Our scientific community faces a sort of paradox. A large bulk of work has been done on data-oriented techniques devised to improve peer reputation and knowledge extraction from data, so as to improve trustworthiness of digital services involving coordination and cooperation among heterogeneous peers. But, perhaps surprisingly, to the best of our knowledge, such techniques have rarely been applied to the (for our own community, crucial) process of reducing noise in the process of peer reviewing our own papers. Goal of this work is to provide initial insights on the applicability of methodologies and tools from inferential statistical to the field of peer review quality control. Our contribution is threefold. First, we propose a statistical model where each technical program committee member (reviewer) is characterized as random noise added to the “actual” value of the paper. Second, we provide an iterative data-oriented approach based on Expectation-Maximization devised to estimate mean value and variance of the noise added by each reviewer; our approach uses only the ratings provided by the reviewers themselves and does not rely on any additional source of a-priori knowledge. Third, we make use of the estimated mean values and variances to improve the accuracy of paper's evaluation and ranking.

Starting from the definition of mutual information, one promptly realizes that the probabilities inferred by Bayesian tracking can be used to compute the Shannon information between the state and the measurement of a dynamic system. In the Gaussian and linear case, the information rate can be evaluated from the probabilities computed by the Kalman filter. When the probability distributions inferred by Bayesian tracking are nontractable, one is forced to resort to approximated inference, which gives only an approximation to the wanted probabilities. We propose upper and lower bounds to the information rate between the hidden state and the measurement based on approximated inference. Application of these bounds to multiplicative communication channels is discussed, and experimental results for the discrete-time phase noise channel and for the Gauss-Markov fading channel are presented.

Numerical upper and lower bounds to the information rate transferred through the additive white Gaussian noise channel affected by discrete-time multiplicative autoregressive moving-average (ARMA) phase noise are proposed in the paper. The state space of the ARMA model being multidimensional, the problem cannot be approached by the conventional trellis-based methods that assume a first-order model for phase noise and quantization of the phase space, because the number of state of the trellis would be enormous. The proposed lower and upper bounds are based on particle filtering and Kalman filtering. Simulation results show that the upper and lower bounds are so close to each other that we can claim of having numerically computed the actual information rate of the multiplicative ARMA phase noise channel, at least in the cases studied in the paper. Moreover, the lower bound, which is virtually capacity-achieving, is obtained by demodulation of the incoming signal based on a Kalman filter aided by past data. Thus we can claim of having found the virtually optimal demodulator for the multiplicative phase noise channel, at least for the cases considered in the paper.

In this paper we focus on a single carrier pilot-assisted transmission scheme where one pilot symbol is periodically inserted in the transmitted sequence on a time-division multiplexing basis. A new equalization scheme, where the knowledge of pilot symbols is exploited by the equalizer to generate an estimate of the noise affecting the symbol to be detected, is introduced and analyzed. The criterion used to compute the equalizer coefficients is the minimization of the mean-square error (MSE). The main new result of our analysis is that the optimal pilot aided equalizer (PAE) can be decomposed as the cascade of an unconstrained minimum MSE (MMSE) linear equalizer (LE) and a data- aided noise estimation filter. This result completes and extends the noise-predictive view of decision feedback equalization to general data-aided equalization. The PAE is compared here to the MMSE- LE and to the MSE decision feedback equalizer on two frequency selective wireless channels.

The hybrid decision feedback equalizer (DFE) is a combined time-frequency domain implementation of the conventional time-domain DFE that is able to provide a good trade-off between performance and computational complexity in single carrier transmission over severely frequency-selective channels. In the hybrid DFE the implementation of the feedforward filter is done in the frequency domain, while the feedback filter (FBF) is implemented in the time-domain. The computation of the coefficients for the two filters is usually done in the same domain where they are implemented. A method for frequency-domain computation of the FBF is proposed in the paper. As is known, the key operation in the computation of the FBF is the spectral factorization. In the paper it is proposed to adopt the (cepstral) method for spectral factorization due to Kolmogoroff, which can be efficiently implemented by using the fast Fourier transform (FFT). The application of the method is considered for highly dispersive channels. By using simulations we show that for this type of channels the performance of the proposed method is virtually the same as that obtained by using time-domain approaches. The advantage of the proposed approach is that the efficient FFT gives a substantial reduction of complexity compared to time-domain methods.

