• Journal of the Korea Society of Computer and Information
• /
• v.17 no.6
• /
• pp.1-11
• /
• 2012

As the process technology scales down, multi-core processors cause serious problems such as increased interconnection delay, high power consumption and thermal problems. To solve the problems in 2D multi-core processors, researchers have focused on the 3D multi-core processor architecture. Compared to the 2D multi-core processor, the 3D multi-core processor decreases interconnection delay by reducing wire length significantly, since each core on different layers is connected using vertical through-silicon via(TSV). However, the power density in the 3D multi-core processor is increased dramatically compared to that in the 2D multi-core processor, because multiple cores are stacked vertically. Unfortunately, increased power density causes thermal problems, resulting in high cooling cost, negative impact on the reliability. Therefore, temperature should be considered together with performance in designing 3D multi-core processors. In this work, we analyze the temperature of the cache in quad-core processors varying cache organization. Then, we propose the low-temperature cache organization to overcome the thermal problems. Our evaluation shows that peak temperature of the instruction cache is lower than threshold. The peak temperature of the data cache is higher than threshold when the cache is composed of many ways. According to the results, our proposed cache organization not only efficiently reduces the peak temperature but also reduces the performance degradation for 3D quad-core processors.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.7 s.349
• /
• pp.1-13
• /
• 2006

In this paper, we introduce a novel key management scheme that is based on the key pre-distribution but provides the key re-distribution method, in order to manage keys for message encryption and authentication of lower-layer sensor nodes on hierarchical mobile sensor networks. The characteristics of our key management are as follows: First, the role of key management is distributed to aggregator nodes as well as a sink node, to overcome the weakness of centralized management. Second, a sink node generates keys using regression model, thus it stores only the information for calculating the keys using the key information received from nodes, but does not store the relationship between a node and a key, and the keys themselves. As the disadvantage of existing key pre-distributions, they do not support the key re-distribution after the deployment of nodes, and it is hard to extend the key information in the case that sensor nodes in the network enlarge. Thirdly, our mechanism provides the resilience to node capture(${\lambda}$-security), also provided by the existing key pre-distributions, and fourth offers the key freshness through key re-distribution, key distribution to mobile nodes, and scalability to make up for the weak points in the existing key pre-distributions. Fifth, our mechanism does not fix the relationship between a node and a key, thus supports the anonymity and untraceability of mobile nodes. Lastly, we compare ours with existing mechanisms, and verify our performance through the overhead analysis of communication, computation, and memory.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.7
• /
• pp.761-768
• /
• 2010

This paper presents an effective method to achieve the wideband waveform for high resolution SAR(Synthetic Aperture Radar) using the frequency multiplication technique. And also this paper analyzes the root causes for the spectral regrowth due to 3rd-order intermodulation in chirp bandwidth expansion scheme using quadrature modulator and frequency multipliers. The amplitude and phase imbalance requirement are defined based on the simulation results in terms of quadrature channel imbalance. This minimizes the degradation of range resolution, peak sidelobe ratio and integrated sidelobe ratio. The wideband chirp generator using the frequency multiplier and memory map scheme was manufactured and the compensation technique was presented to reduce the spectral regrowth of SAR waveform by minimizing the amplitude and phase imbalance. After I and Q channel imbalance adjustment, the carrier level reduces －28.7 dBm to －53.4 dBm. Chirp signal with 150 MHz bandwidth at S-band expands to 600 MHz bandwidth at X-band. The sidelobe levels are reduced by about 8 to 9 dB by compensating the amplitude balance between I and Q channels.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.2B
• /
• pp.36-43
• /
• 2005

This paper suggests a method to effectively apply an application model of fuzzy-neural network to the optimal load distribution algorithm, considering the complication and non-linearity of the web server environment. We use the clustering web server in the linux system and it consists of a load balancer that distributes the network loads and some of real servers that processes the load and responses to the client. The previous works considered only with the scrappy decision information such as the connections. That is, since the distribution algorithm depends on the input of the whole network throughput, it was proved inefficient in terms of performance improvement of the web server. With the proposed algorithm, it monitors the whole states of both network input and output. Then, it infers CPU and memory states of each real server and effectively distributes the requests of the clients. In this paper, the proposed model is compared with the previous method through simulations and we analysis the results to develop the optimal and intelligent load balancing model.

