noise in speech enhancement. However, a limitation of this method is the slow tracking of speech onsets that generates a transient distortion to the speech. The proposed method modifies the DD a priori SNR estimator by employing an adaptive smoothing factor based on the instan-taneous a posteriori SNR and an adaptive sigmoid function to obtain an improved tracking of speech onsets. Experimental results using several instrumental measures show the ability of the proposed method to preserve weak speech components when compared to two reference methods while maintaining the advantage of the DD approach in eliminating the musical noise.

The paper proposes a convex combination fusion function based on a sigmoid function for the estimation of the a priori SNR in a speech enhancement framework with critical frequency band processing. The proposed method does not only eliminate the one frame delay generated by the well-known decision directed approach but also increases the adaptation speed during abrupt changes in the SNR estimation. As a result, the advantage of low musical noise has been maintained while more weak speech components have been preserved. Experimental results using instrumental and subjective measures also indicate improvement in speech quality compared to the reference methods.

The proportionate normalized least-mean-squares (PNLMS) algorithm is commonly used in acoustic echo cancellation (AEC) context. It provides faster initial convergence and tracking rates compared to the NLMS algorithm for the case of sparse echo impulse responses. The improved PNLMS algorithm (IPNLMS) has been proven to be more powerful than PNLMS by exploiting new rules for computing the weight of each step-size corresponding to each adaptive filter coefficient. However, the application of the PNLMS and the IPNLMS algorithms for adaptive feedback control (AFC) in hearing aids (HAs) is still limited due to high correlation between the loudspeaker and incoming signals. This paper proposes implementations of the PNLMS/IPNLMS algorithms for AFC using the prediction error method (PEM) for hearing aids. The proposed methods have been evaluated for both speech and music incoming signals. Simulation shows that the proposed methods have faster initial convergence and tracking than the PEM using the NLMS algorithm (PEM-NLMS).

Variable step-size (VSS) schemes are popular to use in acoustic echo cancellation (AEC) contexts. However, their effective implementation in adaptive feedback cancellation (AFC) for hearing aids is still challenging due to the correlation between microphone and loudspeaker signals. We propose an improved practical VSS algorithm (IPVSS) which uses a variable step-size with upper and lower limits to control the update equation of an adaptive filter. The proposed algorithm is implemented for feedback cancellation using the prediction error method. As a result, the overall system has a fast convergence rate, a high tracking rate and a low steady-state error. The performance of the proposed approach has been evaluated for both speech and music incoming signals. The simulation results show that the proposed approach outperforms a system which only utilizes either the lower or the upper fixed step-size used in the IPVSS.

Prediction error method (PEM) is popularly applied to acoustic feedback cancellation (AFC) in hearing aids. Commonly, this method uses normalized least mean square (NLMS) adaptive filter to estimate the coefficients of the real feedback path. A disadvantage of NLMS algorithm is to provide a slow convergence rate when coloured incoming signals are used. To address this problem, we propose a simple but effective way to apply an affine projection algorithm (APA) to acoustic feedback cancellation using PEM. Performance of the proposed method is evaluated for speech incoming signal in both cases of using/not using a probe noise. Simulation results show that the proposed method outperforms the PEM using NLMS in both terms of misalignment and added stable gain.

The limiting feature of hearing aids is acoustic feedback. This feedback problem has been approached by using an acoustic feedback canceller. This well known identification in a loop problem, will give a biased solution due to correlation between the desired and loudspeaker signals. Speech and music signals have long tails in the correlation. As a result, the performance of the system is considerably degraded and under certain conditions the cancellation system will be unstable. The two-microphone techniques have the potential to significantly reduce this problem. This paper introduces the applications of uniform and non-uniform sub-band techniques into the two-microphone acoustic feedback cancellation (AFC-2mics) to decorrelate input signals and individually adapt solutions in those bands. The system has been evaluated in terms of Misalignment (MisAL) and Maximum Stable Gain (MSG) using both male and female speech input signals. The simulation results show that our proposed methods provide better performance than the other existing methods.

The paper presents results of an investigation of a MU-MIMO system suitable for providing broadband services. The base station has a circular array and each user has a single high gain antenna. Two optimum zero forcing beamformers are considered and semi-orthogonal user selection is used to schedule users in the downlink. A key point of the investigation is to determine the maximum number of users that can be scheduled for simultaneous transmission in order to achieve high spectral efficiency and still maintain acceptable performance in terms of sum rate and also all user rates.

