Mark A. Wickert

Implementation of a real-time frequency-selective RF channel simulator using a hybrid DSP-FPGA architecture

A low cost frequency-selective RF channel simulator architecture is explored in this paper. The technique of quadrature amplitude modulation (QAM) by independent low-pass filtered white Gaussian noise sources forms a rational function approximation (RFA) to the desired Doppler spectrum for flat Rayleigh fading. To simulate frequency-selective fading, this QAM/RFA architecture may be extended by combining delayed outputs from multiple flat fading generators. In this paper, the noise shaping filter considered is in the form of an infinite-impulse-response digital filter followed by an interpolator (upsampler) using linear interpolation. The performance requirements are those in the standard channel simulator section of TIA IS-55-A. The system is implemented almost entirely in the digital domain by use of IF sampling, with the signal processing performed in a high-end floating-point digital signal processor and a field-programmable gate array. The theoretical performance of the simulator is studied with respect to the TIA standard, and limitations of the hardware prototype are identified. A system capable of simulating 12 delay taps, with a processing bandwidth of 5 MHz, can be built at about one-tenth the cost of commercially available channel simulators of comparable performance.

Minimization of detection of symbol-rate spectral lines by delay and multiply receivers

The detectability of the symbol rate of a digitally modulated signal by a delay and multiply receiver is investigated. Two signal pulse types are considered: Nyquist pulses and rectangular pulses. Since many digital transmission systems use rectangular pulse shapes, symbol rate-line detectability is considered for receivers optimized for rectangular pulses. For a receiver configuration that is reasonable for rectangular pulses, it is found that with a Nyquist pulse parameter alpha =0.25, the rate-line signal-to-noise ratio is down about 20 dB from that for a rectangular pulse signal. >

Probability of error analysis for FHSS/CDMA communications in the presence of fading

Expressions are found for the probability of error of a slow frequency-hopped spread-spectrum (FHSS) code-division multiple-access (CDMA) system using noncoherent M-ary frequency-shift-keyed (FSK) data modification in the presence of slow-nonselective Rayleigh or single-term Rician fading. The analysis is general enough for the consideration of the near/far problem under the specified channel conditions. Comparisons are made between the error expressions developed here and previously published upper bounds. It is found, under certain conditions, that the previous upper bounds on the probability of error may exceed the true probability of error by an order of magnitude. >

Nonstationary moments of a random binary pulse train

A method for finding the nonstationary moments of a random binary pulse train with arbitrary pulse shape is described and used to find the joint moments of the pulse train with a delayed version of itself. These moments are used in the analysis of nonlinear synchronization circuits. The method uses the moment generating function and a recursive relationship for moments and cumulants. Example derivations are given for moments up to eighth order. >

Implementation of a real-time, frequency selective, RF channel simulator using a hybrid DSP-FPGA architecture

A low cost, frequency selective, RF channel simulator architecture is explored. The system is implemented almost entirely in the digital domain by use of IF sampling with the signal processing performed in a high-end floating point DSP and an FPGA. The prototype system costs less than

Direct-sequence spread-spectrum using directionally constrained adaptive beam forming to null interference

3900 and is capable of simulating three delay taps. It is believed that a system capable of simulating 12 delay taps with a bandwidth of 5 MHz could be built at a cost less than

Minimization of Detection of Baud Rate Spectral Lines by Delay and Multiply Receivers

2000, at least an order of magnitude cheaper than commercially available channel simulators of comparable performance.

Joint maximum-likelihood parameter estimation for burst DS spread-spectrum transmission

A direct-sequence spread-spectrum communication system using an adaptive antenna array to null interference is examined. The system model is a modification of Comptons (1978) system, using a digital implementation and a directional constraint. Computer simulations are used to compare the behavior of the directionally constrained model with Comptons model with emphasis on spread-spectrum code synchronization. In the directed system, nulling of a tone jammer occurs without respect to direct-sequence code acquisition, and nulling of an off-beam desired signal occurs very slowly. Theoretical results are obtained by using the system covariance matrix, and incorporating the effects of spread-spectrum filtering to compute the system response as a function of time. The simulated results indicate that incomplete directional knowledge may be exploited to enhance signal quality and speed up code acquisition. >

Practical Volterra equalizers for wideband satellite communications with TWTA nonlinearities

The detectability of the baud rate of a digitally modulated signal that uses Nyquist pulses is investigated. The intercept receiver is assumed to be a delay and multiply circuit followed by a complex correlator. Since many digital transmission systems use rectangular pulse shapes, Nyquist pulse rate line detectability is considered for receivers optimized for rectangular pulses. For a receiver configuration that is reasonable for rectangular pulses, it is found that with a Nyquist pulse with parameter ¿ = 0.25, the rate line signal-to-noise ratio is down about 20 dB from that for a rectangular pulse signal.

Channel Estimation and Mitigation Techniques for OFDM in a Doppler Spread Channel

We first derive the joint optimal maximum-likelihood (ML) estimator of the carrier phase, both Doppler shift and Doppler rate, and the spreading code delay for a short burst direct sequence/spread spectrum (DS/SS) transmission in the absence of the data modulation. The typical burst duration is three data bit periods (60 ms). The performance of the joint estimator is analytically derived separately for high and low carrier-to-noise ratio (CNR) values. A suboptimal ML estimator based on the method of averaged periodogram is proposed for the data modulation present case, allowing the joint estimation of the data bit values. Then the above parameters are assumed correctly estimated and a segmentation approach is adopted, deriving the optimal joint ML estimator for the bit synchronization epoch and data. Simulations show the joint estimators perform reliable down to a CNR of approximately 30 dB/spl middot/Hz.