• 전자공학회논문지
• /
• 제50권4호
• /
• pp.52-58
• /
• 2013

캐시 메모리의 성능에 큰 영향을 미치는 요소 중 하나인 캐시 교체 알고리즘 중에서 가장 최적의 성능을 가지는 알고리즘은 LRU알고리즘이다. LRU알고리즘은 데이터의 temporal locality특성이 강한 프로그램에서 좋은 성능을 보여주지만, 그렇지 않은 프로그램에서는 많은 캐시 미스를 발생시킨다. 본 논문에서는 LRU알고리즘의 이러한 단점을 개선하기 위한 새로운 카운터 기반 교체 알고리즘인 DCR(Dynamic Counter based Replacement) 알고리즘을 제안한다. 본 논문에서는 캐시에 저장된 이후로 교체되기 전까지 다시 사용되지 않는 데이터인 zero reuse line의 발생 추이를 관찰함으로써 프로그램의 temporal locality 특성이 시간에 따라 동적으로 변화함을 보였다. 그리고 이에 착안하여 제안하는 DCR 알고리즘은 주기적으로 zero reuse line의 수를 카운트하여 프로그램의 temporal locality 변화에 대응할 수 있도록 하였다. DCR 알고리즘은 기존의 LRU알고리즘과 비교하여 최대 2.7%, 평균 0.47%의 미스율 감소를 보였다.

• 한국위성정보통신학회논문지
• /
• 제8권4호
• /
• pp.1-5
• /
• 2013

Cognitive Radio는 유휴 스펙트럼을 찾아 환경에 맞는 통신방식과 주파수 대역폭을 능동적으로 판단해 재활용하는 지능적인 간섭회피로 방식으로 스펙트럼을 공유하여 전파자원효율을 극대화 하는 기술이다. 기존의 협력 스펙트럼 센싱은 협력하는 사용자들의 채널 상태만을 고려하여 임계치를 정하고 있으며, 모두 같은 고정된 오경보 확률을 적용하여 센싱하게 된다. 오경보 확률은 오류 검출과 관계가 있으며, 고정된 오경보 확률은 검출확률의 감소의 문제를 가진다. 따라서 본 논문에서는 CR사용자의 채널상태를 고려할 뿐만 아니라 채널 상태에 따른 다른 오경보 확률을 차등 적용하여 협력 스펙트럼 센싱을 하는 시스템 모델을 제안하고 분석한다. 시뮬레이션 결과를 통해 검출확률 향상을 보인다.

• Journal of Communications and Networks
• /
• 제15권3호
• /
• pp.283-291
• /
• 2013

In this paper, we propose new multiuser detectors (MUDs) based on compressed sensing approaches for the large-scale multiple antenna systems equipped with dozens of low-power antennas. We consider the scenarios where the number of receiver antennas is smaller than the total number of users, but the number of active users is relatively small. This prior information motivates sparsity-embracing MUDs such as sparsity-embracing linear/nonlinear MUDs where the detection of active users and their symbol detection are employed. In addition, sparsity-embracing MUDs with maximum a posteriori probability criterion (MAP-MUDs) are presented. They jointly detect active users and their symbols by exploiting the probability of user activity, and it can be solved efficiently by introducing convex relaxing senses. Furthermore, it is shown that sparsity-embracing MUDs exploiting common users' activity across multiple symbols, i.e., frame-by-frame, can be considered to improve performance. Also, in multiple multiple-input and multiple-output networks with aggressive frequency reuse, we propose the interference cancellation strategy for the proposed sparsity-embracing MUDs. That first cancels out the interference induced by adjacent networks and then recovers the desired users' information by exploiting the low user activity. In simulation studies for binary phase shift keying modulation, numerical evidences establish the effectiveness of our proposed MUDs exploiting low user activity, as compared with the conventional MUD.

• 한국환경과학회지
• /
• 제29권11호
• /
• pp.1109-1123
• /
• 2020

In this study, the amount of harmful fungi and bacteria contained in the drainage and culture medium from the paprika hydroponic facility is identified. In addition, by proposing the necessity of effective purification of discharged drainage, this study attempted to confirm the possibility of drainage reuse. Finally, this study provides basic data on the basis for calculating the need for purification facilities in the future, as well as improvements in horticulture facility for sustainable agriculture. As a result of the analysis, a total of 12 types of fungi were detected in paprika medium and 10 types of fungi were detected in the drainage, and their densities were 130 and 68, respectively. Among the fungi detected in the media and drainage of the paprika hydroponic facility, the fungi with the highest detection frequency are Fusarium, Phytophthora, and Pythium. In the case of bacteria, a total of 2 types of bacteria were detected in the paprika facilities, and the density was 28 and 23, respectively. Therefore, in order to reuse the drainage and settle the circulating hydroponic cultivation system, a water treatment process capable of appropriate treatment is required.

• 대한임베디드공학회논문지
• /
• 제17권4호
• /
• pp.217-228
• /
• 2022

Currently, there are increasing demands for applying deep neural networks (DNNs) in the embedded domain such as classification and object detection. The DNN processing in embedded domain often requires custom hardware such as NPU for acceleration due to the constraints in power, performance, and area. Processing DNN models requires a large amount of data, and its seamless transfer to NPU is crucial for performance. In this paper, we developed a cycle-accurate NPU simulator to evaluate diverse NPU microarchitectures. In addition, we propose a novel technique for reducing the number of memory accesses when processing convolutional layers in convolutional neural networks (CNNs) on the NPU. The main idea is to reuse data with memory interleaving, which recycles the overlapping data between previous and current input windows. Data memory interleaving makes it possible to quickly read consecutive data in unaligned locations. We implemented the proposed technique to the cycle-accurate NPU simulator and measured the performance with LeNet-5, VGGNet-16, and ResNet-50. The experiment shows up to 2.08x speedup in processing one convolutional layer, compared to the baseline.

