查詢結果分析
來源資料
相關文獻
- 類神經網路在多位元錯誤校正上之應用
- 鑰匙辨識之評估與比較
- Neural Network Procedures for Taguchi's Dynamic Problems
- 泛歐式數位行動電話之通道編碼技術
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
- 類神經網路(Neural Networks)的種類及其在影像處理上的應用
- C++Fuzzy類神經網路物件導向發展系統之建立
- 臺灣汽保費率之估計--對數線性費率模式與類神經網路之比較
頁籤選單縮合
題 名 | 類神經網路在多位元錯誤校正上之應用 |
---|---|
作 者 | 簡鼎立; 樓康寧; | 書刊名 | 技術學刊 |
卷 期 | 10:4 1995.12[民84.12] |
頁 次 | 頁437-444 |
分類號 | 312.2 |
關鍵詞 | 循環碼; 最小距離; 錯誤校正; 類神經網路; Cyclic code; Minimum distance; Error correcting neural network; |
語 文 | 中文(Chinese) |
中文摘要 | 一個 (n, k) 循環碼編碼後,具有2□個位元的字碼,資料位元長度k,同位元檢查長度n-k,最小距離d□及可錯誤校正位元數t之間存在著互動關係。傳統的錯誤較正使用n-k位元的信種計算器並透過表格找出發生錯誤位元的位置,以達到錯誤校正之目的,而其缺點除了串列處理速度慢以外,且當d□愈大時結構將較為複雜。本文嚐試著提出一個三層前向認知器類神經網路以達到錯誤校正之功能。經理論推導顯示此類神經網路除了訓練容易外並展現其平行處理及錯誤校正之功能,且在d□愈大時更顯示其結構比傳統方法簡單;而實際模擬湖試後亦得到在t位元錯誤內,其錯誤校正率為百分之百。 |
英文摘要 | After encoding a (n,k) cyclic code, a set of 2□ code word of length n is generated at the encoder output. There are mutual relation among the k information bits, n-k parity-check bits, minimum distance d□ and error correcting capability t. The conventional way of error correcting is accomplished by computing the syndrome of the received sequence and using the table to find the position of error bits. Except that it took long time to process the sequence, another disadvantage of this way is that the cyclic structure will become more complicated as the d□ become large. This paper tries to provide a three-layer forward perceptron neural network with the capability of error correcting. By this theory, this kind of neural network is not only capable of the functions of parallel process and error correcting but also easy to train. This network also shows that the cyclic structure is simpler than the conventional way as d□ becomes large. From the simulation results, it shows that its error correcting rate is definitely 100% within t error bits. |
本系統中英文摘要資訊取自各篇刊載內容。