• 제목/요약/키워드: Decoder complexity

검색결과 350건 처리시간 0.024초

네트웍 반향제거기의 성능 향상 (Performance Improvement of the Network Echo Canceller)

  • 유재하
    • 음성과학
    • /
    • 제11권4호
    • /
    • pp.89-97
    • /
    • 2004
  • In this paper, an improved network echo canceller is proposed. The proposed echo canceller is based on the LTJ(lattice transversal joint) adaptive filter which uses informations from the speech decoder. In the proposed implementation method of the network echo canceller, the filer coefficients of the transversal filter part in the LTJ adaptive filter is updated every other sample instead of every sample. So its complexity can be lower than that of the transversal filter. And the echo cancellation rate can be improved by residual echo cancellation using the lattice predictor whose order is less than 10. Computational complexity of the proposed echo canceller is lower than that of the transversal filter but the convergence speed is faster than that of the transversal filter. The performance improvement of the proposed echo canceller was verified by the experiments using the real speech signal and speech coder.

  • PDF

A Study on Complexity Reduction for Space-Time Block Codes Decoder in MIMO OFDM systems

  • Pham, Van-Su;Le, Minh-Tuan;Mai, Linh;Yoon, Gi-Wan
    • 한국정보통신학회:학술대회논문집
    • /
    • 한국해양정보통신학회 2005년도 춘계종합학술대회
    • /
    • pp.117-121
    • /
    • 2005
  • In this work, we first present our study on the decoding schemes for space-time block code as well as our comments on their complexity. Then, we propose a new modified complex sphere decoding scheme, which has lower computational load than that of conventional complex sphere decoders. In the proposed scheme, the boundary for searching is defined by the intersection of an approximate polygon of searching circle and disk of lattice constellation. Therefore, the proposed scheme can reduce the searching boundary and it can avoid missing searching points as well. The performance of the proposed scheme, which is verified by computer simulations, consolidates our scheme.

  • PDF

Complexity Reduction for Space-Time Block Codes Decoder in MIMO OFDM Systems

  • Pham Van-Su;Le Minh-Tuan;Mai Linh;Yoon Giwan
    • Journal of information and communication convergence engineering
    • /
    • 제3권2호
    • /
    • pp.72-75
    • /
    • 2005
  • In this work, we first present our study on the decoding schemes for space-time block code as well as our comments on their complexity. Then, we propose a new modified complex sphere decoding scheme, which has lower computational load than that of conventional complex sphere decoders. In the proposed scheme, the boundary for searching is aided with the intersection of approximate polygon of searching circle and disk of constellation. Therefore, the proposed scheme can reduce the searching boundary while can avoid missing searching points. The performance of the proposed scheme is verified by computer simulation that consolidates our scheme.

A Comparative Study of List Sphere Decoders for MIMO Systems

  • Pham, Van-Su;Yoon, Giwan
    • 한국정보통신학회:학술대회논문집
    • /
    • 한국해양정보통신학회 2009년도 춘계학술대회
    • /
    • pp.143-146
    • /
    • 2009
  • In this paper, we investigated the list sphere decoders (LSD) for multiple-input multiple-output (MIMO) systems. We showed that the ordering procedures play an important role in LSD in order to achieve the low complexity without degrading the bit-error-rate (BER) performance. Then, we proposed a novel ordering algorithm for the LSD which uses a look-up table and simply comparative operations. Comparative results in terms of BER performance and computational complexity are provided through computer simulations.

  • PDF

Zero forcing based sphere decoder for generalized spatial modulation systems

  • Jafarpoor, Sara;Fouladian, Majid;Neinavaie, Mohammad
    • ETRI Journal
    • /
    • 제41권2호
    • /
    • pp.145-159
    • /
    • 2019
  • To reduce the number of radio frequency (RF) chains in multiple input multiple output (MIMO) systems, generalized spatial modulation (GSM) techniques have been proposed in the literature. In this paper, we propose a zero-forcing (ZF)-based detector, which performs an initial pruning of the search tree that will be considered as the initial condition in a sphere decoding (SD) algorithm. The proposed method significantly reduces the computational complexity of GSM systems while achieving a near maximum likelihood (ML) performance. We analyze the performance of the proposed method and provide an analytic performance difference between the proposed method and the ML detector. Simulation results show that the performance of the proposed method is very close to that of the ML detector, while achieving a significant computational complexity reduction in comparison with the conventional SD method, in terms of the number of visited nodes. We also present some simulations to assess the accuracy of our theoretical results.

