We propose ThaiLMCut, a semi-supervised approach for Thai word segmentation which utilizes a bi-directional character language model (LM) as a way to leverage useful linguistic knowledge from unlabeled data. After the language model is trained on substantial unlabeled corpora, the weights of its embedding and recurrent layers are transferred to a supervised word segmentation model which continues fine-tuning them on a word segmentation task. Our experimental results demonstrate that applying the LM always leads to a performance gain, especially when the amount of labeled data is small. In such cases, the F1 Score increased by up to 2.02%. Even on abig labeled dataset, a small improvement gain can still be obtained. The approach has also shown to be very beneficial for out-of-domain settings with a gain in F1 Score of up to 3.13%. Finally, we show that ThaiLMCut can outperform other open source state-of-the-art models achieving an F1 Score of 98.78% on the standard benchmark, InterBEST2009.

The complexity-performance trade-off is a fundamental aspect of the design of low-density parity-check (LDPC) codes. In this paper, we consider LDPC codes for the binary erasure channel (BEC), use code rate for performance metric, and number of decoding iterations to achieve a certain residual erasure probability for complexity metric. We first propose a quite accurate approximation of the number of iterations for the BEC. Moreover, a simple but efficient utility function corresponding to the number of iterations is developed. Using the aforementioned approximation and the utility function, two optimization problems w.r.t. complexity are formulated to find the code degree distributions. We show that both optimization problems are convex. In particular, the problem with the proposed approximation belongs to the class of semi-infinite problems which are computationally challenging to be solved. However, the problem with the proposed utility function falls into the class of semi-definite programming (SDP) and thus, the global solution can be found efficiently using available SDP solvers. Numerical results reveal the superiority of the proposed code design compared to existing code designs from literature.

By means of a signal representation in a signal space spanned by a few time-limited exponential basis-functions, and data-aided maximum-likelihood techniques, novel methods for the joint and nonjoint estimation of carrier-phase and symbol-timing for general coherent CPM (continuous phase modulation) receivers are proposed. The estimators are quite simple, especially in obtaining symbol-timing. Simulation results show that synchronizers using such estimation structures exhibit an excellent steady-state behavior. Whereas a fast acquisition of the carrier-phase is always possible, joint synchronizer loops tend to hang up for CPM-schemes with partial response signaling. Several ways to overcome the hang-up problem are discussed.>

By separating the two complexity problems connected with optimum coherent CPM (continuous phase modulation) receivers, namely, the number of linear filters and the number of memory states required, a straightforward procedure for reducing the complexity is proposed and its usefulness is evaluated. A concise analysis of the inherent trellis encoder associated with CPM is given. It is then demonstrated that for almost all schemes of interest in practice, it is quite sufficient for the receiver to implement only four or six linear filters which represent proper reference signals. For a reduction in the number of memory states, decision-feedback sequence estimation together with minimization of the unprocessed intersymbol interference is proposed. Modifications of this procedure allow a state reduction without losses or with only negligible losses because only error events with large distances are affected. Combinations of these methods make possible an almost continuous tradeoff between receiver complexity and SNR (signal/noise ratio) losses. Several examples are considered for which evaluations of minimum Euclidean distances and simulation results are given.>