SPTM-8.1

The Detection of Radar Pulse Sequences by Means of a Continuous Wavelet Transform
Douglas E Driscoll, Stephen D Howard (Defence Science Technology Organisation, Commonwealth of Australia)

A wavelet transform is introduced as a means of detecting the characteristic scale or period T of a radar pulse sequence in an incoming stream of pulses. A particular choice of mother wavelet contains a fixed number, M, of cycles of a complex exponential, which provides a resolution of 1/M times T. The transform operates on interleaved pulse sequences and, by thresholding, determines T for each component sequence. The detector is robust against missing pulses and timing jitter and is sensitive to simple, staggered and complex pulse sequences. The method is an improvement on other established approaches, such as the time-difference-of-arrival (TDOA) histogram and the periodogram.

SPTM-8.2

Analysis of spectral and wavelet-based measures used to assess cardiac pathology
Conor J Heneghan (University College Dublin), Steven B Lowen, Malvin C Teich (Boston University)

In recent studies of heart rate variability in humans, it has been demonstrated that the scale-dependent wavelet transform (WT) standard deviation [sigma_wav(m)] of the interbeat intervals can be used to discriminate patients with certain forms of cardiac dysfunction from normal subjects. This paper forges an explicit link between this measure and a corresponding spectral measure, which is also shown to provide discrimination between the two classes of data. The statistics of the estimator for sigma_wav(m) are obtained in the analytically simplest case, including expressions for its bias and variance. Numerical simulations are provided to support the theoretical analysis. We compare the bias, variance, and frequency resolution of WT and spectral measures, and conclude that the former appears more suited to our particular circumstances.

SPTM-8.3

Optimal DMT Transceivers over Fading Channels
Yuan-Pei Lin (Dept. Electrical and Control Engr. National Chiao Tung Univ.), See-May Phoong (Dept. of EE & Inst. of Comm Engr. National Taiwan Univ.)

Recently DFT based discrete multitone modulation (DMT) systems have been widely applied to various applications. In this paper we study a broader class of DMT systems using more general unitary matrices instead of DFT matrices. For this class we will show how to design the optimal DMT systems over fading channels with colored noise. Examples will be given to show the improvement over the traditional DFT based DMT system. In addition we introduce a modified DFT based DMT system. The new system has the same complexity but better noise rejection property.

SPTM-8.4

Narrowband Channel Extraction for Wideband Receivers
Matthew L Welborn (MIT Lab for Computer Science)

One of the most computationally intensive processing stages of a wideband digital receiver is the extraction of a narrowband channel from a wideband input signal. In implementations that compute the convolutional sum, the computation is proportional to the bandwidth of the input signal. This paper shows how to break this dependence, reducing the limiting factor to the requirement to maintain a sufficient output signal-to-noise ratio (SNR). This paper describes two complementary algorithms for efficient channel extraction in wideband receivers. The first allows the required frequency translation to be performed at the lower sample rate of the channel filter output. The second algorithm decouples the effect of interference rejection from SNR improvement and improves the computational efficiency of filtering by using only a subset of the input samples. Additionally, we present a simple model to quantify the effects of this technique and experimental verification using a wideband software radio receiver.

SPTM-8.5

Performance analysis of multicarrier modulation systems using cosine modulated filter banks
Subbarao S Govardhanagiri (Boston University), Tanja Karp (Mannheim University), Peter Heller (Aware Inc.), Truong Nguyen (Boston University)

In this paper we compare the performance of biorthogonal cosine-modulated transmultiplexer filter banks with today's multicarrier modulation systems whose transceivers are based on DFT. In contrast to early works on transmultiplexer filter banks that concentrated on the derivation of perfect reconstruction constraints of the filter bank or prototype design, this study takes into consideration a typical twisted pair copper line transmission channel into consideration and examines the influence of different system parameters as filter length, number of channels, and the overall system delay on the distortion at the receiver. Biorthogonal filter banks have the advantage that filter length and overall system delay can be chosen independently. Restricting the equalizer at the receiver to a single scalar tap per subchannel, we show that cosine-modulated filter banks outperform DFT based multicarrier systems without a guard interval and obtain a similar performance to DFT based systems with a guard inverval and time domain equalization but at a lower computational cost and a higher throughput data rate.