The paper deals with carrier recovery based on pilot symbols in single-carrier systems. The system model considered in the paper includes the channel additive white noise and the phase noise that affects the local oscillators used for up/down-conversion. Wiener's method is used to determine the optimal filter in estimation of phase noise assuming that a sequence of equally spaced pilot symbols is available. Our analysis allows to capture the cyclostationary performance of the estimate, a phenomenon that is not considered in the previous literature. In the paper, closed-form formulas for the transfer function of the optimal filter and for the mean-square phase error are derived for the case where the phase noise is modelled as random phase walk. For this case, a suboptimal filter is proposed. Numerical results are presented to substantiate the analysis.

A new scheme of synchronous CDMA is introduced in the paper. The new scheme is based on a code made by all the cyclical translations of a basic sequence having constant amplitude and white discrete spectrum. Such a code is proposed here for the first time as a code for CDMA. According to the proposed scheme, a cyclic prefix is appended to the multiplexed signal. The proposed scheme has a property that none of the known CDMA schemes has: in a multipath environment, it allows multiuser interference to become cyclic intersymbol interference. Noticeably, the memory of the finite state machine that describes the ISI model is equal to memory of the multipath channel. The main advantage of our proposed scheme is that optimal and suboptimal detectors can be obtained from detectors proposed in the past for the ISI channel, which are much easier to implement than conventional multiuser detectors of classical CDMA schemes. Another advantage of our scheme is that it leads naturally to a signal processing architecture similar to that of OFDM systems, hence based on the efficient FFT/IFFT algorithm.

In this paper two symbol-level soft-decision decoding algorithms for Reed-Solomon codes, derived form the ordered statistics (OS) and from the generalized minimum-distance (GMD) decoding methods, are presented and analyzed. Both the OS and the GMD algorithms are based on the idea of producing a list of candidate code words, among which the one having the larger likelihood is selected as output. We propose variants of the mentioned algorithms that allow to finely tune the size of the list in order to obtain the desired decoding complexity. The method proposed by Agrawal and Vardy for computing the error probability of the GMD algorithm is extended to our decoding methods. Examples are presented where these algorithms are applied to singly-extended Reed-Solomon codes over GF(16) used as outer codes in a 128-dimensional coded modulation scheme that attains good performance, with manageable decoding complexity.

The design of the loop filter for the linearized model of the discrete-time phase-lock loop is investigated in the paper. Phase noise, which always affects oscillators, is modelled as a stationary random waveform and is taken into account in the design. Being the phase noise a predictable random process, optimization of the loop filter is worked out in the light of linear prediction. The main result of the paper is that the optimal loop filter is the causal and minimum phase factor of the spectrum of phase noise plus white additive noise. From this result they descend a lower bound on the variance of the phase error and a closed-form expression for the optimal bandwidth of the loop. Exploiting our analysis, we show that the popular perfect second order loop is optimal in tracking oscillators affected by random frequency walk plus random phase walk. Also, similarities and differences between the discrete-time PLL and the continuous-time PLL are investigated in the paper.

This paper presents an efficient implementation of a frequency domain linear equalizer for WCDMA systems. A significant complexity reduction with respect to traditional equalizers can be achieved by means of frequency domain block processing. After deriving the analytical formulation of the equalizer coefficients, it is shown that the performance of the frequency domain equalizer (FDE) is substantially equivalent to that of the classical time domain equalizer (TDE) over a wide range of operating conditions so that no penalty has to be paid for the complexity reduction. Starting from a pure parallel implementation of the FDE, it is shown that block processing for multi-code downlink WCDMA can be conveniently performed in a serial fashion, through properly reusing small processing elements while exploiting the pipeline in order to reduce the subsequent latency overhead.

