• Speech Sciences
• /
• v.6
• /
• pp.219-233
• /
• 1999

This paper presents a real-time implementation of a speaker independent speech recognition system based on a discrete hidden markov model(DHMM). This system is developed for a car navigation system to design on-chip VLSI system of speech recognition which is used by fixed point Oak DSP core of DSP GROUP LTD. We analyze recognition procedure with C language to implement fixed point real-time algorithms. Based on the analyses, we improve the algorithms which are possible to operate in real-time, and can verify the recognition result at the same time as speech ends, by processing all recognition routines within a frame. A car noise is the colored noise concentrated heavily on the low frequency segment under 400 Hz. For the noise robust processing, the high pass filtering and the liftering on the distance measure of feature vectors are applied to the recognition system. Recognition experiments on the twelve isolated command words were performed. The recognition rates of the baseline recognizer were 98.68% in a stopping situation and 80.7% in a running situation. Using the noise processing methods, the recognition rates were enhanced to 89.04% in a running situation.

• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.119-122
• /
• 2000

Recently, many complex DSP (Digital Signal Processing) algorithms have being realized on RISC CPU due to good compilation, low power consumption and large memory space. But, real-time implementation of multiple DSP algorithms on RISC requires the minimum and efficient memory usage and the lower occupancy of CPU. In this thesis, the original floating-point code of MPEG-1 audio decoder is converted to the fixed-point code and then optimized to the efficient assembly code in time-consuming function in accord with RISC feature. Finally, compared with floating-point and fixed-point, about 30 and 3 times speed enhancements are achieved respectively. And 3~4 times memory spaces are spared.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.2
• /
• pp.365-374
• /
• 1996

Finite wordlength effects of the FS-1016 CELP(Code Excite Linear Prediction) vocoder algorithm) is analyzed, and a block floating-Point implementation method is employed to improve the fixed-point performance. An efficient run-time integer wordlength estimation algorithm is developed, and the overall system performance. An efficient run-time integer wordlength estimation algorithm is developed, and the overall system performance is verified in real-time using a TMS320C50 emulation board. Autoscaler software that conducts simulation-based automatic scaling to provide a floating-point like programming environment is used for this application development.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.1211-1214
• /
• 1998

Real-time implementation of Conjugate-Structure Algebraic CELP(CS-ACELP) is presented. ITU-T Study Group(SG) 15 has standardized the CS-ACELP speech coding algorithm as G.729. A real-time implementation of the CS-ACELP is achieved using 16 bit fixed point DSP16210 Digital Signal Processor (DSP) of Lucent Technologies. The speech coder has been implemented in the bit-exact manner using the fixed point CS-ACELP C source which is the part of the G.729 standard. To provide a multi-channel vocoder solution to digital communication system, we try to minimize the complexity(e.g., MIPS, ROM, RAM) of CS-ACELP. Our speech coder shows 15.5 MIPS in performance which enables 4 channel CS-ACELP to be processed with one DSP16210.

• Journal of Satellite, Information and Communications
• /
• v.7 no.1
• /
• pp.92-96
• /
• 2012

This paper presents return link demodulation algorithm designed for hardware implementation using HDL that include complex signal processing work reliably even in low SNR. Simulation results show that performance of Es/$N_0$ has improved within 0.5dB at the point of uncoded BER $10^{-3}$ compared to the ideal QPSK signal. In addition, fixed-point simulation and HDL implementation performance compared to the simulation we can see that there is no difference.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.1
• /
• pp.95-103
• /
• 2009

We improve the performance of WCDMA repeater by cancelling the feedback interference radio signal under the fixed point implementation. Floating-point DSP or FPGA to implement the ICS algorithm may have an disadvantage of high cost, To solve this problem, we suggest the ICS algorithm based on LMS under fixed point operation, and show the validity of our results by comparing with the floating-point results through numerical simulation.

• Proceedings of the IEEK Conference
• /
• 2003.07a
• /
• pp.517-520
• /
• 2003

Orthogonal Frequency Division Multiplexing (OFDM) is a promising technology for high speed multimedia communication in a frequency selective multipath channel. In this paper, Software IPs for the synchronizer of IEEE 802.11a Wireless LAN system are designed and optimized for TI's TMS320C6201 fixed point DSP. As a result of the execution cycles of the target DSP for each functions of the system, an efficient HW/SW partitioning method can be considered.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.1
• /
• pp.1-8
• /
• 2018

Recently, the convolutional neural network shows high performance in many computer vision tasks. However, convolutional neural networks require enormous amount of operation, so it is difficult to adopt them in the embedded environments. To solve this problem, many studies are performed on the ASIC or FPGA implementation, where an efficient representation method is required. The fixed-point representation is adequate for the ASIC or FPGA implementation but causes a performance degradation. This paper proposes a separate optimization of representations for the convolutional layers and the batch normalization layers. With the proposed method, the required bit width for the convolutional layers is reduced from 16 bits to 10 bits for the ResNet-50 neural network. Since the computation amount of the convolutional layers occupies the most of the entire computation, the bit width reduction in the convolutional layers enables the efficient implementation of the convolutional neural networks.

• The KIPS Transactions:PartA
• /
• v.16A no.3
• /
• pp.159-168
• /
• 2009

In this paper, we present an efficient way of implementing OpenGL ES 1.1 standard for the environments with hardware-supported OpenGL API, such as desktop PCs. Although OpenGL ES was started from the existing OpenGL features, it becomes a new three-dimensional graphics library customized for embedded systems through introducing fixed-point arithmetic operations, buffer management with fixed-point data type supports, completely new texture mapping functionalities and others. Currently, it is the official three dimensional graphics library for Google Android, Apple iPhone, PlayStation3, etc. In this paper, we achieved improvements on the arithmetic operations for the fixed-point number representation, which is the most characteristic data type for OpenGL ES. For the conversion of fixed-point data types to the floating-point number representations for the underlying OpenGL, we show the way of efficient conversion processes even with satisfying OpenGL ES standard requirements. We also introduced a simple memory management scheme to mange the converted data for the buffer containing fixed-point numbers. In the case of texture processing, the requirements in both standards are quite different and thus we used completely new software-implementations. Our final implementation result of OpenGL ES library provides all of over than 200 functions in OpenGL ES 1.1 standard and completely passed its conformance test, to show its compliance with the standard. From the efficiency viewpoint, we measured its execution times for several OpenGL ES-specific application programs and achieved at most 33.147 times improvements, to become the fastest one among the OpenGL ES implementations in the same category.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.11C
• /
• pp.1491-1497
• /
• 2004

The psychoacoustic model in MPEG audio layer-III (MP3) encoder is optimized for the fixed-point processing. The optimization process consists of determining the data word length of arithmetic unit and the algorithm for transcendental functions that are often used in the psychoacoustic model. In order to determine the data word length, we defined a statistical model expressing the relation between the fixed-point operation errors of the psychoacoustic model and the probability of alteration of the allocated bits doe to these errors. Based on the simulations using this model, we chose a 24-bit data path and constructed a 24-bit fixed-point MP3 encoder. Sound quality tests using the constructed fixed-point encoder showed a mean degradation of -0.2 on ITU-R 5-point audio impairment scale.