Affine projection algorithms (APA) have been widely employed for acoustic echo cancellation (AEC) since they provide a natural trade-off between convergence speed and computational complexity. However, their application in Acoustic Feedback Cancellation (AFC) in hearing aids so far has been limited due to lack of performance improvement in one microphone settings. A two microphone technique was recently proposed to provide an improved cancellation for a larger class of input signals. This paper proposes to use APA in the two microphone closed-loop feedback cancellation. It is shown that APA has significantly improved the misalignment and maximum stable gain of the system. Moreover, a new variable Gaussian step-size control (VGSS) for APA is also proposed. The simulation results indicate an improvement in convergence speed of the proposed algorithms as compared to previously suggested methods.

This paper presents a design of microphone array system that is intended for conference room applications. Different microphone array configurations have been considered in this work to demonstrate the importance of array layout. The filter-and-sum broadband beamformer is designed by using weighted least square (WLS) optimization method. As existing near-field broadband beamformers are sensitive to changes in microphone characteristics and position, robust designs need to be considered that include microphone characteristics mismatch (gain and phase) in the design. Furthermore, it is shown that the aperture size is important for the designed beamformer performance. Golden section search technique is used to find optimal aperture size and weights. The important roles that microphone array layout, robust design and aperture size play in the design of the broadband beamformer are demonstrated.

Zero-forcing beamforming (ZFBF) under perantenna power constraint (PAPC) is a popular research topic. A survey is carried out in this paper to show the important results in the literature in this area. Particularly, it has been found that there are two research lines for this topic. One research line is to find the optimal solutions based on the interior point method. Another line is focused on the low complexity and parallel implementations. Important results are presented for both lines.

Zero-forcing beamforming (ZFBF) is a popular pre-coding scheme for MIMO systems. Most of the studies in the literature are under total power constraints. However, the perantenna power constraints (PAPC) are more realistic. The state-of-the-art method is interior point method which is expensive to realize in practice due to the high computational complexity. Hence, a low complexity zero-forcing precoding scheme under the per-antenna power constraints is proposed in this paper. This is achieved by introducing a regularized dual method. Simulations are carried out to show the effectiveness of the proposed method, which has a low computational complexity. In addition, the algorithm can be implemented in parallel to further reduce the computational complexity.

In industrial noise environments, the use of assistive listening headsets is a means to provide adequate access to voice communication while wearing hearing protection. This paper presents a performance evaluation and comparison of two different methods to provide the binaural speech enhancement in real industrial noise scenarios. The investigated binaural methods based on differential beamforming and multichannel Wiener filter show different strengths and weaknesses. A transient noise suppression algorithm is also proposed and evaluated. Performance evaluation shows that this algorithm, together with the binaural multi-channel Wiener filter approach, can successfully reduce the hammering noise. This can be observed from the PESQ scores and the signal characteristics.

This paper extends the design of frequency invariant beamformer by incorporating robustness in the design formulation to cater for array element mismatches and other non-ideal characteristics. In this approach, a 2-norm constraint on the filter weight is imposed into the design optimization problem. The constraints limit the norm of the beamformer coefficients and thus reduce the sensitivity of the beamformer towards the non-ideal characteristics of the microphones. Design examples and comparisons are presented to illustrate the effectiveness of our approach.

This paper presents an adaptive coherence-based beamforming algorithm that enhances a target speech signal from an arbitrary direction of arrival (DOA) in the azimuthal plane, while localising and suppressing the interfering speech signal and noise from other directions. In these complex noisy acoustic environments, spatially selective speech enhancement schemes provide significant advantages over single channel enhancement methods. Experimental results demonstrate that the proposed method achieves considerable improvement in Perceptual Evaluation of Speech Quality (PESQ). Moreover, the proposed algorithm is directly suited for real-time implementation.

The real-time implementation of the existing multi-channel Wiener filter (MWF) algorithms suffer from performance degradation due to the lack of robustness against estimation errors of the second-order statistics. The reasons are twofold: one, the estimation of the statistics relies on real voice activity detector (VAD), which often fails in adverse environments. Second, the MWF solutions involve estimation of the second order clean speech statistics, which also exaggerates the errors. This paper presents an MWF algorithm that requires neither VAD nor clean speech statistics. Performance evaluation under real scenarios shows that the proposed method outperforms the conventional MWF solution in terms of the trade-off between noise reduction and speech distortion.

