• Title/Summary/Keyword: 완화와 적응

Search Result 190, Processing Time 0.028 seconds

Efficient Data Preprocessing Scheme for Audio Deep Learning in Solar-Powered IoT Edge Computing Environment (태양 에너지 수집형 IoT 엣지 컴퓨팅 환경에서 효율적인 오디오 딥러닝을 위한 데이터 전처리 기법)

  • Yeon-Tae Yoo;Chang-Han Lee;Seok-Mun Heo;Na-Kyung You;Ki-Hoon Kim;Chan-Seo Lee;Dong-Kun Noh
    • Annual Conference of KIPS
    • /
    • 2023.05a
    • /
    • pp.81-83
    • /
    • 2023
  • 태양 에너지 수집형 IoT 기기는 주기적으로 재충전되는 태양 에너지의 특성상, 에너지 소모를 최소화하기보다는 수집된 에너지를 최대한 유용하게 사용하는 것이 중요하다. 한편, 데이터 기밀성과 프라이버시, 응답속도, 비용 등의 이유로 클라우드가 아닌 데이터 소스 근처에서 머신러닝을 수행하는 엣지 AI에 대한 연구도 활발한데, 그 중 하나는 여러 IoT 장치들이 수집한 오디오 데이터를 활용하여, 다양한 AI 응용들을 IoT 엣지 컴퓨팅 환경에서 제공하는 것이다. 그러나, 이와 관련된 많은 연구에서, IoT 기기들은 에너지의 제약으로 인하여, 엣지 서버(IoT 서버)로의 센싱 데이터 전송만을 수행하고, 데이터 전처리를 포함한 모든 AI 과정은 엣지 서버에서 수행한다. 이 경우, 엣지 서버의 과부하 문제 뿐 아니라, 학습 및 추론에 불필요한 데이터까지도 서버에 그대로 전송되므로 네트워크 과부하 문제도 야기한다. 또한, 이를 해결하고자, 데이터 전처리 과정을 각 IoT 기기에 모두 맡긴다면, 기기의 에너지 부족으로 정전시간이 증가하는 또 다른 문제가 발생한다. 본 논문에서는 각 IoT 기기의 에너지 상태에 따라 데이터 전처리 여부를 결정함으로써, 기기들의 정전시간 증가 문제를 완화시키면서 서버 집중형 엣지 AI 환경의 문제들(엣지 서버 및 네트워크 과부하)을 완화시키고자 한다. 제안기법에서 IoT 장치는 기기가 기본적으로 동작하는 데 필요한 에너지 외의 여분의 에너지 양을 예측하고, 이 여분의 에너지가 있는 경우에만 이를 사용하여 기기에서 전처리 과정, 즉 수집 대상 소리 판별과 잡음 제거 과정을 거친 후 서버에 전송함으로써, IoT기기의 정전시간에 영향을 주지 않으면서, 에너지 적응적으로 데이터 전처리 위치(IoT기기 또는 엣지 서버)를 결정하여 수행한다.

Clinical Approach for Thyroid Radiofrequency Ablation (갑상선 고주파 절제술을 위한 임상진료)

  • Jung Suk Sim
    • Journal of the Korean Society of Radiology
    • /
    • v.84 no.5
    • /
    • pp.1017-1030
    • /
    • 2023
  • Radiofrequency ablation (RFA) is a non-surgical treatment for symptomatic, benign thyroid nodules. This treatment works by heating and destroying the nodule tissue, which results in reduction of its size and alleviation of the symptoms involved. RFA is indicated for nodules which are confirmed to be benign on two or more cytological or histological examinations, and which result in clinical symptoms requiring medical treatment. It is associated with good short-term outcomes on one-year follow-up; however, 20%-30% of the nodules regrow after more than three years. Therefore, on the basis of long-term follow-up, management of regrowth is key to patient care following RFA. Regrowth is more likely to occur in nodules that are large in size prior to RFA, and in those with high or increased vascularity. Recently, new techniques such as hydrodissection, artery-first ablation, and venous ablation have been introduced to inhibit regrowth. In addition, appropriate criteria for additional RFA should be applied to manage regrowth and prolong its therapeutic effects. RFA is essentially an alternative to surgery; therefore, the ultimate goal of this procedure is to avoid surgery permanently, rather than to achieve temporary effects.