High-Speed Transformer for Panoptic Segmentation

  • Baek, Jong-Hyeon;Kim, Dae-Hyun;Lee, Hee-Kyung;Choo, Hyon-Gon;Koh, Yeong Jun
    • 방송공학회논문지
    • /
    • 제27권7호
    • /
    • pp.1011-1020
    • /
    • 2022
  • Recent high-performance panoptic segmentation models are based on transformer architectures. However, transformer-based panoptic segmentation methods are basically slower than convolution-based methods, since the attention mechanism in the transformer requires quadratic complexity w.r.t. image resolution. Also, sine and cosine computation for positional embedding in the transformer also yields a bottleneck for computation time. To address these problems, we adopt three modules to speed up the inference runtime of the transformer-based panoptic segmentation. First, we perform channel-level reduction using depth-wise separable convolution for inputs of the transformer decoder. Second, we replace sine and cosine-based positional encoding with convolution operations, called conv-embedding. We also apply a separable self-attention to the transformer encoder to lower quadratic complexity to linear one for numbers of image pixels. As result, the proposed model achieves 44% faster frame per second than baseline on ADE20K panoptic validation dataset, when we use all three modules.

Turbo coded BICM-ID의 복잡도 개선 기법 (A Low-Complexity Turbo coded BICM-ID System)

  • 강동훈;이용욱;오왕록
    • 전자공학회논문지
    • /
    • 제50권8호
    • /
    • pp.21-27
    • /
    • 2013
  • 본 논문에서는 Turbo coded BICM-ID (bit-interleaved coded modulation with iterative decoding)의 구현 복잡도를 개선하기 위한 기법을 제안한다. 터보 부호는 Shannon 한계에 근접하는 우수한 성능을 나타내는 오류정정부호 (error correction code)이며 터보 부호와 고차원 변조 (high order modulation)를 결합하여 전송효율을 개선할 수 있다. BICM (bit-interleaved coded modulation)은 채널 부호와 고차원 변조 사이에 이진 인터리버를 사용하는 기법이며 복호기와 복조기 간에 정보를 주고받으며 반복 복호 (iterative decoding)를 수행하는 BICM-ID를 이용하여 성능을 개선 할 수 있다. BICM-ID는 BICM과 비교하였을 때 반복 복호로 인하여 비트오율 (bit error rate) 성능이 개선되지만 구현 복잡도가 크게 증가하는 단점이 있다. 본 논문에서는 Turbo coded BICM-ID에서 구현 복잡도를 개선하기 위한 기법을 제안한다. 제안하는 기법은 기존에 제안된 Turbo coded BICM-ID 기법과 유사한 비트오율 (bit error rate) 성능을 나타내면서 구현 복잡도를 크게 감소시킬 수 있는 장점이 있다.

단일 엔코더 및 디코더를 이용하는 다시점 비디오 시스템 (Multi-View Video System using Single Encoder and Decoder)

