Fuzzy systems design: direct and indirect approaches

A systematic classification of the data-driven approaches for design of fuzzy systems is given in the paper. The possible ways to solve this modelling and identification problem are classified on the basis of the optimisation techniques used for this purpose. One algorithm for each of the two basic categories of design methods is presented and its advantages and disadvantages are discussed. Both types of algorithms are self-learning and do not require interaction during the process of fuzzy model design. They perform adaptation of both the fuzzy model structure (rule-base) and the parameters. The indirect approach exploits the dual nature of Takagi-Sugeno (TS) models and is based on recently introduced recursive clustering combined with Kalman filtering-based procedure for recursive estimation of the parameter of the local sub-models. Both algorithms result in finding compact and transparent fuzzy models. The direct approach solves the optimisation problem directly, while the indirect one decomposes the original problem into on-line clustering and recursive estimation problems and finds a sub-optimal solution in real-time. The later one is computationally very efficient and has a range of potential applications in real-time process control, moving images recognition, autonomous systems design etc. It is extended in this paper for the case of multi-input–multi-output (MIMO systems). Both approaches have been tested with real data from an engineering process.     

Envelope Dynamic Ratio Quantizer

The envelope dynamic ratio quantizer (envelope DRQ) is an adaptive quantizer for speech signals. By utilizing the envelope of the speech, nonlinear elements, a fixed quantizer and a simple predictor, a closed-loop adaptive quantizer emerges having a high constant SNR over a wide dynamic range. The theory of the quantizer is presented, together with computer simulation results which show an improvement.compared to the one word memory APCM system. Finally, the simplicity of implementing the envelope DRQ is described.     

Sequential Adaptive Predictors for ADPCM Speech Encoders

The sequential gradient estimation predictor is compared in detail to the stochastic approximation predictor, and both are evaluated in an ADPCM codec. A switched predictor having two coefficients is then described for use in a DPCM-AQF codec. This predictor divides the range of the correlation coefficient of the speech signal into zones, and as the correlation coefficient changes zones, the predictor coefficients undergo a substantial modification. By this method the adaptation rate of the predictor is improved, particularly during transitions between unvoiced and voiced sounds.     

Pitch synchronous 1st-order linear d.p.c.m. system

A system called p.s.f.o.l.d. is described which exploits the correlation between successive pitch periods of a speech signal. This system is a differential one and can employ various types of encoders. We describe a p.s.f.o.l.d. system using a 1st-order d.p.c.m. encoder and show that for a speech utterance this system has a peak signal/noise ratio which is 6 dB larger, and has an increase in dynamic range of 13 dB, compared with a 1st-order d.p.c.m. codec.     

Pitch synchronous differential predictive encoding system

A pitch synchronous differential predictive encoding system (p.s.d.p.e.) is described, which reduces the dynamic range of voiced speech to a value similar to that of unvoiced speech. As a consequence, the signal encoded has a smaller dynamic range than the speech signal and results in an improvement in the signal/noise ratio for a given transmitted number of bits per sample. This improvement is approximately 8 dB compared with an a.d.p.c.m. codec, when the p.s.d.p.e. system uses an adaptive p.c.m. encoder and the transmission rate is 3 bit/sample.     

Simulation of photon-limited images using video data

Poisson or shot noise is a major degrading factor in low-light and infrared imaging. The authors show how images from a standard video camera can be artificially degraded to simulate the effect of Poisson noise. A specific algorithm is given, together with details of the computational cost     

Noise reduction in ADPCM AQJ systems using quantiser correction at the receiver

A technique is considered which reduces the output noise in ADPCM-AQJ speech coders. Computer simulation results indicate a subjectively perceptible enhancement in the quality of the recovered speech.     

Gradient-based multiresolution image fusion

A novel approach to multiresolution signal-level image fusion is presented for accurately transferring visual information from any number of input image signals, into a single fused image without loss of information or the introduction of distortion. The proposed system uses a "fuse-then-decompose" technique realized through a novel, fusion/decomposition system architecture. In particular, information fusion is performed on a multiresolution gradient map representation domain of image signal information. At each resolution, input images are represented as gradient maps and combined to produce new, fused gradient maps. Fused gradient map signals are processed, using gradient filters derived from high-pass quadrature mirror filters to yield a fused multiresolution pyramid representation. The fused output image is obtained by applying, on the fused pyramid, a reconstruction process that is analogous to that of conventional discrete wavelet transform. This new gradient fusion significantly reduces the amount of distortion artefacts and the loss of contrast information usually observed in fused images obtained from conventional multiresolution fusion schemes. This is because fusion in the gradient map domain significantly improves the reliability of the feature selection and information fusion processes. Fusion performance is evaluated through informal visual inspection and subjective psychometric preference tests, as well as objective fusion performance measurements. Results clearly demonstrate the superiority of this new approach when compared to conventional fusion systems.     

Frequency Compression of 7.6 kHz Speech Into 3.3 kHz Bandwidth

Telephone channels restrict the bandwidth of speech signals to approximately 0.3-3.3 kHz, with the consequence that the intelligibility of unvoiced sounds may be significantly impaired. To prevent this band limitation of unvoiced sounds while still confining the speech to the telephonic bandwidth, we propose a scheme which, on recognizing the presence of unvoiced sounds extending to 7.6 kHz, frequency maps them into the band 0.3-3.3 kHz. Four mapping laws are considered and the unvoiced speech is compressed using each law. Frequency demapping is employed, and the law that has the best spectral match to the speech spectrum is selected. Voiced speech is band limited from 0.3 to 3.3 kHz. Results measured over 16 ms, a phoneme, and word durations indicate that the adaptive frequency mapping algorithm significantly enhances the recovered speech compared to telephonic speech. Informal listening experiences support these findings.     

Embedding Data Into Pictures by Modulo Masking

Three systems are proposed for embedding data into industrial quality monochrome analog pictures. The video signal on each scan line is sampled, and a data bit is inserted into a block of three or five pels by modulo masking scrambling the luminance level of only one pel in the block. Prior to transmission, the combined data and video sequence is converted into a continuous signal with a bandwidth that is no greater than that of the original video signal. Using six images each containing 65 536 pels, Systems 1 and 2 embedded an average of 17 430 and 8713 bits per image, while System 3 accommodated data at a constant rate of 21 760 bits/image. The data embedding procedures of Systems 1, 2, and 3 operated with average picture SNR's of 41, 44, and 30 dB, respectively, when the transmission channel was ideal. When the transmission was over a channel composed of a second-order Butterworth filter plus additive noise that yield a channel SNR of 40 dB, no bit errors occurred but System 3 offered the greater safety margin to bit errors than Systems 1 and 2.