It is very important to correct the synchronization errors that occur if the receiver is not completely synchronized with the transmitter. However, it is impossible to achieve perfect synchronization in many communication systems.

Synchronization errors due to the insertion and deletion of symbols can cause serious performance degradation if the coding scheme does not possess the capability to recover from such errors.

Pak Jang Son, a researcher at the Faculty of Communications, has proposed a method for correcting synchronization errors that occur in Turbo code sequence using the algebraic operation principle of Scarce-State-Transition (SST) Viterbi decoder.

The encoding/decoding procedures of Turbo code in this method are similar to the common case, except making corrupted bit positions caused by bit insertion/deletion in received Turbo code sequence take their original positions.

The simulation result shows that the proposed method is capable of correcting synchronization errors occurring in the received Turbo code sequences.

Unlike the Optical Character Recognition (OCR) in traditional document images, video subtitle text recognition is still a challenging task in the fields of computer vision and artificial intelligence because of complex background, different size and color of fonts, low contrast between texts and background regions and different location of subtitle texts in video frames. Due to these factors, the traditional OCR developed for document images has proved to be inappropriate for video subtitle texts.

In order to solve this problem, unified models for both detecting and recognizing texts have been proposed. However, these methods have the disadvantage of low accuracy in the recognition of unknown words or sequences of arbitrary characters.

Yun Chol Song, a researcher at the Faculty of Information Science and Technology, has designed and implemented two types of models―one is a video subtitle text detection model, a combination of a convolutional neural network model and an adaptive scale fusion model, and the other is a subtitle text recognition model, a combination of a convolutional neural network model and a BLSTM model.

The results of the experiments have demonstrated that the subtitle text detection accuracy in the video is 93.4% and the subtitle text recognition accuracy is 94.6%, which proves to be superior to the previous methods.

Since the rotary angular sensor is a main sensor for measuring accurate positions and directions, it plays an important role in various applications such as robot control, cameras and industrial machines.

Generally, the rotary angular sensor can be divided into sensors based on the changes in electric field and magnetic field, and optical sensors according to the underlying physical principle. Electric field-based and magnetic field-based sensors have their own drawbacks respectively.

However, optical sensors operating in visible or infrared light have the advantage of contactless measurements and insensitivity to electromagnetic fields. The sensitivity of an optical sensor might be reduced by external light, but this problem can be solved using a housing and pulsed light. Today, optical angular sensors are indispensable displacement/position sensors for their high resolution, light weight and insensitivity to electromagnetic fields.

Jo Myong Jin, a researcher at the Semiconductor Institute, has proposed a new absolute rotary angular sensor using 1D PSD (position sensitive detector).

Sinusoidal signals with rotation angles are output from the 1D PSD by the sinusoidal path of the optical disc, which allows easy reduction of 1-bit measurement errors present in the binary code disc.

The proposed method is simple in design and the output signal is a combination of an absolute binary code and an analog signal, which can achieve high resolution. In addition, it has the advantage of high linearity and accuracy.