• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.101-107
• /
• 2009

Growth in the VLSI process and design technology is resulting into a continuous increase in the number of IPs on a chip to form a system. Because of many IPs on a single chip, efficient communication between IPs is essential. We propose a dynamically reconfigurable 3-layer bus structure which can adapt to the pattern of data transmission to achieve an efficient data communication between various IPs. The proposed 3-layer bus can be reconfigured to multi-single bus mode, and single-multi bus mode, thus providing the benefits of both single-bus and multi-bus modes. Experimental results show that the flexibility of the proposed bus structure can reduce data transmission time compared to the conventional fixed bus structure. We incorporated the proposed bus structure in a JPEG system and verified that the proposed structure achieved an average of 22% improvement in time over the conventional fixed bus structure.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2632-2640
• /
• 2011

In this paper, we implemented a memory controller which can accommodate data processing blocks. The memory controller is arbitrated by the internal arbiter which receives request signals from masters and sends grant and data signals to masters. The designed memory controller consists of Master Interface, Master Arbitrator, Memory Interface, Memory accelerator. It was designed using VHDL, and verified using the memory model of SAMSING Inc. For FPGA synthesis and verification, Quartus II of ATERA Inc. was used. The target device is Cyclone II. For simulation, ModelSim of Cadence Inc was used.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.6
• /
• pp.301-309
• /
• 2001

본 논문에서는 RSA 암호 시스템의 핵심 과정인 모듈로 멱승 연산에 대한 새로운 하드웨어 구조를 제시한다. 본 방식은 몽고메리 곱셈 알고리즘을 사용하였으며 기존의 방법들이 데이터 종속 그래프(DG : Dependence Graph)를 수직으로 매핑한 것과는 달리 여기서는 수평으로 매핑하여 1차원 선형 어레이구조를 구성하였다. 그 결과로 멱승시에 중간 결과값이 순차적으로 나와서 바로 다음 곱셈을 위한 입력으로 들어갈 수 있기 때문에 100%의 처리율(throughput)을 이룰 수 있고, 수직 매핑 방식에 비해 절반의 클럭 횟수로 연산을 해낼 수 있으며 컨트롤 또한 단순해지는 장점을 가진다. 각 PE(Processing Element)는 2개의 전가산기와 3개의 멀티플렉서로 이루어져 있고, 암호키의 비트수를 k비트라 할 때 k+3개의 PE만으로 파이프라인구조를 구현하였다. 1024비트 RSA데이터의 암호 똔느 복호를 완료하는데 2k$^2$+12k+19의 클럭 수가 소요되며 클럭 주파수 100Mhz에서 약 50kbps의 성능을 보인다. 또한, 제안된 하드웨어는 내부 계산 구조의 지역성(locality), 규칙성(regularity) 및 모듈성(modularity) 등으로 인해 실시간 고속 처리를 위한 VLSI 구현에 적합하다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1992.04b
• /
• pp.415-424
• /
• 1992

본 논문에서는 정수선형계획법을 반도체 설계자동화과정에 이용한 예를 보인다. 반도체 설계자동화과정은 매우 여러 단계를 거치게 되는데, 본 논문에서는 상위수준 합성중 스케쥴링(scheduling)문제에 정수선형계획법을 응용하였다. 여기서 스케쥴링 문제는 설계자동화의 초기단계에서 알고리듬으로 주어진 입력을 하드웨어 요소들로 표현하는 과정에서 매 제어단계(control step)에서 수행하여야 할 연산내용을 결정하는 문제이다. 스케쥴링의 목적함수는 주어진 제어단계 갯수내에서 하드웨어 비용의 최소화이다. 이를 위해 우선 ASAP(As Soon As Possible)과 ALAP(As Late As Possible)방법을 이용하여 매 연산의 수행시작이 가능한 가장 빠른 시간과 가장 늦은 시간을 구한다. 이 두 시간 사이가 각 연산의 time frame이 되며 이를 이용하여 스케쥴링 문제를 정수 선형 계획법으로 공식화하여 풀었다. 이 공식화는 chaining, multicycle연산, pipeline data path, pipeline기능 유닛등에도 일반화하여 적용가능함을 보인다. 실험을 통해 본 공식화 방법이 기존 알고리듬에 의한 해보다 우수한 해를 제공함을 보인다. 비교를 위해 잘 알려진 benchmark회로인 bandpass filter를 이용하였는데 이 회로는 8개의 덧셈, 7개의 뺄셈 및 12개의 곱셈연산을 포함하고 있다. 제시된 알고리듬은 이 회로를 8개의 제어단계내에 총비용 675 (연산별 하드웨어 비용은 라이브러리로 주어짐)로 스케쥴링하였는데 이는 기존의 최상의 결과인 685보다 우수한 결과이다.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2004.11a
• /
• pp.666-670
• /
• 2004