In this paper, a modified a priori SNR estimator is proposed for speech enhancement. The well-known decision-directed (DD) approach is modified by matching each gain function with the noisy speech spectrum at current frame rather than the previous one. The proposed algorithm eliminates the speech transient distortion and reduces the impact from the choice of the gain function towards the level of smoothing in the SNR estimate. An objective evaluation metric is employed to measure the trade-off between musical noise, noise reduction and speech distortion. Performance is evaluated and compared between a modified sigmoid gain function, the state-of-the-art log-spectral amplitude estimator and the Wiener filter. Simulation results show that the modified DD approach performs better in terms of the trade-off evaluation.

A noise estimation algorithm is proposed for single channel speech enhancement. By comparing the noise estimate with the short term noise and speech at every time frame, the noise estimate is efficiently updated by using a fixed step-size. The step size is optimized based on the speech quality performance and the noise tracking capability. The proposed technique is capable of tracking noise spectrum variations, while remaining robust to the speech onsets. In addition, the noise estimator requires low computational complexity, which makes it effective for real time implementation in battery operated equipment. Simulation results show that the proposed method can achieve good speech quality and efficient noise tracking performance when compared to existing methods.

Large area (LA) sequences form a class of ternary spreading sequences which exhibit an interference free window. In addition, these sequences have low correlation properties outside this window. Work to date has concentrated on examining the parameters and performance of LA sequences with reference to only a small number of example families. In this paper we develop general conditions which an LA family must satisfy. The development of these conditions allows for the production of computationally efficient tests to determine whether a given family is an LA family. In particular, these tests can form the basis for algorithms to construct LA families, allowing for a larger number of families with the potential for higher energy efficiency than those of previous work.

Multirate signal processing is important in a wide range of signal processing applications. Drawbacks when using multirate processing are mainly related to aliasing, reconstruction effects and the delays. In this paper, the authors investigate the design of uniform DFT filter bank with finite precision prototype filters and a minimum total number of power-of-two for the coefficients. The design problem is formulated as a mixed integer optimization problem. This problem can then be solved by using e.g. the mean field annealing algorithm. Design examples show that the filter bank can be designed with less total number of power-of-two than the quantized filter bank while achieving approximately the same performance

This paper investigates the problem of enhancing a desired speech source corrupted by a stationary noise. A beamformer structure is proposed which combines an adaptive constrained beamformer with a post-filtering technique. The adaptive constrained beamformer is designed to spatially enhance the desired source from the observed signal. After the adaptive constrained beamformer, a post-filter is employed to increase noise suppression capability. Experimental results in a real car situation show that the proposed structure achieves a good noise suppression level with a low distortion.

Sensor arrays are used in various applications such as radar, sonar, communications, radio links and voice input systems. In this paper, we propose a new variable array structure, which allows array characteristics such as the look direction or the passband/stopband region to be varied easily. Most importantly, these characteristics can be changed by varying only a single steering or tuning parameter. The design problem is formulated to minimize the integral squared error between the array response and the desired response, which is reduced to a quadratic optimization problem. Design examples are presented to show that the look direction of the array can be designed to effectively steer over the angles of the required range.

The number of spreading sequences required for direct sequence code division multiple access (DS-CDMA) systems depends on the number of simultaneous users on the channel. The correlation properties of the sequences used affect the bit error rate of the system. Often a sequence family provides more sequences than are required and in many cases the selection of the employed sequences is a computationally intensive task. In this paper, a branch and bound algorithm is presented to optimise the subset of available sequences, given the required subset size. In contrast to previous approaches, the resulting subset is guaranteed to be optimal. Numerical results are presented to demonstrate the improved performance of this algorithm over previous work.

This paper presents a unifying technique for robust feedback filter design in the decision feedback equaliser (DFE). The robustness is needed for combatting error propagation effects which is the major limitation when using DFEs. Numerous constraint strategies for the design of robust DFEs are available in the literature, however, they have been presented in their own ad-hoc context. Consequently there is a desire to give a unified approach to robust feedback filter design. The approach in the current presentation is to first develop a mean square error (MSE) measure which is a function of the feedforward and feedback filters. This measure is then combined with a feedback filter constraint using four existing norm constraint approaches (the L/sub 1/, L/sub 2/, L/sub /spl infin// and the "two tap" norm). The use of second order cone programming allows these existing norm constraint strategies to be unified and solved in a computationally efficient manner. The evaluation is performed with respect to the impact on error propagation, through the mean and the variance of the burst error length.

