Parametric yield estimation is a critical task for design and validation of analog and mixed-signal (AMS) circuits. However, the computational cost for yield estimation based on Monte Carlo (MC) analysis is often prohibitively high, especially when multiple circuit performances and/or environmental corners (e.g., voltage and temperature corners) are considered. In this paper, a novel statistical method named correlation-aided yield estimation (C-YES) is proposed to reduce the computational cost for parametric yield estimation. Our proposed approach exploits the fact that multiple circuit performances over different environmental corners are often correlated. Hence, we can accurately predict the performance value at one corner from the simulation results for other performances and/or corners. Based upon this observation, instead of running a large number of MC simulations to cover all performances and corners, an efficient algorithm is developed to select a small set of the most “informative” simulations that should be performed for yield estimation. Our numerical experiments show that for parametric yield estimation with multiple circuit performances and environmental corners, C-YES achieves 6.5-9.3× runtime speedups over other conventional methods.

Interference is a major limiting factor for wireless communications. When interference is independent and identically distributed (i.i.d.), its statistics is invariant with time. The conventional statistical analysis can be used to mitigate the interferences. However, when the interferences are non-i.i.d., few signal processing techniques are available for mitigating the interferences. The matched filter technology is probably the best signal processing technique. In this presentation, a least cross-correlation criterion based rotation-matched filter is presented that can mitigate the interferences and achieve the optimal signal to interference-plus-noise ratio (SINR). Based on the simulation study it is better than the matched filter by 5∼10 dB.

Identifiability is a fundamental issue in blind signal processing. It is well known that there are two types of inherent ambiguities: the scalar and permutation ambiguities. However, this is obtained under the condition that no a priori information of input signals is assumed. We show that if the information of source constellation is exploited, with a general constellation, there is an inherent discrete phase ambiguity, which is uniformly spaced on the unit circle. We also illustrate our identifiability results to blind channel estimation in orthogonal frequency division multiplexing (OFDM) systems.

Blind channel estimation is a promising technique to reduce the pilot overhead. Unfortunately, most existing algorithms suffer from the scalar ambiguity problem, and hence only achieve semi-blind identification. In this paper, we show that with the information of source constellation, the phase of the ambiguous scalar can be divided into a fractional part and an integer part. Then we propose a multiple-constellation scheme enabling totally blind identification regardless of constellation type for OFDM systems. The necessary and sufficient condition for eliminating the scalar ambiguity is given. An application example shows that our scheme can help other algorithms circumvent the annoying ambiguity.

The current multiple accesses, such as CDMA, are designed for multi-user wireless communication systems in the licensed spectrum, where uncoordinated interferences can be kept to a minimum. On the other hand, the uncoordinated interferences do exist in the unlicensed spectrum, and they compromise the orthogonal properties and hence degrade the performance of the current multiple accesses. In this presentation, the Autocorrelation-Division Multiple Access (ADMA) is presented, which achieves the best SINR among all passive pre-/post-filters, by eliminating all MAI and CCI, and retaining the signal power when the channel is lossless without enhancing the noise power.

A total rejection of interferences under any noise power is difficult, if not impossible, by current filtering technology. In this presentation, we will show that this is indeed achievable by a new filtering technique that extracts the part of the statistics from the received signals, which depend only on the interferences, not on the signals and noises. The algorithm is blind to the signal and interference channels; and hence it needs not to be updated when these channels vary with time. It is also important when the channels cannot be or is too costly to be estimated by pilot signals. The computation is based on statistics that is noise-invariant and hence its accuracy is not impacted by the noise power. Simulation confirmed what is proven that interferences are totally rejected over a wide range of noise power.

A covariance matrix based anti-interference algorithm is proposed for SIMO OFDM systems. This algorithm constructs the interference-blocking filter without the information of the inferences. This algorithm is insensitive to the time synchronization error and to the carrier frequency synchronization error. Therefore, it is especially useful for the case when interferences are strong and unknown, such as the case in the unlicensed spectrum.

A covariance-based algorithm is proposed to find the interference blocking filter for single-input multiple-output orthogonal frequency division multiplexing (SIMO OFDM) systems. The algorithm uses only the auto-covariance matrix of the received signal. It does not require precise carrier synchronization and time synchronization, so it is applicable to an environment with uncooperative high power interferences

An anti-jamming filtering technique is presented that it fully eliminates the jamming (or noncooperative) signal and its performance is noise-independent. This technique uses multiple receivers; a filtering process explores the nonoverlapping properties of the signals in the autocorrelation domain; and a matching of the input and the output statistics. A limited simulation study supports the theory.

The autocorrelation matching method is a second order statistics-based blind MIMO FIR channel equalization technique designed for wireless communication systems using multiple receiving antennas. This paper presents a closed-form solution to the optimal zero-forcing equalizer for the autocorrelation matching method. It can recover a transmitted signal with maximum SNR from an unknown MIMO FIR channel based on the autocorrelation coefficients of the transmitted signal and the autocorrelation matrices of the received signals. Numerical results obtained from intensive computer simulations are also given in this paper, which verify the effectiveness and accuracy of the closed-form solution.