Pre-processing Scheme for Indoor Precision Tracking Based on Beacon (비콘 기반 실내 정밀 트래킹을 위한 전처리 기법)

  • Hwang, Yu Min;Jung, Jun Hee;Shim, Issac;Kim, Tae Woo;Kim, Jin Young
    • Journal of Satellite, Information and Communications
    • /
    • v.11 no.4
    • /
    • pp.58-62
    • /
    • 2016
  • In this paper, we propose a pre-processing scheme for improving indoor positioning accuracy in impulsive noise channel environments. The impulsive noise can be generated by multi-path fading effects by complicated indoor structures or interference environments, which causes an increase in demodulation error probability. The proposed pre-processing scheme is performed before a triangulation method to calculate user's position, and providing reliable input data demodulated from a received signal to the triangulation method. Therefore, we studied and proposed an adaptive threshold function for mitigation of the impulsive noise based on wavelet denoising. Through results of computer simulations for the proposed scheme, we confirmed that Bit Error Rate and Signal-to-Noise Ratio performance is improved compared to conventional schemes.

A Spatially Adaptive Post-processing Filter to Remove Blocking Artifacts of H.264 Video Coding Standard (H.264 동영상 표준 부호화 방식의 블록화 현상 제거를 위한 적응적 후처리 기법)

  • Choi, Kwon-Yul;Hong, Min-Cheol
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.33 no.8C
    • /
    • pp.583-590
    • /
    • 2008
  • In this paper, we present a spatially adaptive post-processing algorithm for H.264 video coding standard to remove blocking artifacts. The loop filter of H.264 increases computational complexity of the encoder. Furthermore it doesn't clearly remove the blocking artifacts, resulting in over-blurring. For overcoming them, we combine the projection method with the Constraint Least Squares(CLS) method to restore the high quality image. To reflect the Human Visual System, we adopt the weight norm CLS method. Particularly pixel location-based local variance and laplacian operator are newly defined for the CLS method. In addition, the fact that correlation among adjoining pixels is high is utilized to constrain the solution space when the projection method is applied. Quantization Index(QP) of H.264 is also used to control the degree of smoothness. The simulation results show that the proposed post-processing filter works better than the loop filter of H.264 and converges more quickly than the CLS method.

UWB RADAR based Modified Adaptive CFAR Algorithm for improved safety of Personal Rapid Transit (무인 궤도 차량의 안전성 제고를 위한 UWB 레이더 기반 적응형 CFAR 알고리즘)

  • Hong, Seok-Gon;Kim, Baek-Hyun;Jeong, Rag-Gyo;Kwak, Kyung-Sup
    • The Journal of The Korea Institute of Intelligent Transport Systems
    • /
    • v.12 no.1
    • /
    • pp.28-42
    • /
    • 2013
  • Personal Rapid Transit(PRT) is a new unmanned transportation system using electricity. The purpose of the PRT is relieving the congestion of city traffic and connecting between inner city and airport, high-speed railroad. PRT requires to develop devices for the guarantee of safety and reliability. PRT as the mean of rail transportation must be equipped with control system for front rail sensing. Ultra Wide Band(UWB) radar system is suitable for PRT's detection because it has the advantage of low power consumption, low interference and high resolution. In this paper, an improved adaptive Constant False Alarm Rate(CFAR) algorithm is proposed and studied in various noise environments. The proposed algorithm improves performance in various noise environments compared to the Mean Level CFAR algorithms and other adaptive CFAR algorithms.

Modified Adaptive Random Testing through Iterative Partitioning (반복 분할 기반의 적응적 랜덤 테스팅 향상 기법)

  • Lee, Kwang-Kyu;Shin, Seung-Hun;Park, Seung-Kyu
    • Journal of the Institute of Electronics Engineers of Korea CI
    • /
    • v.45 no.5
    • /
    • pp.180-191
    • /
    • 2008
  • An Adaptive Random Testing (ART) is one of test case generation algorithms that are designed to detect common failure patterns within input domain. The ART algorithm shows better performance than that of pure Random Testing (RT). Distance-bases ART (D-ART) and Restriction Random Testing (RRT) are well known examples of ART algorithms which are reported to have good performances. But significant drawbacks are observed as quadratic runtime and non-uniform distribution of test case. They are mainly caused by a huge amount of distance computations to generate test case which are distance based method. ART through Iterative Partitioning (IP-ART) significantly reduces the amount of computation of D-ART and RRT with iterative partitioning of input domain. However, non-uniform distribution of test case still exists, which play a role of obstacle to develop a scalable algerian. In this paper we propose a new ART method which mitigates the drawback of IP-ART while achieving improved fault-detection capability. Simulation results show that the proposed one has about 9 percent of improved F-measures with respect to other algorithms.

