선택한 단어 수는 입니다.
최소 단어 이상 선택하여야 합니다.
최소 단어 이상 선택하여야 합니다.
선택한 단어 수는 30입니다.
최대 10 단어까지만 선택 가능합니다.
최대 10 단어까지만 선택 가능합니다.
제안 방법
- Based on the simulation results, which show that a satisfactoe BER performance can be achieved when the training size is sufficient, this paper proposes a new decoding structure employing the block size that is set to a fraction of the training size. In the proposed MAP decoder structure, the memory area has been reduced by 5.
- The proposed scheme requires memory for storing the forward state metric and the branch metrics. However, the memory requirement in the proposed scheme is reduced whencompared to the previous scheme due to the employment of the reduced block size.
- From these observations, training size of five times the constraint size is considered to be sufficient. The training process determines the initial backward state metrics. When the block size is large enough, the backward state metrics become reliable as the decoding process continues regardless of the initial error.
- To avoid the BER performance degradation, a training process is introduced. The training process determines the reliable initial backward state metrics by calculating the backward state metrics in the next block without storing their values in the memory. For successive decoding, a parallel invocation of backward state metric calculation process and the training process is required using two processors [5], [6].
- A small block size implies less memory requirement. This paper proposes a new scheme where the block size is only a fraction of the training size. Figure 7 shows the proposed block-wise MAP decoding scheme where the block size is set to 1/μ of the training size.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.