We propose blind multiuser detection schemes with antenna arrays, which is based on signal subspace estimation. They are a multichannel extension of the decorrelating and minimum mean-square-error detectors, and therefore they share their immunity to near-far effects. The blind scheme may be seen as an extension of the results in of Wang and Poor (see IEEE Trans. Inform. Theory, vol.44, p.677-90, 1998). However, it is seen that compared with the latter results when spatial diversity is considered, the proposed spatio-temporal detectors offer, with little attendant increase in computational complexity, a better performance. A blind adaptive implementation based on a new orthogonal PAST (projection approximation subspace tracking) algorithm, which is shown to be efficient for subspace tracking, is proposed. Also, we develop a blind estimation of the spatial signature based on the orthogonality between noise and signal subspaces. It is seen that the blind adaptive multiuser detection and blind spatio-temporal signature estimation can he integrated jointly.

In the presence of partial-band Gaussian interference, the frequency-hopping spread-spectrum system, without and with optimum diversity, performs worse than the frequency-diversity spread-spectrum system proposed by Kaleh (see ibid., p.886-93, vol.44, no.7, 1996) The latter offers the advantage of coherent reception with coherent diversity combining.

Frequency diversity for a phase-coherent frequency-hopping spread-spectrum system is proposed. The frequency band of the communicator is partitioned into L disjoint subbands on which L replicas of the communicator signal are simultaneously transmitted; each signal replica hops independently in its subband. Therefore, both phase continuity and coherent optimal combining of the received replicas are possible. Optimum diversity is thus at least 6 dB better than the conventional noncoherent FH-SS system. We show that a smart jammer should be present in all subbands. We also give a sufficient condition that forces the jammer to spread its interference signal on the whole communicator frequency band.

A spread spectrum system to counter bandlimited Gaussian interference is proposed. The optimum receiver for this system is easy to build. Frequency diversity which allows the receiver to distinguish unjammed signal replicas from their jammed versions is used. The system can also resist bandlimited partial-time jamming. The only choice left to a smart jammer to maximize the error probability is to spread its signal like the communicator. The optimum receiver, which jointly performs symbol detection and interference rejection, is derived. Side information needed by this receiver can easily be estimated. However, if the interference bandwidth is narrow compared to the signal bandwidth, side information on noise and interference levels is not needed by a simpler and near-optimum receiver. Bit-error probability is evaluated for quaternary phase shift keying (QPSK) modulation and compared to that of direct sequence spread spectrum. We also propose the use of polyphase filters for simple system implementation.

We consider block and convolutional codes for improving the reliability of data transmission over the binary precoded noisy (1-D) partial response channel. We concentrate on a class of codes for which the maximum likelihood decoder, matched to the encoder, precoder, and the channel has the same trellis structure as the encoder. Thus, doubling the number of states due to the channel memory is avoided, We show that the necessary and sufficient condition to belong to this class is that all codewords be of the same parity. The Reed-Muller and Golay codes belong to this class.

Joint detection is a means to overcome the near-far problem and to improve the spectrum efficiency in the DS-CDMA uplink in terrestrial mobile communications. Detection requires knowledge of the channel parameters which are distinct for every user. Due to inter-user interference (IUI) conventional estimation of these parameters is unreliable if some signals are weaker than others. Consequently, the performance of a subsequent joint detector gets degraded. We study a new method to jointly estimate these parameters for fading multipath channels. The joint channel estimator uses the RAKE outputs at the symbol rate. These samples are multiplied by pilot or detected symbols. Before low-pass filtering a decorrelator removes the IUI. The resulting channel parameter estimates are shown to be independent of the power levels of the interfering users and therefore near-far resistant. Simulation results evaluating the mean-square error are given for the conventional and the proposed joint estimation of multipath channels with Doppler fading. We also investigate the influence of the estimator on the detector performance.

In a block transmission system the information symbols are arranged in the form of blocks separated by known symbols. Such a system is suitable for communication over time-dispersive channels subject to fast time-variations, e,g., the HF channel. The known reliable receiver for this system is the nonlinear data-directed estimator (NDDE). This paper presents appropriate equalization methods for this system. A nonstationary innovations representation based on Cholesky factorization is used in order to define a noise whitener and a maximum-likelihood block detector. Also block linear equalizers and block decision-feedback equalizers are derived. For each type we give the zero-forcing and the minimum-mean-squared-error versions. Performance evaluations and comparisons are given. We show that they perform better than conventional equalizers. As compared to the NDDE, the derived block decision-feedback equalizers perform better and are much less complex. Whereas the NDDE uses the Levinson algorithm to solve M/2 Toeplitz systems of decreasing order (where M is the number of symbols per block), the derived equalizers need to process only one Toeplitz system. Moreover, the Schur algorithm, proposed for Cholesky factorization allows us to further reduce the complexity.>