Broadband wireless systems require low complexity equalizers which can cope with long channel impulse responses and severe intersymbol interference (ISI). In this paper, a frequency domain (FD) approach for adaptive blind equalization for multiple-input multiple-output (MIMO) communication systems is proposed. As an initial step, a time domain (TD) block based updated adaptation is first developed which updates the equalizer coefficients once per each block of data symbols. As such, this algorithm can be equivalently implemented in the FD to reduce computational complexity associated with the symbol by symbol TD update. Simulation results show that the proposed algorithms can successfully separate and equalize the received signals with a significantly lower computational complexity than a corresponding TD approach

This paper investigates the application of filter bank design techniques on the DFT based transmultiplexers to ameliorate the effects of in-band aliasing and consequently improve the performance of the system. With this approach, one is allowed flexibility in the design of the synthesis and analysis filters. Hence, a suitable prototype filter can be selected and specific values assigned to its parameters, namely the cut-off frequency, main lobe width and side lobe amplitudes so as to construct an ideal filter for optimal recovery. Investigation results depict that increasing the cut-off frequency of the prototype filter by a small increment delta, enhances system performance through the widening of the stopband tolerance region which reduces the effect of aliasing. Simulations also show that using a Kaiser prototype filter with a wide main lobe and low side lobe amplitudes generates the most favorable results. Moreover, it was observed that the performance of the transmultiplexer can be improved by increasing the number of sub-carriers and oversampling factor. The evaluation of the transmultiplexer structure for this paper was assessed in terms of its bit error rate (BER) performance

This paper investigates the short-term power spectral estimation of the source of interest and the interference in noisy environment for hands-free communication. Models of probability density function for the source, interference and noise covariance matrices are proposed for the evaluation of the power spectral parameters. The analysis are performed to obtain the Cramer-Rao bounds for the evaluation and an exact maximum likelihood (ML) estimation is formulated based on the proposed model. Evaluation of an adaptive beamformer with ML estimation shows significant suppression levels for noise and interference whilst maintaining a low source signal distortion

The paper proposes a new adaptive algorithm for speech enhancement in hands-free communication systems. The scheme aims to enhance a target signal corrupted by background noise and by acoustic echo. The main idea is the incorporation of both the source and the interference short-term spectral amplitudes in the solution. These spectral components are estimated iteratively to provide a spectrally optimized constraint while the background noise statistics are estimated and updated continuously. Evaluations in a hands-free car environment show that the proposed algorithm is superior in suppressing both the noise and the interference even in a double-talk situation.

The paper presents a new adaptive beamformer employing updates for both the source and interference short term power spectral densities (PSDs). These PSD updates track the variations in the spectral content, thereby yielding a spectrally optimized constraint for each time instant. Consequently, probability density function models for the source, interference and noise covariance matrices are proposed for the estimation of the PSD parameters. An optimization problem is formulated based on the proposed model. Evaluation in a real car environment shows significant suppression levels for the noise and interference by employing the proposed beamformer whilst maintaining low source signal distortion.

The performance of direct sequence code division multiple access (DS-CDMA) systems depends on the autocorrelation and cross-correlation properties of the deployed spreading sequences. Since good autocorrelation characteristics come at the expense of the cross-correlation properties and vice versa, a combination of these two measures needs to be optimized when designing a DS-CDMA system. In this paper, we consider the design of complex-valued spreading sequences with optimized correlation characteristics. In particular, the maximum nontrivial aperiodic correlation values are used to specify the cost or fitness function for the optimization problems. A genetic algorithm is presented for the design of polyphase sequences, namely, Oppermann sequences and modified Walsh-Hadamard sequences. It can be seen from these applications that the genetic algorithm is well suited to efficiently design polyphase sequences especially when the number of parameters for the optimization problem is large.

We investigate the design of complex-valued spreading sequences with respect to a combination of different correlation properties. The two classes of complex sequences considered are the Oppermann sequences, which offer a wide range of correlation properties, and the modified Walsh-Hadamard sequences, which have been shown to offer good correlation properties. Since the number of parameters for the optimization problem is large, especially for the modified Walsh-Hadamard sequences, it is difficult if not impossible to use global optimization methods for solving such problems. Thus, we propose to transform the problem with continuous variables to another problem with discrete variables. This problem can then be solved efficiently using a genetic algorithm. These types of algorithms have been successfully applied in various areas, such as neural networks; however, to the authors best knowledge, their use for the design of complex spreading sequences has been rather sparse.

In this paper we introduce periodic Oppermann sequences, which constitute a special class of polyphase sequences. The properties of these sequences are presented, and indicate that periodic Oppermann sequences are suitable for combination to generate families of longer sequences. Numerical examples show that periodic Oppermann sequences can be designed for ranging or synchronisation applications or for supporting multiple access spread spectrum communication systems.

