摘要:

The paper explores the use of 2-dimensional cepstral-time features for the utilisation of correlation among successive speech spectral vectors, within a hidden-Markov-model (HMM) framework. A cepstral-time-feature matrix is obtained from a 2-dimensional discrete cosine transform of a spectral-time matrix. Advantages of cepstral-time features are that cepstral-time-feature matrices are a simple and robust method of representing short-time variation of speech spectral parameters; a cepstral-time matrix contains information on the transitional dynamics of feature vectors within the matrix; speech recognition based on cepstral time matrices is more robust in noisy environments; and use of a matrix ofMcepstral vectors implies a minimum HMM-state duration constraint ofMvector units. A simple framework investigated in the paper for applications of cepstral-time features is a finite-state-matrix quantiser (FSMQ), a special case of the HMM. It is used for initialisation of the training phase of HMMs.

DOI:10.1049/ip-i-2.1993.0046

出版商:IEE    

年代:1993

数据来源: IET

摘要:

A newly developed PARCOR coefficients quantisation scheme is presented which is referred to as trellis in-loop PARCOR coefficients quantisation. In the in-loop quantisation scheme, each PARCOR coefficient is determined and quantised inside the analysis computation loop prior to the high order one. Therefore, the quantisation error is taken into account at each stage of lattice filter analysis. This scheme, when combined with trellis structure and associated with appropriate search algorithms, can achieve lower distortions than the conventional scalar quantisation. Simulation results show that by incorporating trellis structure into the in-loop quantisation scheme and using (M, L)-search algorithm withM= 4 andL= 8, an average gain of six bits per frame over the traditional methods can be achieved.

DOI:10.1049/ip-i-2.1993.0047

出版商:IEE    

年代:1993

数据来源: IET

摘要:

A subcodebook searching (SCS) algorithm is developed for fast VQ encoding of images. This algorithm allows searching only a portion of the codebook to find the minimum distortion codeword of an input vector. In comparison with other existing fast VQ encoding algorithms, this method requires the least number of multiplications as well as the least total number of operations. The requirements of extra memory in the encoder and precomputation in the training stage of the SCS algorithm are also limited when the codebook size is not too large.

DOI:10.1049/ip-i-2.1993.0048

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The author presents the theory for developing optimal convolution filters for the identification of wide linear features and an algorithm for distinguishing them from other types of image feature. Some filters are given and the results of applying the algorithm to some real images are shown. These filters, and the algorithm, are appropriate for use in knowledge based systems which aim at identifying roads, canals, hedges and rivers in images, or in digitised maps.

DOI:10.1049/ip-i-2.1993.0049

出版商:IEE    

年代:1993

数据来源: IET

摘要:

A new simple algorithm which allows the design of a trellis diagram for array codes is introduced. Based on this algorithm, the trellis decoding of array codes with maximum likelihood decoding (MLD) performance and relatively low complexity is proposed. It is shown that the trellis structure of array codes allows the application of combined coding and modulation (CCM) with higher information digit per channel symbol rate than for conventional CCM methods, based on convolutional codes. The bounds on the transmission rates for different types of CCM signals, based on convolutional and array codes, have been obtained and presented in the paper.

DOI:10.1049/ip-i-2.1993.0050

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The high throughput achieved by type-II hybrid automatic repeat request (ARQ) protocols can be enhanced by using the inner structure of a codeword in a concatenated coding scheme. The paper presents a new type-II hybrid scheme based on concatenated codes to which two retransmission strategies have been applied. The results on throughput bounds show that the proposed protocols are effective even in the case of communication channels with high error rates.

DOI:10.1049/ip-i-2.1993.0051

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The type-II hybrid automatic repeat request (ARQ) system is an adaptive error control system and has been given considerable attention in the recent years. The main feature of this system is the application of a rate 1/2 block or convolutional error correction code and parity retransmission. The author presents a new parity retransmission system using simple product codes. The throughput efficiency of the proposed system is analysed for the binary symmetric channel (BSC). Numerical results indicate that the proposed system outperforms the type-II hybrid ARQ system.

DOI:10.1049/ip-i-2.1993.0052

出版商:IEE    

年代:1993

数据来源: IET

摘要:

A new multichannel adaptive error-correction system has been investigated for use over noisy time-varying channels. Information is fed at a fixed overall rate over a number of parallel channels, with the amount of information per channel being determined on the basis of channel quality. The system is shown to be superior to that obtainable with a fixed-rate forward error-correction systems, and in addition the proposed system is shown to have a performance superior to that obtainable with classical diversity systems operating at the same rate.

DOI:10.1049/ip-i-2.1993.0053

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The authors discuss how an asymptotically efficient simulation technique, originally proposed for binary communication systems, can be applied also to modulation systems with a large number of signalling waveforms, retaining much of the efficiency of the method. Basically, the probability of error is decomposed into sums of terms corresponding to all possible errors. Then each term is evaluated by a series ofad hocexperiments, designed as if the modulation system were binary. For a given computational effort, each experiment is performed an appropriate number of times to minimise the variance of the estimated probability of error. When the number of waveforms is very large, we show that random selection of the experiments can provide an unbiased form of stochastic truncation of the sum. The paper presents examples of application to block codes, convolutional codes, and trellis coded modulation. These can be easily handled, provided sufficient information about the code structure is available.

DOI:10.1049/ip-i-2.1993.0054

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The authors deal with an efficient high-speed packet switching fabric suitable for applications in future high speed networks. An advanced implementation architecture based on shared buffers at each input is studied. An important result is that the proposed switching fabric achieves optimum throughput-mean switching delay performance jointly with reduced buffering requirements and without having to resort to a faster switching fabric.

DOI:10.1049/ip-i-2.1993.0055

出版商:IEE    

年代:1993

数据来源: IET