An improved resource allocation scheme is proposed in this paper which uses genetic algorithms (GAs) in conjunction with the recently developed plane cover multiple-access (PCMA) scheme in order to maximize the attainable capacity of packet-based wireless cellular networks. The studied problem has been proven to be in the class of nondeterministic polynomial (NP)-hard problem, therefore, the powerful search capability of the GA is a key factor in improving the performance of cellular resource allocation. Computer simulation results suggest that the proposed approach outperforms the "uniform" and the "greedy" algorithm-based "min " methods in terms of the number of serviced users.

An Adaptive minimum bit error rate (MBER) based linear multiuser detection scheme is proposed for an asynchronous multicarrier code-division multiple-access (MC-CDMA) system in frequency selective Rayleigh fading channels. In general, a MC-CDMA system employs the minimum mean square error (MMSE) receiver due to its good performance and amenability to adaptive implementation. However, the MSE cost function is nut optimal in terms of the bit error probability performance of the system. The proposed multiuser detector (MUD) that directly minimizes the BER can significantly improve the performance by employing an adaptive stochastic gradient algorithm based on the estimation of kernel density function in order to adjust the weights in the receiver. Simulation results demonstrate a clear performance improvement by using the proposed multiuser detection scheme.

Abstract-Recently, a robust version of the linear decorrelating detector (LDD) based on the Huber's M-estimation technique has been proposed. In this paper, we first demonstrate the use of a three-layer recurrent neural network (RNN) to implement the LDD without requiring matrix inversion. The key idea is based on minimizing an appropriate computational energy function iteratively. Second, it will be shown that the M-decorrelating detector (MDD) can be implemented by simply incorporating sigmoidal neurons in the first layer of the RNN. A proof of the redundancy of the matrix inversion process is provided and the computational saving in realistic network is highlighted. Third, we illustrate how further performance gain could be achieved for the subspace-based blind MDD by using robust estimates of the signal subspace components in the initial stage. The impulsive noise is modeled using non-Gaussian alpha-stable distributions, which do not include a Gaussian component but facilitate the use of the recently proposed geometric signal-to-noise ratio (G-SNR). The characteristics and performance of the proposed neural-network detectors are investigated by computer simulation.

Multiuser communications channels based on code division multiple access (CDMA) technique exhibit non-Gaussian statistics due to the presence of highly structured multiple access interference (MAI) and impulsive ambient noise. Linear adaptive interference suppression techniques are attractive for mitigating MAI under Gaussian noise. However, the Gaussian noise hypothesis has been found inadequate in many wireless channels characterized by impulsive disturbance. Linear finite impulse response (FIR) filters adapted with linear algorithms are limited by their structural formulation as a simple linear combiner with a hyperplanar decision boundary, which are extremely vulnerable to impulsive interference. This raises the issues of devising robust reception algorithms accounting at the design stage the non-Gaussian behavior of the interference. We propose a multiuser receiver that involves an adaptive nonlinear preprocessing front-end based on a multilayer perceptron neural network, which acts as a mechanism to reduce the influence of impulsive noise followed by a postprocessing stage using linear adaptive filters for MAI suppression. Theoretical arguments supported by promising simulation results suggest that the proposed receiver, which combines the relative merits of both nonlinear and linear signal processing, presents an effective approach for joint suppression of MAI and non-Gaussian ambient noise.

The ability to determine the size distribution of a field of bubbles has a wide range of practical applications. Bubbles have traditionally been detected by using either optical or acoustic techniques. Bubbles exhibit strong acoustic resonance properties dependent on their size. By probing a field of bubbles it is possible to form an estimate of the range of sizes present by analysing the scattered data. There are a number of acoustic bubble sizing methods currently available, some of which exploit the nonlinear dynamic behaviour of the bubble. By modeling this behaviour, improvements may be made to the estimation techniques. This paper presents estimates for the quadratic transfer function of a single bubble based on the Volterra series. A method of estimating the Volterra kernels using a neural network is presented which significantly reduces the quantity of data required to form an estimate of the transfer function.

This paper presents the development and performance of blind algorithms for a spatial diversity scheme to enable reliable data telemetry over a long range underwater acoustic channel. A number of Bussgang based stochastic gradient algorithms were tested for this multipath channel with additive white and coloured shipping noise. Both simulation and real experimental tests have shown that a significant improvement is obtained by utilising the spatial diversity of the long range channel and the ability of the combiner to perform joint equalisation and carrier phase tracking.

A concurrent error detection (CED) scheme has been proposed by Jou and Abraham (1988)for a radix-2 FFT network. The method uses a coding scheme to ensure that all the modelled faults are detectable. The same method has been applied to the prime factor algorithm (PFA) FFT network. The results show that the CED scheme is applicable to multi-factor PFA networks and it is efficient, only, for large N. Since the PFA is a minimum multiplication algorithm, it is logical to adopt a minimum multiplication CED scheme. For this purpose, a suitable CED scheme has been developed such that the data coding is no longer needed which results in less hardware and time overheads. The new scheme provides a stage-by-stage error detection without causing time delay and additional roundoff noise. The hardware overhead is substantially reduced with improved error detection performance since the scheme needs to consider the roundoff error of only one stage instead of the entire FFT network.

This paper describes a parallel architecture that has been developed to perform real-time synthetic aperture sonar imaging as part of the Acoustical Imaging Development (ACID) project. The project has successfully developed a synthetic aperture sonar system for producing high resolution images of the sea floor and that has been tested during a series of sea trials in May 1993 off the south coast of France. This paper describes the synthetic aperture processing system developed by the University of Newcastle upon Tyne and its use of transputer modules and associated devices in order to obtain real-time imaging performance, the software structure of the processing system and the load balancing techniques that have been developed in order to provide efficient processing. The use of a parallel distributed architecture has also allowed a processing system that can readily be extended to deliver greater computational power in the future. Images produced by the synthetic aperture processor from data collected from around the Toulon coastal region are presented. These images highlight the improvement in azimuth resolution that can be obtained from synthetic aperture processing over conventional sidescan sonars.

An outline is given of the optimal and hierarchical control methods which are particularly effective and efficient for solving some fermentation problems and which can be implemented using a microcomputer system. The literature of the fermentation models which encompass major biological features usually encountered in a fermentation process are documented. The principal features, advantages, and limitations of the microcomputer systems are pointed out. Finally, the practical application of the optimal of fermentation processes using a microcomputer system published previously by the authors are listed.

The application of hierarchical (or multilevel) techniques to solving nonlinear optimal control problems in fermentation systems using microprocessors described. All results presented refer to batch gluconic acid fermentation, and they show why these techniques are more advantageous and attractive than the traditional single-level optimization method. In particular, the simulation results and conclusions show that hierarchical techniques result in the control computer requiring lower computational power and memory capacity.