DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Chung, Sae-Young | - |
dc.contributor.advisor | 정세영 | - |
dc.contributor.author | Ahn, Jungsik | - |
dc.contributor.author | 안정식 | - |
dc.date.accessioned | 2017-03-29T02:38:55Z | - |
dc.date.available | 2017-03-29T02:38:55Z | - |
dc.date.issued | 2014 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=657480&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/221798 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학과, 2014.2 ,[iv, 22 p. :] | - |
dc.description.abstract | This thesis is separated into two parts. In the first part, we consider a lossy joint source-channel coding system, where the encoder chooses an action sequence that affects the formation of the channel states, and then creates the channel input sequence based on the state sequence. The decoder chooses an action that affects the nature of the side information, and then reconstructs the source with a side information. This configuration combines the channel model of pure channel coding with action-dependent states and the source model of pure source coding with a side information vender at the decoder. Our setting extends the previously considered communication systems where the encoder has access to the channel state information (CSI) and the decoder has access to the side information that is correlated to the source to a more general system where the encoder and decoder generate action sequences that affect the nature of the state and side information. We prove a separation theorem for this communication system. In the second part, we develop an error correction code for the flash memory using a 2-stage coding system. The flash memory is a non-volatile memory and is widely used in many mobile devices since it has good characteristics such as portability, good power consumption rate, etc. But the relatively high cost is one of its weaknesses. For solving this problem, multiple levels per cell are used. But this method leads to increased raw error rate and it requires more powerful coding. We use the fact that the flash memory read operation is easy and less noisy, we propose the following 2-stage coding system. After writing the flash memory once and reading its output, the encoder performs another writing. We design the 2-stage coding scheme that matches with the characteristics of the flash memory and has simple encoding and decoding. We show using simulations the performance of the proposed 2-stage coding scheme is better than that of the one-stage scheme. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | action | - |
dc.subject | joint source-channel coding | - |
dc.subject | separation theorem | - |
dc.subject | 2-stage coding | - |
dc.subject | flash memory | - |
dc.subject | 액션 | - |
dc.subject | 조인트 소스 채널 코딩 | - |
dc.subject | 세퍼레이션 정리 | - |
dc.subject | 두 단계 코딩 | - |
dc.subject | 플래시 메모리 | - |
dc.title | On joint source-channel coding with action-dependent side information and development of the error correction code using a 2-stage coding system | - |
dc.title.alternative | 액션-의존 보조 정보 시스템의 조인트 소스-채널 코딩과 2-단계 코딩 시스템을 이용한 오류 정정 부호 개발 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학과, | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.