The authors present a method for computing the probability of undetected error for hybrid ARQ over the Gilbert-Elliott channel (Gilbert, E.N., Bell Syst. Tech. J., vol.39, p.1253-65, 1960; Elliott, E.O., Bell Syst. Tech. J., vol.42, p.1977-97, 1963). The method described by Elliott is extended to the case where the code is used for error detection and error correction simultaneously (hybrid ARQ, type-I). This analytical approach to the evaluation of the probability of undetected error allows the throughput to be calculated as a function of the code rate and length, and systems to be designed that maximize the throughput, given the channel parameters and the desired undetected error probability.

Turbo equalization is an iterative equalization decoding scheme for detecting encoded data transmitted over channels that introduce intersymbol interference. In such a scheme, the complexity of the optimal soft-input soft-output (SISO) equalizer, which implements the BCJR a posteriori probability computation algorithm, is one of the major concerns. We extend the reduced state sequence detection (RSSD) algorithm to a BCJR-SISO equalizer matched to a non binary modulation. The resulting reduced-state SISO (RS-SISO) algorithms can take full advantage from the flexibility offered by RSSD in reducing the number of trellis states. However, extension of RSSD algorithm for obtaining the RS-SISO equalizers requires attention in the recombination procedure of the forward and backward metrics.

Decision feedback equalization is a popular method for signal detection that allows a good trade off between complexity and performance. The complexity of the decision feedback equalizer (DFE) is mainly concentrated in the feedforward filter, which is often realized in FIR form. When the number of taps of the FIR filter needed to obtain nearly optimal performance becomes large, one may take advantage of the efficiency of the FFT/IFFT algorithm realizing the feedforward filter in the discrete frequency domain. In this paper we consider the minimum mean square error (MMSE) DFE with feedforward filter in the discrete frequency domain, assuming perfect knowledge of a static channel. We move from the observation that the optimal MMSE-DFE feedforward filter is IIR, and point out that the classical scheme of unconstrained frequency domain filtering with overlap/save digital signal processing may induce substantial increase in the error rate when the approximation of the linear convolution between the received signal and the mentioned IIR, to a circular convolution is poor. Our main finding is that the error rate can be significantly improved by deliberately augmenting the power of the noise in the computation of the transfer function of the frequency domain feedforward filter.

For multiple-input multiple-output (MIMO) spatial multiplexing systems the complexity of the maximum likelihood detector (MLD) can be prohibitively extensive when the number of transmitting antennas and constellation points is high. To simplify the MLD many linear and nonlinear techniques have been proposed. In this paper the principle of reduced state sequence detection, based on mapping by set partitioning, is applied to perform detection in MIMO systems. Simulation results are presented for the proposed suboptimal detection algorithm.

The design of transmit finite-impulse response (FIR) filters with few coefficients in frequency division multiplexing data transmission systems is considered. In these systems, quality objectives are imposed for transmission over channels affected by additive white Gaussian noise (AWGN) and over channels affected by AWGN plus adjacent channel interference (ACI). The goal of this letter is: given that an adaptive receive filter, possibly cooperating with an adaptive decision feedback equalizer, is used for the AWGN channel and for the AWGN plus ACI channel, what is the best fixed FIR transmit filter to use for both channel cases? This goal is achieved by optimizing the compromise between the performance of the system on the two mentioned channels. Also, the advantages of the proposed design over a rival method based on a fixed receive filter are demonstrated.

This correspondence presents performance analysis of symbol-level soft-decision decoding of q-ary maximum-distance-separable (MDS) codes based on the ordered statistics algorithm. The method we present is inspired by the one recently proposed by Agrawal and Vardy (2000), who approximately evaluate the performance of generalized minimum-distance decoding. The correspondence shows that in our context, the method allows us to compute the exact value of the probability that the transmitted codeword is not one of the candidate codewords. This leads to a close upper bound on the performance of the decoding algorithm. Application of the ordered statistics algorithm to MDS codes is not new. Nevertheless, its advantages seem not to be fully explored. We show an example where the decoding algorithm is applied to singly extended 16-ary Reed-Solomon (RS) codes in a 128-dimensional multilevel coded-modulation scheme that approaches the sphere lower bound within 0.5 dB at the word error probability of 10/sup -4/ with manageable decoding complexity.

