D.G.M. Cruickshank

An FFT-Based Approach for Fast Acquisition in Spread Spectrum Communication Systems

This paper describes a non-coherent technique forfast acquisition of direct sequence spread spectrum(DS/SS) signals in low earth orbit (LEO) satellitecommunication scenarios. Large Doppler offsets areinherent to such environments and are likely to causemajor problems during the code acquisition phase dueto the introduced frequency ambiguity. In the presentpaper we discuss the use of a set of partialcorrelators for code phase acquisition, combined witha fast Fourier transform (FFT) for the purpose of simultaneous Doppler estimation. We will show that theuse of this architecture largely accelerates thesynchronisation process compared to conventionaltechniques over a wide range of Doppler offsets.Furthermore we discuss the implementation of azero-padded FFT that increases the frequencyresolution to such an extent, that acquisition timesidentical to scenarios without existing Doppler offsetcan be achieved.

A robust location estimator architecture with biased Kalman filtering of TOA data for wireless systems

This paper presents a robust location estimator architecture for a time difference of arrival (TDOA) system, which mitigates non-line of sight (NLOS) and multipath errors, using biased Kalman filtering of time of arrival (TOA) estimates. With additional Kalman filtering of location estimates a high degree of location accuracy is demonstrated. Simulations of an idle period downlink (IPDL) UMTS system give an accuracy to within 50 m at the 67th percentile for most scenarios.

RBF based receivers for DS-CDMA with reduced complexity

This paper presents a Mahalanobis based radial basis function network (RBF) receiver structure with reduced complexity. It is also illustrated, how the RBF receiver could be employed as a multiuser detector. The RBF structure has superior performance over linear receiver structures and is equivalent to MLSD in an AWGN channel. However, a drawback is its complexity in multipath channels. Ideas from pattern recognition are exploited to reduce its complexity for multipath scenarios with little performance loss.

Doppler estimation for fast acquisition in spread spectrum communication systems

This paper describes a non-coherent technique which employs a fast Fourier transform (FFT) for the fast acquisition of direct sequence spread spectrum signals in the presence of large Doppler shifts. The application of an FFT to code acquisition results in decreased acquisition time and can improve the probability of detection. The technique discussed in this paper uses a set of partial correlators and a zero-padded FFT to reduce the search region for code acquisition whilst maintaining good frequency resolution for Doppler offset. This approach will prove most pertinent in future re-configurable terminals.

Volterra Based Receivers for Ds-Cdma

This paper presents a Volterra function based filter as the receiver for a direct-sequence-code-divisionmultiplex-access (DS-CDMA) system. The envisaged application would be cellular telephony, where the receiver will be required to be adaptive due to channel effects. The results obtained from the Volterra filter are compared against the linear minimum-meansquare-error (MMSE) filter and the nonlinear radialbasis-function (RBF) network in a stationary multipath channel. Further, the computational complexity will be investigated and compared against the MMSE.

Optimal and adaptive FIR filter receivers for DS-CDMA

We calculate the optimum (Weiner) FIR filter for a DS-CDMA system in Gaussian noise using Gold codes. We consider adaptive filter approximations to the Weiner filter, showing that an RLS algorithm can provide a good approximation to the Weiner filter with fast convergence in the stationary case. The outline of a method for extending this work to the non-stationary cellular channel is also given. We concentrate our results on the filters ability to suppress multiaccess interference (MAI). The choice of an FIR filter enables us to calculate optimal filter in the presence of MAI and additive noise using Wiener filter theory and to describe a method for extending the optimisation process to include the multipath cellular channel.

Variable length equalizers for broadband mobile systems

The very high bit rates of modern mobile systems (2.5G and 3G) demand powerful channel distortion mitigation, and for this role, equalization is a possible candidate. In this paper the importance of the length of linear equalizers is analyzed. It is shown that accurate control of equalizer length can improve the systems performance and also reduce the handset power consumption.

Transmitter based inverse filters for MAI and ISI mitigation in a TDD/CDMA downlink

Recently, research has been focused on techniques that simplify the mobile receiver structure in a CDMA system by means of signal precoding. These methods assume knowledge of the channel at the transmitter which is feasible in a time division duplex (TDD) system where the channel is assumed to be the same on uplink and downlink. A novel algorithm based on prefiltering the spread signal in the base station is described. The prefilter taps are determined with an MMSE criterion and the necessary power constraint. We show that this technique increases the system capacity with a low computational cost compared with existing schemes. The portable units structure is reduced to the conventional receiver for a single user in an AWGN channel, a matched filter. Thus no channel estimation, equalisation or multiuser detection is required at the mobile. The theoretical analysis is supported with simulation results where the performance is compared to other recently published methods.

A Unified Approach to Dynamic Length Algorithms for Adaptive Linear Equalizers

The number of taps is an important factor which affects nearly all the performance and complexity measures of adaptive equalizers. Recently, dynamic length algorithms for the adaptive finite impulse response (FIR) filters have attracted considerable attention and several methods have been proposed. Nevertheless, the relationship among these techniques is still unclear which makes the performance comparison remaining difficult. In this paper, a unified explanation of these techniques is presented which summarizes all proposed algorithms and provides a platform for their performance comparison. Moreover, a new type of dynamic length algorithms, the varying threshold dynamic length algorithm (VT-DLA) is proposed. Simulation results show that the performance of the new algorithm outperforms that of existing algorithms

A chip based multiuser detector for the downlink of a DS-CDMA system using a folded state-transition trellis

The maximum-likelihood sequence estimation (MLSE) multiuser detector has very good performance but at the expense of the exponentially increasing number of states. We propose a scheme that exploits the feature of power-equality in the downlink to reduce the number of states involved. By using all the possible discrete chip values as the states in the transition trellis diagram and operating the Viterbi algorithm at the chip level, the number of states is reduced from 2/sup K/ to K+1 where K is the number of users.