Gonçalo Nuno Gomes Tavares

Visual neuroprosthesis: a non invasive system for stimulating the cortex

This paper describes a complete visual neuroprosthesis wireless system designed to restore useful visual sense to profoundly blind people. This visual neuroprosthesis performs intracortical microstimulation through one or more arrays of microelectrodes implanted into the primary visual cortex. The whole system is composed by a primary unit located outside the body and a secondary unit, implanted inside the body. The primary unit comprises a neuromorphic encoder, a forward transmitter, and a backward receiver. The developed neuromorphic encoder generates the spikes to stimulate the cortex by approximating the spatio-temporal receptive fields characteristic response of ganglion cells. Power and stimuli information are carried to inside the cranium by means of a low-coupling transformer, which establishes a wireless inductive link between the two units. The secondary unit comprises a forward receiver, microelectrode stimulation circuitry and a backward transmitter that is used to monitor the implant. Address event representation is used for communicating spike events. Data is modulated with binary frequency-shift keying and differential binary phase-shift keying in the forward and in the backward directions, respectively. A prototype of the proposed system was developed and tested. Experimental results show that the spikes to stimulate the visual cortex are accurately generated and that the efficiency of the inductive link is relatively high, about 28% in average for 1 cm intercoil distance providing a power of about 50 milliwatts to the secondary implanted unit. Application specific integrated circuits were designed for this secondary unit, showing that, with current technology, it is possible to implement such a unit, respecting the power constraints.

A Mobile-to-Mobile Fading Channel Simulator Based on an Orthogonal Expansion

This paper presents a Rayleigh fading mobile-to- mobile channel simulator based on a modified Karhunen-Loeve orthogonal expansion of a complex Gaussian fading process. The method is similar to the well-known IFFT method but with a different frequency mask. The simulation accuracy is assessed by the computation of power margins (for the autocorrelation) and of Kullback-Leibler divergence (for the envelope probability density function) of the simulated fading process. The results demonstrate a good agreement with the theory and a slight improvement relative to the IFFT method.

The true Crame/spl acute/r-Rao lower bound for data-aided carrier-phase-independent time-delay estimation from linearly Modulated waveforms

In this paper, we present the derivation and analysis of the true Crame/spl acute/r-Rao lower bound (CRB) for the variance of unbiased, data-aided (DA) symbol-timing estimates, obtained from a block of K samples of a linearly modulated information signal, transmitted through an additive white Gaussian noise channel with random carrier phase. We consider a carrier-phase-independent time-delay estimation scenario wherein the carrier phase is viewed as an unwanted or nuisance parameter. The new bounds require only a moderate computational effort and are tighter than the CRB for the variance of unbiased time-delay estimates obtained under the assumption that the carrier phase is known. These bounds are particularly useful to assess the ultimate accuracy that can be achieved by pilot-assisted symbol synchronizers. Conversely, they may be used to evaluate data sequence suitability for the purpose of time-delay estimation. Comparison of the actual variance of a DA feedforward timing estimator with the new bounds show that these are attainable by practical synchronizers.

On the true Cramer-Rao lower bound for data-aided carrier-phase-independent frequency offset and symbol timing estimation

In this letter we present new and simple closed form expressions for the true Cramer-Rao lower bound (CRB) for data-aided (DA) joint and individual carrier frequency offset and symbol timing estimation from a linearly modulated waveform transmitted over an AWGN channel. The bounds are derived under a carrier-phase-independent (CPI) estimation strategy wherein the carrier phase is viewed as a nuisance parameter and assumed to have a worst-case noninformative uniform distribution over [-¿, ¿]. The computation of these CRBs requires only a single numerical integration. In addition, computationally simpler yet highly accurate asymptotic lower bounds are presented. As particularizations, new bounds for individual CPI frequency estimation with known symbol timing from M-PSK and continuous wave (CW) signals are also reported.