In code division multiple access (CDMA) systems transmitting over time-varying multipath channels, both intersymbol interference (ISI) and multiple access interference (MAI) arise. The conventional suboptimum receiver consisting of a bank of matched filters is often inefficient because interference is treated as noise. The optimum multi-user detector is too complex to be implemented at present. Four suboptimum detection techniques based on zero forcing (ZF) and minimum mean square error (MMSE) equalization with and without decision feedback (DF) are presented and compared. They combat both ISI and MAI. The computational complexity of all four equalizers is essentially the same and they are independent of the size of the data symbol alphabet. It is shown that all four equalizers considerably outperform the conventional receiver. The MMSE equalizers are superior to the corresponding ZF equalizers. Furthermore, the equalizers with DF perform better than the corresponding equalizers without DF. An algorithm termed channel sorting is introduced to reduce the impairing effect of error propagation on the equalizers with DF.>

The theory of hidden-Markov models is used to formulate an iterative method for the joint estimation of timing phase offset and the detection of symbols carried by a time-unsynchronized PAM signal block. Timing phase offset estimation is based on the maximum likelihood criterion. Posterior probabilities of symbols are calculated and used to perform symbol-by-symbol decisions minimizing the symbol error probability.>

We present an iterative method for joint channel parameter estimation and symbol selection via the Baum-Welch algorithm, or equivalently the Expectation-Maximization (EM) algorithm. Channel parameters, including noise variance, are estimated using a maximum likelihood criterion. The Markovian properties of the channel state sequence enable us to calculate the required likelihood using a forward-backward algorithm. The calculated likelihood functions can easily give optimum decisions on information symbols which minimize the symbol error probability. The proposed receiver can be used for both linear and nonlinear channels. It improves the system throughput by making saving in the transmission of known symbols, usually employed for channel identification. Simulation results which show fast convergence are presented.>

We derive a Chernoff upper bound for the pairwise error probability in the presence of an additive white Gaussian noise and a Rayleigh or Rice correlated fading. The bound is useful for situations where perfect interleaving cannot be achieved. We use it to determine some indications in the design of optimum trellis coded modulation for correlated fading channels.>

The performance of binary partial response continuous phase modulation (with index 0.5) using a differentially coherent receiver depends on the choice of the receiver filter. An optimum MMSE design method for this filter is presented. The receiver filter is equivalent to the cascade of a matched filter and an equalizer in order to reduce inherent intersymbol interference (ISI). It is shown that performance degradation with respect to that of the differential binary phase shift keying (BPSK) system is due to inherent ISI contained in the signal and also to noise enhancement and correlation caused by the receiver filter. The bit error probability on the Gaussian channel is calculated by assuming that ISI is Gaussian. The Gaussian minimum shift keying (MSK) signal is used for illustration.>

The author extends to the case of minimum-shift-keying (MSK) modulation the differentially coherent reception theory established for phase-shift-keying modulation. A novel differentially coherent detector for MSK is thus derived. The receiver filter is equivalent to the cascade of a matched filter and an equalizer in order to suppress inherent intersymbol interference. It is shown that performance can be improved when the delay between signals, multiplied by the differential detector, is increased from one to M bit time intervals. This decreases the effect of noise correlation and, thus, the bit error probability. The bit error probability of the proposed receiver is calculated. It is found that almost all potential improvement due to the delay M is obtained with M=3.>

By using a pulse-amplitude-modulation representation of binary continuous-phase-modulation signals, the authors develops a novel optimum Viterbi sequence detector and a near-optimum Viterbi receiver with low complexity. For modulation index 0.5, where a linear receiver can be used, a minimum-mean-squared-error linear receiver filter is derived. The performance of all of these is analyzed, using the Gaussian minimum-shift-keying signal (GMSK) for illustration. It is shown that a GMSK receiver consisting of two matched filters and a four-state Viterbi algorithm performs with less than 0.24-dB degradation compared with the optimal receiver. The linear receiver is optimum for all values of E/sub b//N/sub 0/ (bit-energy-to-noise one-sided spectral density ratio). A design method for its filter is given. The filter is equivalent to a cascade of a matched filter and a Wiener filter estimator. Both upper and lower bounds for the bit-error probability are calculated. Simulation results which confirm the analysis are given.>