  • 김수학;김윤;김만배
    • 방송공학회논문지
    • /
    • 제11권1호
    • /
    • pp.116-129
    • /
    • 2006
  • 인터넷을 통한 데이터 전송 기술의 발전은 다양한 실감 콘텐츠를 확산시키는 역할을 하고 있고, 이러한 실감 콘텐츠의 하나로 다시점 비디오가 연구되고 있다. 다시점 비디오는 여러 대의 카메라로부터 획득한 영상을 사용자에게 전달해 원하는 시점에서 영상을 시청할 수 있도록 하는 차세대 영상이다. 다시점 영상을 공급하기 위해서는 일반적으로 카메라 개수와 같은 개수의 엔코더와 디코더가 필요하며, 또한 상대적으로 큰 전송 대역폭을 필요로 한다. 본 논문에서는 상기 문제점을 해결하기 위해서 단일 엔코더와 디코더만을 이용하는 다시점 비디오 시스템을 제안한다. 부호화에서는 다시점 YUV 비디오를 GOP 단위로 믹싱하여 l 개의 YUV 스트립을 만든 후에, 단일 H.264/ AVC 엔코더로 부호화하여 압축 비트열을 생성한다. 또한 복호화도 단일 디코더 및 이를 제어하는 스케줄러로 구성된다. 스케줄러는 디코더 이용률과 프레임 스킵을 활용하여 각 시점 영상들이 동등한 복호화 프레임 개수를 가지게 한다. 또한 프레임 스킵에서는 비용 함수를 이용하여 H.264/AVC 프로화일에 따른 효율적인 프레임 선택 방법을 제안한다. 다양한 메인 및 베이스라인 프로화일 다시점 영상들을 활용하여 제안 방법의 성능을 검증하였다. 실험 결과는 각 시점 영상이 동등한 디코더 이용률을 얻는 것을 보여준다. 또한 rate-distortion 커브를 이용하여 제안한 방식과 동시 방식의 성능을 검증하였다.

웨이블렛 변환을 이용한 오디오 코딩 (Audio Signal Coding Using Wavelet Transform)

  • 배석모;김도형;정재호
    • 한국음향학회지
    • /
    • 제16권4호
    • /
    • pp.64-70
    • /
    • 1997
  • 본 논문은 MPEG의 서브밴드 필터링을 웨이블렛 변환으로 대체하였을 때 얼마만큼의 계산량이 감소하는 가를 연구하였다. MPEG 오디오에서는 16비트 PCM 오디오 데이타를 입력으로 받아 서브밴드 필터링후 서브밴드 샘플을 양자화하여 전송하는 것을 기본으로 한다. MPEG의 서브밴드 필터링의 경우 32 대역의 등간격으로 분할한다. 이 경우 32개의 필터가 필요하고 각 필터는 512의 길이를 갖는 필터를 사용한다. 본 연구에서는 분할 대역을 6개로 하고 웨이블렛 필터중 가장 짧은 Haar 필터를 사용하였다. 제안된 시스템은 256kbps 이상의 전송율에서는 MPEG 오디오와 비슷한 수준의 CD 음질을 유지하였으며, 계산량 비교결과 부호화기는 약 40%, 복호화기는 약 70%의 감소를 보였다.

  • PDF

Movement Detection Using Keyframes in Video Surveillance System

  • Kim, Kyutae;Jia, Qiong;Dong, Tianyu;Jang, Euee S.
    • 한국방송∙미디어공학회:학술대회논문집
    • /
    • 한국방송∙미디어공학회 2022년도 하계학술대회
    • /
    • pp.1249-1252
    • /
    • 2022
  • In this paper, we propose a conceptual framework that identifies video frames in motion containing the movement of people and vehicles in traffic videos. The automatic selection of video frames in motion is an important topic in security and surveillance video because the number of videos to be monitored simultaneously is simply too large due to limited human resources. The conventional method to identify the areas in motion is to compute the differences over consecutive video frames, which has been costly because of its high computational complexity. In this paper, we reduced the overall complexity by examining only the keyframes (or I-frames). The basic assumption is that the time period between I-frames is rather shorter (e.g., 1/10 ~ 3 secs) than the usual length of objects in motion in video (i.e., pedestrian walking, automobile passing, etc.). The proposed method estimates the possibility of videos containing motion between I-frames by evaluating the difference of consecutive I-frames with the long-time statistics of the previously decoded I-frames of the same video. The experimental results showed that the proposed method showed more than 80% accuracy in short surveillance videos obtained from different locations while keeping the computational complexity as low as 20 % compared to the HM decoder.

  • PDF