• Design & Manufacturing
• /
• v.11 no.2
• /
• pp.51-56
• /
• 2017

Nitinol, a shape memory alloy (SMA), is manufactured from titanium and nickel and it used in various fields such as electrical applications, micro sensors. It is also recommended as a material in medical for implant because it has excellent organic compatibility. Nitinol is intended to be inserted into the human body, products require a high-quality surface and low residual stress. To overcome this problems, explore electrolyte polishing (EP) is being explored that may be appropriate for use with nitinol. EP is a particularly useful machining method because, as a non contact machining method, it produces neither machining heat nor internal stress in the machined materials. Sandpaper polishing is also useful machining method because, as a contact machining method, it can easily good surface roughness in the machined materials. The electrolyte polishing (EP) process has an effect of improving the surface roughness as well as the film polishing process, but has a characteristic that the residual stress is hardly generated because the work hardened layer is not formed on the processed surface. The sandpaper polishing process has the effect of improving the surface roughness but the residual stress remains in the surface. We experimented with three conditions of polishing process. First condition is the conventional polishing. Second condition is the electrochemical polishing(EP). And Last condition is a mixing process with the conventional polishing and the EP. Surface roughness and residual stress of the nitinol before a polishing process were $0.474{\mu}mRa$, -45.38MPa. Surface roughness and residual stress of the nitinol after mixing process of the conventional polishing and the EP were $1.071{\mu}mRa$, -143.157MPa. Surface roughness and residual stress of the nitinol after conventional polishing were $0.385{\mu}mRa$ and -205.15MPa. Surface roughness and residual stress of sandpaper and EP nitinol were $1.071{\mu}mRa$, -143.157MPa. The result shows that the EP process is a residual stress free process that eliminates the residual stress on the surface while eliminating the deformed layer remaining on the surface through composite surface machining rather than single surface machining. The EP process can be used for biomaterials such as nitinol and be applied to polishing of wafers and various fields.

• Journal of Biosystems Engineering
• /
• v.29 no.4
• /
• pp.341-346
• /
• 2004

Recent livestock people concern not only increase of production, but also superior quality of animal-breeding environment. So far, the optimization of the breeding and air environment has been focused on the production increase. In the very near future, the optimization will be emphasized on the environment for the animal welfare and health. Especially, cattle farming demands the precision livestock farming and special attention has to be given to the management of feeding, animal health and fertility. The management of individual animal is the first step for precision livestock farming and animal welfare, and recognizing each individual is important for that. Though electronic identification of a cattle such as RFID(Radio Frequency Identification) has many advantages, RFID implementations practically involve several problems such as the reading speed and distance. In that sense, computer vision might be more effective than RFID for the identification of an individual animal. The researches on the identification of cattle via image processing were mostly performed with the cows having black-white patterns of the Holstein. But, the native Korean and Japanese cattle do not have any definite pattern on the body. The purpose of this research is to identify the Japanese black cattle that does not have a body pattern using computer vision technology and neural network algorithm. Twelve heads of Japanese black cattle have been tested to verify the proposed scheme. The values of input parameters were specified and then computed using the face images of cattle. The images of cattle faces were trained using associate neural network algorithm, and the algorithm was verified by the face images that were transformed using brightness, distortion, and noise factors. As a result, there was difference due to transform ratio of the brightness, distortion, and noise. And, the proposed algorithm could identify 100% in the range from -3 to +3 degrees of the brightness, from -2 to +4 degrees of the distortion, and from 0% to 60% of the noise transformed images. It is concluded that our system can not be applied in real time recognition of the moving cows, but can be used for the cattle being at a standstill.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.06a
• /
• pp.257-257
• /
• 2007

고밀도 FeRAM (Ferroe!ectric Random Access Memory) 소자를 개발하기 위해서는 강유전체 물질을 이용한 안정적인 스텍형의 커패시터 개발이 필수적이다. 특히 $(Bi,La)_4Ti_3O_{12}$ (BLT) 강유전체 물질을 이용하는 경우에는 낮은 열처리 온도에서도 균질하고 높은 값의 잔류 분극 값을 확보하는 것이 가장 중요한 과제 중의 하나이다. 불행히도, BLT 물질은 a-축으로는 약 $50\;{\mu}C/cm^2$ 정도의 높은 잔류 분극 값을 갖지만, c-축 방향으로는 $4\;{\mu}C/cm^2$ 정도의 낮은 잔류 분극 값을 나타내는 동의 강한 비등방성 특성을 보인다. 따라서 BLT 박막에서 각각 입자들의 크기 및 결정 방향성을 세밀하게 제어하는 것은 무엇보다 중요하다. 본 연구에서는 16 Mb의 1T/1C (1-transistor/1-capacitor) 형의 FeRAM 소자를 BLT 박막을 적용하여 제작하였다. 솔-젤 (sol-gel) 용액을 이용하여 스핀코팅법으로 BLT 박막을 증착하고, 후속 열처리 공정을 RTP (rapid thermal process) 공정을 이용하여 수행하였다. 커패시터의 하부 전극 및 상부 전극은 각각 Pt/IrOx/lr 및 Pt을 적용하였다. 반응성 이온 에칭 (RIE: reactive ion etching) 공정을 이용하여 커패시터를 형성시킨 후, 32k-array (unit capacitor: $0.68\;{\mu}m$) 패턴에서 측정한 스위칭 분극 (dP=P*-P^) 값은 약 $16\;{\mu}C/cm^2$ 정도이고, 웨이퍼 내에서의 균일도도 2.8% 정도로 매우 우수한 특성을 보였다. 그러나 단위 셀들의 특성을 평가하기 위하여 bit-line의 전압을 측정한 결과, 약 10% 정도의 커패시터에서 불량이 발생하였다. 그리고 이러한 불량 젤들은 매우 불규칙적으로 분포함을 확인할 수 있었다. 이러한 불량 원인을 파악하기 위하여 양호한 젤과 불량이 발생한 셀에서의 BLT 박막의 미세구조를 분석하였다. 양호한 셀의 BLT 박막 입자들은 불량한 셀에 비하여 작고 비교적 균일한 크기를 갖고 있었다. 이에 비하여 불량한 셀에서의 BLT 박막에는 과대 성장한 입자들이 존재하고 이에 따라서 입자 크기가 매우 불균질한 것으로 확인되었다. 또 이러한 과대 성장한 입자들은 거의 모두 c-축 배향성을 나타내었다. 이상의 실험 결과들로부터, BLT 박막을 이용하여 제작한 FeRAM 소자에서 발생하는 불규칙한 셀 불량의 주된 원인은 c-축 배향성을 갖는 과대 성장한 입자의 생성임을 알 수 있었다. 즉 BLT 박막을 이용하여 FeRAM 소자를 제작하는 경우, 균일한 크기의 입자 및 c-축 배향성의 입자 억제가 매우 중요한 기술적 요소임을 알 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.1
• /
• pp.178-184
• /
• 2015