A New Type of Normalized LMS Algorithm Based on the Kalman Filter

While the LMS algorithm and its normalized version (NLMS), have been thoroughly used and studied, and connections between the Kalman filter and the RLS algorithm have bean established, the connection between the Kalman filter and the LMS algorithm has not received much attention. By linking these two algorithms, a new normalized LMS algorithm can be derived that has some advantages to the classical one. Firstly, its stability is guaranteed since it is a special case of the Kalman filter. Secondly, it suggests a new way to control the step size that results in optimum convergence for a large range of input signal powers, that occur in many applications. Finally, it prevents measurement noise amplification that occurs in the NLMS algorithm for low order filters, like the ones used in OFDM equalization systems.

RF-link for cortical neuroprosthesis

This paper describes an RF-link system suitable for intracortical neuroprosthesis. The wireless coupling minimizes the risk of infection and improves the patient health level. Data and power are transmitted by means of an RF low-coupling transformer. Binary FSK modulation is used in the forward link (from outside to implanted prosthesis) and BPSK is used in the backward link (implanted prosthesis to outside). Experimental results are presented, and important concerns regarding the magnetic coupling are addressed.

A new ML-based data-aided feedforward symbol synchronizer for burst-mode transmission

A data-aided (DA) feedforward symbol synchronizer based upon the maximum-likelihood (ML) criterion is proposed. The algorithm is carrier-phase independent and is designed for operation in burst-mode, narrow-band communication systems employing linear modulation. The accuracy of the proposed algorithm is assessed by simulation using QPSK signaling. The results show that the performance of the proposed synchronizer is very close to the theoretical minimum limit (variance of the estimates near the Cramer-Rao lower bound) and is also significantly better than that of other popular circuits such as the square-law synchronizer.

On the Statistics of the Sum of Squared Complex Gaussian Random Variables

In this letter, we derive new results for the statistics of the complex random variable <i>z</i>=^DeltaSigma_n=1 ^N <i>x</i> _n ²=<i>z</i> ₁+<i>jz</i> _Q = <i>re</i> ^jphi where {<i>x</i> _n} is a set of mutually independent complex-valued Gaussian random variables with arbitrary means and equal variances. Each random variable <i>x</i> _n is assumed to have independent real and imaginary components with equal variance for all <i>n</i>. Expressions are derived for the joint probability density function (pdf) of (<i>z</i> ₁,<i>z</i> _Q), for the joint pdf of (<i>r</i>,phi) and also for the marginal pdf of the modulus <i>r</i>. An useful Fourier series expansion for the pdf of the phase phi is also derived. As an application of the results, a theoretical performance analysis of the well-known nondata-aided Viterbi and Viterbi feedforward carrier phase estimator operating with BPSK signals is presented. In particular, the expressions for the exact pdf, variance, and equivocation probability of the carrier phase estimates are derived.

On the Generation of Correlated Gaussian Random Variates by Inverse DTF

In this paper the problem of generating a stationary band-limited Gaussian random vector with arbitrary complex autocorrelation by the inverse discrete Fourier transform (IDTF) algorithm is considered. Instead of using the classical frequency mask (FM), determined from samples of the (band-limited) target power spectral density (PSD) of the process, a new FM is obtained by matching the autocorrelation obtained with the IDFT algorithm to a desired arbitrary autocorrelation. Example results presented show that the new FM is able to significantly increase the autocorrelation accuracy of the generated process at no additional online computational cost.

On the Miller-Chang lower bound for NDA carrier phase estimation

In this letter, we derive a new Miller-Chang lower bound (MCB) for the variance of unbiased non-data-aided (NDA) carrier phase estimates obtained from a block of N samples of a linearly modulated pulse-amplitude modulation or quadrature-amplitude modulation information signal, transmitted through an additive white Gaussian noise channel. This bound is tighter than the corresponding Crame/spl acute/r-Rao lower bound (CRB) for data-aided or continuous-wave carrier phase estimation (CW-CRB), particularly when N is small. For some given N and sufficiently high signal-to-noise ratios, the MCB is tighter than the corresponding true NDA CRB. The main limitation of this new MCB is that its application is restricted to carrier phase estimators which are unbiased for all possible values of the received symbol sequence.