The authors propose and analyze a reduced complexity ordered statistics decoding algorithm for maximum-distance separable (MDS) codes. The proposed algorithm provides a high degree of flexibility in the maximum number of candidate code words tested by the algorithm which yields different trade-offs between complexity and performance between orders of reprocessing. An upper bound to the error probability is calculated by extending the method proposed by D. Agrawal and A.Vardy (IEEE Trans. Inf. Theory, vol.46, p.60-83, 2000) for generalized-minimum-distance decoding. We present an application to singly-extended Reed-Solomon codes over GF(16) in a 128-dimensional multilevel coded modulation scheme that approaches the sphere lower bound within about 0.5 dB for a word error rate of 10/sup -4/ with manageable decoding complexity.

A new technique for the control of the peak power in OFDM modulation, called pulse superposition, is presented. The proposed technique adds to the original OFDM signal a peak-reducing signal composed of Gaussian impulses centered on the peaks of the OFDM signal in the time domain. The technique combines effectiveness and low complexity, requires only a modest amount of side information and does not lead to signal-to-noise ratio reduction. In this paper, simulation results are presented.

The authors describe a multilevel coded modulation scheme that has been designed for being implemented on FPGA. The goal of the design is to obtain good performance while keeping decoding delay and computational complexity as low as possible. This goal is obtained by a two-level scheme, where the partition chain of the 8-dimensional Gosset lattice is associated to a combination of 16-ary convolutional and block codes, at the first and second level respectively. The motivations behind the design choices are illustrated and the implementation on FPGA is presented. The authors conclude by analyzing the performance of the system and by showing experimental results.

In digital radio transmission over frequency selective channels, the minimum mean-square error decision feedback equalizer (MMSE-DFE) is widely recognized as an efficient equalization scheme. In order to compute the coefficients of the feedforward and feedback filters of the MMSE-DFE, both the channel impulse response (CIR) and the variance of the noise have to be estimated. The estimate of the CIR is usually performed by the standard least-square method, where a known training sequence is employed. Then, the estimated CIR is used to evaluate the variance of the noise. The use of the two above estimates for MMSE-DFE is studied. In particular, an unbiased estimate of the variance of the noise is described.

In signal equalization, a technique that allows reduction of the number of states of the Viterbi detector is the delayed decision feedback sequence detector (DDFSD). In order to achieve good performance, it is essential to operate, before the DDFSD, an appropriate prefiltering of the received sequence. This paper is devoted to performance evaluation of the DDFSD when the feedforward filter of a minimum mean square error decision feedback equalizer is adopted as a prefilter. A truncated version of the union bound is used to approximate the bit error rate. The analysis includes a method for determining the error events that dominate the bound.

In signal equalization, a detection technique that allows reduction of the number of states of the Viterbi (1979) detector is the delayed decision feedback sequence detector (DDFSD). In order to achieve good performance, it is crucial to operate an appropriate prefiltering of the received sequence before the DDFSD. The main novelty of the paper is performance evaluation of the DDFSD when the feedforward filter of the minimum mean-square error decision feedback equalizer (DFE) is adopted as prefilter. The union upper bound is used to evaluate the probability of first error event and truncation of the sum appearing in the bound to the error sequences that dominate the performance is discussed. It is also shown that the feedforward filter of the minimum mean-square error DFE leads to maximum likelihood sequence detection with a minimum number of states, which seems to be a novel result.