Based on a new diversity, the diversity of autocorrelation, a new multiple-access scheme, called A-CDMA, is presented here. Like WCDMA, it packs high-density information into transmitted signals. Unlike WCDMA, it can block all the ISI and MAI completely under all noise level. The simulation supports the theory and shows that the total interference due to ISI and MAI is reduced to 1% when SNR varies from 1 to 11. By comparison, it outperforms WCDMA by a factor of 9 in the reduction of ISI and MAI. Its overall performance, as measured by BER is better than WCDMA in general, and by 10-20 dB when SNR ranges from 9 dB to 11 dB.

In this paper, we propose a new iterative algorithm to solve the blind deconvolution problem of MIMO-FIR channels driven by source signals which are temporally second-order uncorrelated but fourth-order correlated and spatially second- and fourth-order uncorrelated. In our new approach, to solve the blind deconvolution problem, we consider two stages: First, filtered source signals are extracted from the mixtures of source signals. Second, the source signals are recovered from the filtered source signals.

This paper presents an application of a new signal processing technique, called the autocorrelation matching (AM) method, to outdoor IEEE 802.11b wireless LAN operation. Since wireless LAN uses unlicensed spectrum, outdoor wireless LAN may suffer strong co-channel interference from each other. In addition, delay spread in the outdoor environment is comparable to symbol duration, and thus channel equalization is needed. The AM method can solve these two problems simultaneously by inserting three steps in the wireless LAN signal processing procedure: (1) filtering the transmitted sequence with a specially designed filter (called a pre-filter) before the spreading process at the transmitter; (2) matching the autocorrelation function of a composite signal constructed from multiple received signals with that of the desired signal after the de-spreading process at the receiver; and (3) filtering the composite signal with a filter that is the inverse of the pre-filter. It is proven that, regardless of the existence of co-channel interference and intersymbol interference, a match of autocorrelation functions between the composite signal and the desired signal assures the composite signal equaling the desired signal up to a phase difference, which can then be identified from the synchronization prefix to every wireless LAN data frame. Some preliminary simulation results demonstrate that the AM method is effective and adaptive to non-stationary co-channel interference.

The goal of the article is to find a minimal amount of statistical information (or a much weaker condition) about source signals for which blind equalization is possible for multiple-input multiple-output finite-impulse response (MIMO-FIR) channels. First, a sufficiently broad framework is set up within which such a theoretical problem is well posed. Within this framework, it is shown that second-order statistics (SOS) alone are not sufficient for blind equalization when the source signals are white. Additional higher order statistics (HOS) are needed. Then we show that the only additional higher order statistical information needed is spatial fourth-order cumulants. Though it has not yet been proved to be minimal, it is interesting to note that this is the same as the weakest known condition on the source signals even for an instantaneous mixture. We then show a necessary and sufficient condition for blind equalization when the source signals are white and spatially fourth-order uncorrelated. Based on this condition, criterion (A) for blind equalization of MIMO-FIR channels is developed by exploiting the temporal fourth-order statistics. Finally, based on this criterion, a new necessary and sufficient condition in terms of cumulants for the blind equalization of MIMO-FIR channels is obtained.

The autocorrelation matching method is a blind signal separation and channel equalization technique for distributed MIMO communication systems over unknown FIR channels using only second order statistics. This method is based on a theoretical discovery, ie, under the condition that the autocorrelation functions of the (multiple) inputs are linearly shift-independent, an input is recovered, up to a unitary factor and a delay, by an output of an MIMO-FIR equalizer if and only if the autocorrelation function of the output matches that of the input. An optimal zero-forcing equalizer is developed to maximize the SNR for the outputs, ie, the recovered inputs. Some preliminary simulation results show that the BER in the recovered inputs is about 3/spl times/10/sup -5/ at the SNR=15 dB. This method has the potential to be applied to cellular wireless communications for the purpose of boosting spectrum efficiency or suppressing co-channel interference.

We consider the blind deconvolution problem of FIR multiple-input multiple-output channel systems. The systems-theoretic foundation is established on the broadest class of equalizable channel systems. This class of equalizable channel systems is then characterized. Based on this characterization, a fundamental factorization property, called irreducible-allpass factorization, is shown. Based on this fact, it is shown that any equalizable FIR system can and only can be converted to a paraunitary system, through the whitening of the observations, using the second-order statistics. The current second order statistics techniques considered only those channel systems that are irreducible, only half of the irreducible-allpass factorization of equalizable channels.

This paper addresses the blind deconvolution of multi-input-multi-output (MIMO) FIR systems driven by white non-Gaussian source signals. First, we present a weaker condition on source signals than the so-called i.i.d. condition so that blind deconvolution is possible. Then, under this condition, we provide a necessary and sufficient condition for blind deconvolution of MIMO FIR systems. Finally, based on this result, we propose two maximization criteria for blind deconvolution of MIMO FIR systems. These criteria are simple enough to be implemented by adaptive algorithms.