SPTM-8.6

Optimal Ladder-Based Biorthogonal Coder
See-May Phoong (Dept. of EE & Inst. of Comm Engr. National Taiwan Univ.), Yuan-Pei Lin (Dept. Electrical and Control Engr., National Chiao Tung Univ.)

In this paper, a novel minimum noise structure is introduced for ladder-based biorthogonal filter banks. The proposed MInimum Noise LAdder-based Biorthogornal (MINLAB) coder ensures that the noise gain of the quantizers is unity, even though the system is not orthonormal. The coding gain of the optimal MINLAB coder is always greater than unity. For both the AR(1) and MA(1) processes, the MINLAB coder with 2 taps outperforms the optimal orthonormal coders of any number of taps. In addition to its superior coding performance, the optimal biorthogonal coder has a very low design and implementational cost. Moreover the proposed coder enjoys many advantages that make it an attractive choice for lossy/lossless data compression.

SPTM-8.7

The Predictive Embedded Zerotree Wavelet (PEZW) Coder: Low Complexity Image Coding with Versatile Functionality
Jie Liang (Texas Instruments, Inc.)

In this paper, we introduced the Predictive Embedded Zerotree Wavelet(PEZW) codec, an image coder that achieves good coding efficiency and versatile functionality with limited complexity requirement. Our complexity analysis showed that the memory requirement of this coder is less than 15k bytes regardless of image sizes. Our simulation results also showed that the coding efficiency of this low complexity coder is competitive with the state of the art of wavelet coders that use whole image buffers. The PEZW coder described in this paper has been adopted in MPEG4 as its still texture coding tool and is currently a proposal to the evolving JPEG2000 standard.

SPTM-8.8

A NEW DIRECTIONAL FILTER BANK FOR IMAGE ANALYSIS AND CLASSIFICATION
Sang-il Park, Mark J.T Smith, Russell M Mersereau (Georgia Institute of Technology)

A new directional filter bank for image analysis and classification is proposed. This paper introduces an improved structure in order to visualize subband outputs of the directional filter banks, while retaining the attractive properties of the original directional filter banks such as 1-D separable filtering, perfect reconstruction, and maximal decimation. Using this structure, any arbitrary $2^n$ band directional filter bank can be implemented by cascading simple directional filter bank blocks, unlike the original structure that needs a parallel structure for visualizing subband outputs. Also, in order to have non-distorted phase information in the subbands for visualization, both FIR and IIR filter prototypes that can be implemented efficiently are provided for linear phase filtering. This paper shows the approach proposed here can be applied to image analysis and classification.

SPTM-8.9

A Modulated Complex Lapped Transform and its Applications to Audio Processing
Henrique S Malvar (Microsoft Research)

This paper introduces a new structure for a modulated complex lapped transform (MCLT), which is a complex extension of the modulated lapped transform (MLT). The MCLT is a particular kind of a 2x oversampled generalized DFT filter bank, whose real part corresponds to the MLT. That property can be used for efficient implementation of joint echo cancellation, noise reduction, and coding, for example. Fast algorithms for the MCLT are presented, as well as examples that show the good performance of the MCLT in noise reduction and echo cancellation.

SPTM-8.10

Time-Frequency Mapping Based on Non-Uniform Smoothed Spectral Representations
Panagiotis D Hatziantoniou, Dionysis Tsoukalas, John N Mourjopoulos (Audio Group, Wire Communications Lab., Electrical & Computer Eng. Dept., University of Patras, GREECE), Soterios Salamouris (Roister Speakers)

For audio and acoustic system responses, the auditory system tends to smooth high frequency spectral regions and to register, mainly, low-frequency components of late echoes in the time domain. To model such processing, a theoretical model for non-uniform spectral smoothing is introduced allowing the choice of arbitrary frequency resolution profiles to match such auditory features. This generalized framework is extended to allow mapping of such smoothing spectra into time domain.

< SPTM-7 SPTM-9 >

Last Update: February 4, 1999 Ingo Höntsch