The most commonly used method for training feature weights in statistical machine translation (SMT) systems is Och’s minimum error rate training (MERT) procedure. A well-known problemwith Och’s procedure is that it tends to be sensitive to small changes in the system, particularly when the number of features is large. In this paper, we quantify the stability of Och’s procedure by supplying diff...

The equalisation topic is well researched and a variety of solutions are available. The MAP sequence detector provides the lowest symbol error rate (SER) attainable, and the MLSE offers a near optimal solution. However, these optimal techniques are not yet practical for high-level modulation schemes, due to their computational complexity. Linear equaliser or linear-combiner DFE are practical sc...

Minimum Error Rate Training (MERT) remains one of the preferred methods for tuning linear parameters in machine translation systems, yet it faces significant issues. First, MERT is an unregularized learner and is therefore prone to overfitting. Second, it is commonly used on a noisy, non-convex loss function that becomes more difficult to optimize as the number of parameters increases. To addre...

In this paper development of a multimodal based biometric fusion system is discussed. A fingerprint recognition system is developed using global singularity features. Mel-frequency Cepstral Coefficients are used to recognise a speaker using the backpropagation artificial neural network. A score level fusion based recognition system is developed using fingerprint and speech match scores and the ...

We consider a multiple antenna system when combined array processing with space–time coding is used. We present variable rate space–time block codes for two, three, and four transmit antennas and optimize the transmit power so that the average bit-error rate (BER) is minimized. Numerical results show that this optimum power allocation scheme provides significant gain over the equal power alloca...