VLSI 소자에서 design rule(D/R)이 작아져 각 단위 Pattern의 size가 작아짐에 따라 aspect ratio가 커지게 되었다. 산화막 contact etch를 하는데 있어 산화막 측벽을 보호하는데, 이러한 보호막은 주로 fluoro-carbon 계열의 polymer precursor들이 사용된다. Aspect ratio(A/R)가 5 이하일 때에는 측벽의 보호막에 의한 바닥 변형이 문제가 되지 않으나, 10 이상의 A/R를 가진 contact에서는 크기가 줄고, 모양이 불균형하게 변하는 바닥 변형을 쉴게 관찰할 수 있다. 이러한 바닥 변형이 커지면 contact 저항이 높아지는 것은 물론이고, 심하게는 하부 pattern과 overlap 불량을 유발할 수 있다. 본 논문에서는 바닥변형을 일으키는 원인을 분석하고 fluoro-carbon 계열의 polymer precursor의 종류$(C_4_F6\;vs.\;C_3F_8)$에 따른 polymer증착 상태 확인 및 pattern비대칭에 따른 바닥 변형의 고찰과 plasma etching 시 H/W 변형을 통해 바닥 변형이 거의 없는 조건을 찾아낼 수 있었다.

• 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.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.41 no.7
• /
• pp.71-79
• /
• 2004

Packet-type data transmission is characterized by the continuous transmission of massive data with relatively constant rate. In such transmission, the dynamic power consumed on buses is influenced by the sequence of transmitted data. A new coding scheme called Sequence-Switch Coding (SSC) is proposed in this paper. SSC reduces the number of bus transitions in the transmission of packet-type data by changing the sending order of the data. Some simple algorithms are presented, In. The simulation results show that SSC outperforms the well-known Bus-Invert Coding with these algorithms. SSC is not a specific algerian but a method to reduce the number of bus-transitions. There could be lots of algorithms for realizing SSC. The variety of SSC algorithms provides circuit designers a wide range of trade-off between performance and circuit complexity.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.39 no.7
• /
• pp.89-96
• /
• 2002

In this paper, the ADI(Adaptive De-interlacing) algorithm is proposed, which improves visually and subjectively horizontal and vertical edges of the image processed by the ELA(Edge Line-based Average) method. This paper also proposes a VLSI architecture for the proposed algorithm and the architecture designed through the full custom CMOS layout process. The proposed algorithm is verified using C and Matlab and implemented using $0.6{\mu}m$ 2-poly 3-metal CMOS standard libraries. For the circuit and logic simulation, Cadence tool is used.

• ETRI Journal
• /
• v.21 no.4
• /
• pp.1-8
• /
• 1999

Thick metal 0.8${\mu}m$ CMOS technology on high resistivity substrate(RF CMOS technology) is demonstrated for the L-band RF IC applications, and we successfully implemented it to the monolithic 900 MHz and 1.9 GHz CMOS LNAs for the first time. To enhance the performance of the RF circuits, MOSFET layout was optimized for high frequency operation and inductor quality was improved by modifying the technology. The fabricated 1.9 GHz LNA shows a gain of 15.2 dB and a NF of 2.8 dB at DC consumption current of 15mA that is an excellent noise performance compared with the offchip matched 1.9 GHz CMOS LNAs. The 900 MHz LNA shows a high gain of 19 dB and NF of 3.2 dB despite of the performance degradation due to the integrating of a 26 nH inductor for input match. The proposed RF CMOS technology is a compatibel process for analog CMOS ICs, and the monolithic LNAs employing the technology show a good and uniform RF performance in a five inch wafer.

• ETRI Journal
• /
• v.39 no.3
• /
• pp.326-335
• /
• 2017

In the uplink transmission of massive (or large-scale) multi-input multi-output (MIMO) systems, large dimensional signal detection and its hardware design are challenging issues owing to the high computational complexity. In this paper, we propose low-complexity hardware architectures of Richardson iterative method-based massive MIMO detectors. We present two types of massive MIMO detectors, directly mapped (type1) and reformulated (type2) Richardson iterative methods. In the proposed Richardson method (type2), the matrix-by-matrix multiplications are reformulated to matrix-vector multiplications, thus reducing the computational complexity from $O(U^2)$ to O(U). Both massive MIMO detectors are implemented using a 65 nm CMOS process and compared in terms of detection performance under different channel conditions (high-mobility and flat fading channels). The hardware implementation results confirm that the proposed type1 Richardson method-based detector demonstrates up to 50% power savings over the proposed type2 detector under a flat fading channel. The type2 detector indicates a 37% power savings compared to the type1 under a high-mobility channel.