In signal equalization, a technique that allows reduction of the number of states of the Viterbi (1979) detector is the delayed decision feedback sequence detector (DDFSD). In order to achieve good performance, it is crucial to operate an appropriate prefiltering of the received sequence before the DDFSD. This paper is devoted to performance evaluation of the DDFSD when the prefilter of a minimum mean square error decision feedback equalizer is adopted. The union upper bound is used to evaluate the probability of first-event error, and truncation of the sum to the error sequences that dominate the performance is discussed.

In signal equalization, a suboptimal technique for reducing the number of states of the Viterbi detector is the delayed decision feedback sequence detector (DDFSD). In order to achieve good performance a mean-square (MS) prefilter is employed before the DDFSD. This paper is devoted to investigate the sensitivity of the MS-DDFSD to a noisy estimate of the variance of the noise. Analytical performance evaluation is addressed. A truncated version of the union bound is used to approximate the bit error rate. The analysis includes a method for determining the terms that dominate the bound.

The paper deals with the design of suboptimal receivers for data transmission over frequency selective channels. The complexity of maximum likelihood sequence estimation (MLSE) turns out be exponential in the channel memory. Hence, when dealing with channels with long memory, suboptimal reception must be considered. Among the suboptimal methods, the delayed decision feedback sequence estimation (DDFSE) plays an important role. This receiver is based on a Viterbi processor where the channel memory is truncated. The memory truncation is compensated by a per-survivor decision feedback equalizer. The DDFSE was originally proposed by Duel-Hallen and Heegard (1989), where the whitened matched filter was considered as a front-end. Our contribution is to extend the principles of MLSE and DDFSE to the case where the mean-square whitened matched filter is adopted as a front-end. Simulation results show that our proposed design of the DDFSE gives substantial benefits when a severe frequency selective channel is considered.

A common model for distributed detection systems is that of several separated sensors each of which measures some observable, quantizes it, and communicates to a fusion center the quantized observation. The fusion center collects the quantized observations and takes the decision. The article deals with the design of the quantizers and of the fusion center under a rate constraint. The system of interest allows soft nonbreakpoint quantizers and nonindependent observations. Our finding is that locally optimal design of the distributed detection system is feasible via alternate minimization of the average misclassification risk.

This paper deals with two innovative schemes of digital adaptive pre-correction for the high power amplifier (HPA) nonlinearity in OFDM transmission systems. In order to evaluate pre-correction performance in terms of BER, total degradation, and normalized power spectral density, a simulator has been implemented including actual solid state power amplifier (SSPA) and travelling wave tube amplifier (TWTA) characteristics. The results thus obtained are compared with limit performance achievable with the clipper (perfectly linear amplifier with saturation). Finally, based on the pre-correction gain parameter, the considerable performance improvement for the pre-corrected system comparing with the non-pre-corrected transmitter is pointed out.

In pattern classification, a decision rule is a labeled partition of the observation space, where labels represent classes. A way to establish a decision rule is to attach a label to each code vector of a vector quantizer (VQ). When a labeled VQ is adopted as a classifier, we have to design it in such a way that high classification performance is obtained by a given number of code vectors. In this paper we propose a learning algorithm which optimizes the position of labeled code vectors in the observation space under the minimum average misclassification risk criterion.

To evaluate an upper bound on error probabilities of signal constellations used for transmission over the additive white Gaussian noise (AWGN) channel, enumeration of all the constellation intradistances is required. These may be infinite in number, for example, when convolutional codes are used and the constellations are lattices. Truncation of the series does not necessarily provide a bound anymore, and must be done with care. Yet the union bound is very simple, as it does not require any further knowledge about the signal constellation than the distance enumerator. In this paper, we describe some methods that can be used to evaluate error probabilities of infinite signal constellations, and that require only a finite number of terms. These methods are applicable, for example, to convolutional codes decoded with a finite-depth Viterbi algorithm and to signal constellations carved from lattices. Coded modulations based on lattices and convolutional or block codes can also be dealt with. As an example of application, we analyze a variable-rate 3-stage coded modulation encoder/decoder, which has been built and is based on a combination of convolutional codes with a single-parity-check block code.