• 국제학술발표논문집
• /
• The 8th International Conference on Construction Engineering and Project Management
• /
• pp.387-396
• /
• 2020

Wearable biosensors have the potential to non-invasively and continuously monitor seniors' stress in their daily interaction with the urban environment, thereby enabling to address the stress and ultimately advance their outdoor mobility. However, current wearable biosensor-based stress detection methods have several drawbacks in field application due to their dependence on batch-learning algorithms. First, these methods train a single classifier, which might not account for multiple subjects' different physiological reactivity to stress. Second, they require a great deal of computational power to store and reuse all previous data for updating the signle classifier. To address this issue, we tested the feasibility of online multi-task learning (OMTL) algorithms to identify multiple seniors' stress from electrodermal activity (EDA) collected by a wristband-type biosensor in a daily trip setting. As a result, OMTL algorithms showed the higher test accuracy (75.7%, 76.2%, and 71.2%) than a batch-learning algorithm (64.8%). This finding demonstrates that the OMTL algorithms can strengthen the field applicability of the wearable biosensor-based stress detection, thereby contributing to better understanding the seniors' stress in the urban environment and ultimately advancing their mobility.

• 대한임베디드공학회논문지
• /
• 제19권1호
• /
• pp.19-25
• /
• 2024

This paper proposes an artificial intelligence-based personal information area object detection optimization method in an embedded system to de-identify personal information in video data. As an object detection optimization method, first, in order to increase the detection rate for personal information areas when detecting objects, a gyro sensor is used to collect the shooting angle of the image data when acquiring the image, and the image data is converted into a horizontal image through the collected shooting angle. Based on this, each learning model was created according to changes in the size of the image resolution of the learning data and changes in the learning method of the learning engine, and the effectiveness of the optimal learning model was selected and evaluated through an experimental method. As a de-identification method, a shuffling-based masking method was used, and double-key-based encryption of the masking information was used to prevent restoration by others. In order to reuse the original image, the original image could be restored through a security key. Through this, we were able to secure security for high personal information areas and improve usability through original image restoration. The research results of this paper are expected to contribute to industrial use of data without personal information leakage and to reducing the cost of personal information protection in industrial fields using video through de-identification of personal information areas included in video data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4331-4346
• /
• 2015

The femtocell network, which is designed for low power transmission and consists of consumer installed small base stations, coexists with macrocells to exploit spatial reuse gain. For its realization, cross-tier interference mitigation is an important issue. To solve this problem, we propose a joint access and power control scheme that requires limited information exchange between the femto and macro networks. Our objective is to maximize the network throughput while satisfying each user's quality of service (QoS) requirement. To accomplish this, we first introduce two distributed interference detection schemes, i.e., the femto base station and macro user equipment based schemes. Then, the proposed scheme dynamically adjusts the transmission power and makes a decision on the access mode of each femto base station. Through extensive simulations, we show that the proposed scheme outperforms earlier works in terms of the throughput and outage probability.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2008년도 제20회 한글 및 한국어 정보처리 학술대회
• /
• pp.63-68
• /
• 2008

인터넷의 발달로 지식의 재사용이 폭발적으로 증가하였다. 이는 지식의 확산이라는 측면에서는 바람직하지만 지식의 도용이라는 문제점을 안고 있다. 따라서 문서의 전부나 일부분을 재사용한 것인지를 판단하고자하는 요구가 증가하고 있다. 본 논문에서는 정보검색 기술을 이용하여 문서에서 부분 문서 재사용 및 표절을 탐색하는 방법을 제안한다. 본 논문에서는 대용량 문서의 고속 탐색을 위해서 원본 문서와 대상 문서를 색인하여 검색에 이용한다. 또한 한글의 언어적 특성을 맞게 어순 변경 비교, 기능어 생략 비교, 갭(gap) 비교 등의 다양한 처리 조건을 제공하여 문서 재사용을 탐색할 수 있다. 실험을 통해서 기존의 시스템보다 정확하게 고속으로 문서 재사용 탐색이 가능함을 보였다. 특히 비교 문서가 증가하더라도 비교 시간이 급격하게 증가하지 않으며, 정보검색 기법을 사용하는 경우 취약하다고 알려져 있는 부분 문서 재사용 탐색에도 견고하며, 처리 조건에 따라 유연하게 문서 재사용 탐색이 가능하다.

• ETRI Journal
• /
• 제38권1호
• /
• pp.30-40
• /
• 2016

Co-channel interference between macro-femtocell networks is an unresolved problem, due to the frequency reuse phenomenon. To mitigate such interference, a secondary femtocell must acquire channel-state knowledge about a co-channel macrocell user and accordingly condition the maximum transmit power of femtocell user. This paper proposes a pilot-based spectrum sensing (PSS) algorithm for overlaid femtocell networks to sense the presence of a macrocell user over a channel of interest. The PSS algorithm senses the pilot tones in the received signal through the power level and the correlation metric comparisons between the received signal and the local reference pilots. On ensuring the existence of a co-channel macrocell user, the maximum transmit power of the corresponding femtocell user is optimized so as to avoid interference. Time and frequency offsets are carefully handled in our proposal. Simulation results show that the PSS algorithm outperforms existing sensing techniques, even at poor received signal quality. It requires less sensing time and provides better detection probability over existing techniques.