This paper proposes a hardware architecture for high performance Deblocking filter(DBF) in High Efficiency Video Coding for UHD(Ultra High Definition) videos. This proposed hardware architecture which has less processing time has a 4-stage pipelined architecture with two filters and parallel boundary strength module. Also, the proposed filter can be used in low-voltage design by using clock gating architecture in 4-stage pipeline. The segmented memory architecture solves the hazard issue that arises when single port SRAM is accessed. The proposed order of filtering shortens the delay time that arises when storing data into the single port SRAM at the pre-processing stage. The DBF hardware proposed in this paper was designed with Verilog HDL, and was implemented with 22k logic gates as a result of synthesis using TSMC 0.18um CMOS standard cell library. Furthermore, the dynamic frequency can process UHD 8k($7680{\times}4320$) samples@60fps using a frequency of 150MHz with an 8K resolution and maximum dynamic frequency is 285MHz. Result from analysis shows that the proposed DBF hardware architecture operation cycle for one process coding unit has improved by 32% over the previous one.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.8A
• /
• pp.739-752
• /
• 2002

Recently, to utilize the limited bandwidth effectively, the concept of partial response (PR) signaling has widely been adopted in both the high-speed data transmission and high-density digital recording/playback systems such as digital microwave, digital subscriber loops, hard disk drives, digital VCR's and digital versatile recordable disks and so on. This paper is concerned with adaptive equalization of partial response channels particularly for the magnetic recording channels. Specifically we study how the PR channel equalizers work for different choices of desired or reference signals used for adjusting the equalizer weights. In doing so, we consider three different configurations that are actually implemented in the commercial products mentioned above. First of all, we show how to compute the theoretical values of the optimum Wiener solutions derived by minimizing the mean-squared error (MSE) at the equalizer output. Noting that this equalizer MSE measure cannot be used to fairly compare the three configurations, we propose to use the data MSE that is computer just before the final detector for the underlying PR system. We also express the data MSE in terms of the channel impulse response values, source data power and additive noise power, thereby making it possible to compare the performance of the configurations under study. The results of extensive computer simulation indicate that our theoretical derivation is correct with high precision. Comparing the three configurations, it also turns out that one of the three configurations needs to be further improved in performance although it has an apparent advantage over the others in terms of memory size when implemented using RAM's for the decision feedback part.

• Journal of Korea Water Resources Association
• /
• v.43 no.8
• /
• pp.747-755
• /
• 2010

Distributed Models have relative weak points due to the amount of computer memory and calculation time required for calculating water flow using a numerical analysis based on kinematic wave theory when compared to the conceptual models used so far. Typically, the distributed models have been mainly applied to small basins. It was necessary to decrease the resolution of the grid to make it applicable for large scale watersheds, and because it would take up too much time to calculate using a higher resolution. That has been one of the more difficult factors in applying the model for actual work. In this paper, MPI (Message Passing Interface) technique was applied to solve the problem of calculation time as it is one of the demerits of the distributed model for performing physical and complicated numerical calculations for large scale watersheds. The comparison studies were performed a single domain and a divided small domain in Yongdam Dam watershed in case of typoon 'Ewiniar' at 2006. They were compared to analyze the application effects of parallelization technique. As a result, a maximum of 10 times the amount of calculation time was saved but keeping the level of quality for discharge by using parallelization code rather than a single processor.