Multipath fading is one of the major impairments encountered in terrestrial digital radio. A common countermeasure to limit the outage time due to multipath is the space diversity technique, which takes its effectiveness from the low correlation between the field samples of two well separated antennas. A very simple and effective blind receiver is proposed for the diversity radio channel. The novelty of the present study is the application of the constant modulus algorithm to joint clock recovery and baseband combining. The effectiveness of our proposal relies upon the synergic action of clock recovery and adaptive baseband combining, which allows optimal equalization of the two-ray diversity channel.

We describe computational techniques for the evaluation of error probability of modulation schemes based on binary lattices and multilevel coding. These techniques stem from the representation of a binary lattice in the form of a trellis whose structure depends on the block codes on which the binary lattice is based. Multilevel coded-modulation schemes can also be described through a trellis; by combining with this the trellis structure of the lattice constellation we obtain a single trellis that can be used for maximum-likelihood decoding using the Viterbi algorithm and for computing the error probability. An upper bound to error probability can be obtained from the transfer function of the trellis with suitable labels, and is applicable to both maximum-likelihood and staged decoding.>

Multipath fading is one of the major impairments encountered in terrestrial digital radio. A common countermeasure to limit the outage time due to multipath is the space diversity technique, which takes its effectiveness from the low correlation between the field samples of two well separated antennas. A very simple and effective blind receiver is proposed for the diversity radio channel. The novelty of the present study is the application of the constant modulus algorithm to joint clock recovery and baseband combining. The effectiveness of our proposal relies upon the synergic action of clock recovery and adaptive baseband combining, which allows optimal equalization of the two-ray diversity channel.

Optimum decision methods are employed to discriminate between in-sync and out-of-sync operations in Viterbi decoders when the received data sequence is given in analog form (soft decisions). The exact solution to this discrimination problem, however, appears too complex in many practical cases. A suboptimum algorithm is then proposed that can be implemented as a hardware device, external to the decoder. This algorithm is recognized as a modified version of an earlier technique known as the "reencoding method". Analysis and simulations indicate that the modification produces remarkably good results not only in comparison with the reencoding method but also with previous techniques discussed in literature.>

LVQ1 is a supervised fine tuning of the Self-Organizing Map, introduced by Kohonen for the classification task. In this paper we present a study about LVQ1. We show that it implements the search of certain saddles on a scalar cost function. The above criterion does not guarantee convergence to a stable solution, as enlightened by a simple example. We also compare the performance of LVQ1 to that of a labeled version of the Self-Organizing Map.>

This paper describes a design method for the synthesis of digital filters in data communication systems. The coefficients in the transmit and receive filters are optimized to meet specified requirements on (i) out-of-band emission, (ii) intersymbol interference, (iii) adjacent channel interference and (iv) thermal noise immunity.>

We study the construction and decoding of binary multilevel coset codes. This construction, originally introduced by Blokh and Zyablov in 1974 and by Zinov'ev in 1976, shows remarkable analogies with most recent schemes of coded modulations. Basic elements of the construction are an inner code, head of a partition chain having suitable distance properties, and a set of outer codes, generally nonbinary. For each partition level there is an outer code whose alphabet has the same order of the partition: in this way it is possible to associate every partition subset to a code symbol. It is well known that these codes can be efficiently decoded by the so called "multistage decoding." We show that good codes (in terms of performance/complexity) can be constructed using Reed-Muller (RM) codes as inner codes. To this aim RM codes are revisited in the framework of the above construction and decoding techniques. In particular we describe a family of decoders for RM codes which include Forney's (1988) and Hemmati's (1989) decoders as special cases. Finally, we present some examples of efficient binary codes based on RM codes, and assess their performance via computer simulation.>

A rotation detection algorithm for solving the phase ambiguity problem in trellis-coded modulations is proposed. This algorithm is simple to implement and may be an efficient alternative to rotationally invariant codes. Its performance has been investigated by simulation and appears promising.>