Numerous spread spectrum applications require families of long spreading sequences, often used at very high chip rates. In this paper, an algebraically simple way of generating long sequences by combining shorter polyphase sequences is presented, aimed at asynchronous spread spectrum systems. The approach is motivated by the fact that polyphase sequences offer increased design options, in terms of the supported range of correlation characteristics, while combination sequences allow for simpler generation of long sequences. It leads to the definition of combination Oppermann sequences, the properties of which are investigated in this paper. Numerical results indicate that the families of these proposed combination sequences provide favourable aperiodic correlations. The presented family of combination Oppermann sequences is therefore suitable for applications that rely on rapid synchronisation and are required to provide multiple access to the system

This paper presents a new space constrained adaptive beamformer employing an updated source power spectral density (PSD). The space constraints are used to capture the target signal spatially and to provide robustness against steering error vectors. The PSD update on the other hand ensures that the spectral information of the desired source is reflected continuously on the space constraints. As such, target signal extraction can be achieved with minimum distortion. The beamformer operates in a subband structure to allow time-frequency operation for each channel, yielding a combination of weighted spatial and temporal filters. Evaluations on real car data show that the proposed algorithm significantly improves the speech intelligibility with noise suppression level up to 21 dB.

In this paper, we present a new subband adaptive beamformer equipped with noise statistics updates. These updates are employed to effectively estimate and track the noise statistics continuously in the solution. Additionally, an update on the source power spectral density (PSD) is incorporated to enhance the timbre of the source of interest. Furthermore, the beamformer is also equipped with a space constraint on the source area to provide robustness against steering vector errors and good capture of the target signal spatially. The entire processing can be viewed as an efficient combination of weighted spatial and temporal filters. Evaluations on real car data with variations in the car noise level show that the proposed scheme achieves a good noise suppression level up to 20 dB.

The performance of a code division multiple access system depends on the correlation properties of the employed spreading code. Low cross-correlation values between spreading sequences are desired to suppress multiple access interference. An autocorrelation function with a distinct peak enables proper synchronization and suppresses intersymbol interference. These requirements, however, contradict each other and a trade-off needs to be established. In this paper, a global two dimensional optimization method is proposed to minimize cross-correlation with auto-correlation allowed to lie within a fixed region. This approach is applied to the design of subsets of complex spreading sequences.

Time domain (TD) adaptive equalization based on the modified recursive least squares (RLS) algorithm has been proposed for multiple-input multiple-output (MIMO) systems. The paper proposes frequency domain (FD) adaptive equalization to reduce the computational complexity associated with the fast converging RLS algorithm. The effect of correlation between different sub-channels of the MIMO systems is also studied. The performance in terms of bit error rate (BER) and the complexity of the FD approach is compared with that of the TD approach. The FD adaptive equalization can obtain significant computational saving over the TD approach with approximately the same BER performance even for a small number of equalizer coefficients.

The performance of a code division multiple access (CDMA) system depends on the correlation properties of the employed spreading sequences. In this paper, we investigate the performance of polyphase spreading sequences in an asynchronous direct sequence (DS) CDMA system. Especially, the equivalent odd and even correlation (EOE) sequences and the Oppermann sequences are examined. Their performance in terms of symbol error rate (SER) on additive white Gaussian noise (AWGN) channels and multiplicative Rayleigh fading channels are compared. Simulation results show that the Oppermann sequences with optimized cross-correlation characteristics outperform the EOE sequences on an AWGN channel and perform slightly better on a Rayleigh fading channel.

We consider a class of FIR filters defined by the first order difference routing digital filter (DRDF) structure and sums of two powers-of-two coefficients. A novel design method is developed for constructing high quality filters with reference to the min-max error criterion. This method is highly efficient in terms of computational time. Simulation studies show a large improvement over existing methods such as quantization. In some cases, the peak ripple magnitude over the stop and pass bands is reduced by up to 13 dB over the quantization method. These results are achieved even for cases involving small number of delays.

In wireless communication, multiple versions of the transmitted signals arrive at the receiver with different attenuations and time delays. Thus, an equalizer with long filter is required at the receiver to reverse the multipath effects. In this paper;, a new non-causal delayless subband equalizer structure is proposed to reduce the computational complexity of the equalizer and to improve the channel tracking capability. This structure avoids the additional delays usually associated with the subband schemes. A design method is formulated to minimize the aliasing effect of the filter bank in the subbands. The significance of employing a non-causal equalizer is also emphasized. Simulation results show that the performance of the new structure is significantly improved by using the design method. Moreover, the complexity of the new subband equalizer is a fraction of the equivalent fullband at the expense of small degradation in the performance.