Previous studies show that equalizers with different equalization delays achieve different performances. The best performed equalizer is not necessarily the one with 0-delay, and often not the case when the channel is non-minimum-phase. In this paper, a blind channel equalization algorithm is presented, by which equalizers with all possible equalization delays can be calculated simultaneously from the second order statistics of the received signals. A (blind) evaluation index is then presented for the purpose of selecting the best equalizer. Simulation shows that the best delayed equalizer performs much better than that for other equalization delays and those given by constant modulus algorithm and linear prediction algorithm.

Conventional blind signal separation algorithms do not adopt any asymmetric information of the input sources, thus the convergence point of a single output is always unpredictable. However, in most of the applications, we are usually interested in only one or two of the source signals and prior information is almost always available. In this paper, a principal independent component analysis (PICA) concept is proposed.We try to extract the objective independent component directly without separating all the signals. A cumulant-based globally convergent algorithm is presented and simulation results are given to show the hopeful applicability of the PICA ideas.

Blind channel equalization has been a very active research topic due to its potential application in mobile communications and digital TV systems. In this paper, we present a new blind zero-forcing equalizer that utilizes second order statistics from the multi-channel configuration. The algorithm is simple and relies only on nullspace decomposition. It can actively select the desired delay of the equalizer output signal. The performance of this new algorithm is demonstrated through simulation examples.

Unlike the conventional stochastic approach, an unsupervised learning algorithm is viewed as a deterministic system. A new concept of time-average invariance is introduced, which is a property of deterministic signals, but plays the role of stochastic signals that are stationary and ergodic. As such, deterministic-based analysis can be used for stochastic-like signals. Consequently, the complexity of convergence analysis is significantly reduced. The simplicity of the main theorem also suggests the possibility for the design of unsupervised learning algorithms. Two examples are given for illustration.

In this paper, we investigate the capability of approximating functions in C(R (n)) by three-layered neural networks with sigmoidal function in the hidden layer. It is found that the boundedness condition on the sigmoidal function plays an essential role in the approximation, as contrast to continuity or monotonity condition. We point out that in order to prove the neural network in the n-dimensional case, all one needs to do is to prove the case for one dimension. The approximation in L(p)-norm (1<p

An unsupervised neural network is constructed for the problem of blind signal separation. It is designed based on the condition that the outputs of the neural network are independent. A study of the stability of the neural network in the sense of expectation is presented. A stability condition on the system matrix A is obtained. Simulation studies show that this neural network is robust. It can separate two signals with strength ratio 100:1.>

Adaptive extraction of principal components of a vector stochastic process is a topic currently receiving much attention. The authors propose a learning algorithm implemented on a neural-like network. This algorithm is shown to be superior to previous ones. The convergence of this algorithm can be proved, but only an outline of the proof is presented.>

A closed-form solution to the parameter identification of multichannel moving average processes is presented. An eigenstructure-based approach is proposed to solve the identification equations involving output cumulants of any order by further exploiting the eigenstructure of the output cumulant matrices. The identification equations are derived without using the Kroneker products, hence greatly reducing the computational complexity. This approach is also computationally simpler than that of L. Tong et al. (1991), and iterations are avoided. In addition, the proposed approach allows one to combine the statistics of different order to achieve better performance. Computer simulation has shown that a much smaller sample size, on the order of 5000, is needed to achieve even better performance than our previous approach.>

An identifiability result for the blind identification of memoryless channels with independent sources is presented. The necessary and sufficient conditions which were obtained by investigating the eigenstructure inherited in the observation cumulants are given. An eigenstructure-based algorithm is proposed to achieve the identification of the parameter matrix whenever such identification is possible. It is shown that the source signals can be extracted up to scale factors and a permutation if and only if all the source signals, but one, are non-Gaussian.>

An iterative algorithm for the identification of multichannel moving average (MA) processes driven by mutually independent and identically distributed (i.i.d.) input signals is proposed. It is shown that the algorithm has a finite-step global convergence property. This algorithm is computationally efficient and numerically stable. Two multichannel MA models, including one nonminimum-phase MA model, are estimated by this algorithm with satisfactory performances. It is shown that this algorithm guarantees a solution of third-order cumulant-based identification equations.>

Two forms of nonlinear-feedback shift-registers are considered. In the Type-I register, the feedback output is added to the shift-register contents at an arbitrary number of stages. In the type-II register, the feedback is input to the first stage only. It is shown that for every Type-I register there is an equivalent Type-II register in the sense that the autonomous state diagrams differ only by a labelling of the states. Moreover, the mapping between equivalent states can always be chosen to be a linear transformation. This theorem is a well-known result in the theory of linear-feedback shift-registers and is thus seen to apply unchanged to the nonlinear case.