Survey on Deep learning-based Content-adaptive Video Compression Techniques (딥러닝 기반 컨텐츠 적응적 영상 압축 기술 동향)

  • Han, Changwoo;Kim, Hongil;Kang, Hyun-ku;Kwon, Hyoungjin;Lim, Sung-Chang;Jung, Seung-Won
    • Journal of Broadcast Engineering
    • /
    • v.27 no.4
    • /
    • pp.527-537
    • /
    • 2022
  • As multimedia contents demand and supply increase, internet traffic around the world increases. Several standardization groups are striving to establish more efficient compression standards to mitigate the problem. In particular, research to introduce deep learning technology into compression standards is actively underway. Despite the fact that deep learning-based technologies show high performance, they suffer from the domain gap problem when test video sequences have different characteristics of training video sequences. To this end, several methods have been made to introduce content-adaptive deep video compression. In this paper, we will look into these methods by three aspects: codec information-aware methods, model selection methods, and information signaling methods.

Gram-Schmidt process based adaptive time-reversal processing (그람슈미트 과정 기반의 적응형 시역전 처리)

  • Donghyeon Kim;Gihoon Byun;J. S. Kim;Kee-Cheol Shin
    • The Journal of the Acoustical Society of Korea
    • /
    • v.43 no.2
    • /
    • pp.184-199
    • /
    • 2024
  • Residual crosstalk has been considered as a major drawback of conventional time-reversal processing in the case of simultaneous multiple focusing. In this paper, the Gram-Schmidt process is applied to time-reversal processing to mitigate crosstalk in ocean waveguides for multiple probe sources. Experimental data-based numerical simulations confirm that nulls can be placed at multiple locations, and it is shown that different signals can be simultaneously focused at different probe source locations, ensuring distortionless responses in terms of active time-reversal processing. This focusing property is also shown to be much less affected by a reduction in the number of receivers than the adaptive time-reversal mirror method. The proposed method is shown to be effective in eliminating crosstalk in passive multi-input multi-output communications using sea-going data.

Malignant Bowel Obstruction in Terminal Cancer Patients (말기암 환자의 악성 장 폐색)

  • Moon, Do-Ho;Choe, Wha-Sook
    • Journal of Hospice and Palliative Care
    • /
    • v.7 no.2
    • /
    • pp.214-220
    • /
    • 2004
  • Purpose: As for the malignant bowel obstruction of terminal cancer patient, a prognosis is relatively bad. Physicians consider palliative procedures or surgery for the quality of life, but sometimes it is hard to decide. After diagnosis of a malignant bowel obstruction in terminal cancer patients, we investigated the clinical characteristics, the prognostic factors and the survival of patients with palliative procedures or surgery. Methods: we retrospectively reviewed the medical records in 40 malignant bowel obstruction patients who had been diagnosed as terminal cancer from May in 2002 to May in 2004. Results: There were 21 males (53%) and 19 females (47%), and median age of patients was $64.1{\pm}1.58$ years. The most common cause of malignant bowel obstruction was colorectal cancer (18 patients, 45%), followed by stomach cancer (11, 28%), pancreatic cancer (4, 10%), others (7, 19%). Metastases were carcinomatosis peritonei (14 patients, 35%), liver (13, 33%). During a bowel obstruction, symptoms were vomiting (15 patients, 38%), abdominal pain (10, 25%), constipation (6, 15%), abdominal distension (5, 13%). Performance status (ECOG) was 2 score (16 patients, 40%), 3 score (20, 50%), 4 score (4, 10%). Palliative procedure group were 30 patients, the others 10. Median survival in palliative procedure group was 142 days, that of no palliation group 30. Median survival time of palliative procedure group from palliative procedures or surgery were significantly higher than that of no palliation group from diagnosis of malignant bowel obstruction. Prognostic factors of palliative procedure group were PS, site of obstruction and primary cancer. Median survival in PS 2, lower GI obstruction and colorectal cancer was higher than PS 3, upper GI obstruction and others, respectively. Conclusion: we recommend aggressively palliative procedures or surgery in malignant bowel obstruction patients diagnosed with terminal cancer if palliative procedures or surgery could be performed effectively.

  • PDF

A Node Mobility-based Adaptive Route Optimization Scheme for Hierarchical Mobile IPv6 Networks (노드 이동성을 고려한 계층적 이동 IPv6 네트워크에서의 적응적 경로 최적화 방안)

  • 황승희;이보경;황종선;한연희
    • Journal of KIISE:Information Networking
    • /
    • v.30 no.4
    • /
    • pp.474-483
    • /
    • 2003
  • The secret sharing is the basic concept of the threshold cryptosystem and has an important position in the modern cryptography. At 1995, Jarecki proposed the proactive secret sharing to be a solution of existing the mobile adversary and also proposed the share renewal scheme for (k, n) threshold scheme. For n participants in the protocol, his method needs O($n^2$) modular exponentiation per one participant. It is very high computational cost and is not fit for the scalable cryptosystem. In this paper, we propose the efficient share renewal scheme that need only O(n) modular exponentiation per participant. And we prove our scheme is secure if less that ${\frac}\frac{1}{2}n-1$ adversaries exist and they static adversary.