학술논문
无线传感器网络中比特错误率估计方法 / Differentia-based bit error rate estimation method in wireless sensor network
Document Type
Academic Journal
Author
Source
激光杂志 / Laser Journal. (10):102-107
Subject
Language
Chinese
ISSN
0253-2743
Abstract
由于无线传感器网络中干扰检测的性能依赖于比特错误率(Bit Error Rate, BER)估计的准确性，而现有的比特错误率估计方法或者难以实现，或者准确性差。针对这个问题，本文提出了采用差异化思想来提高比特错误率估计准确度的方法DEE(Differentiated Error Estimation)。其主要是发送方在数据包中插入具有不同估错能力的多级估错位，并随机均匀地分布所有估错位。接收方借助BER与奇偶校验失败概率的理论关系来估计BER。同时，DEE利用BER非均匀分布特征来优化各级估错位的能力，提高出现概率高的BER的估计准确度，以降低平均估计误差。实验结果表明，与现有方法EEC相比，DEE可将估计误差平均减少约44%。当估错冗余较低时，DEE可将估计误差减少约68%。
The performance of interference detection in wireless sensor network depends on the performance of bit error rate (BER) estimation, however, the existed BER estimation methods are either too complicate to imple-ment or with low precision. To solve the problem, differentia error estimation (DEE) method is proposed to enhance the precision of BER. The main idea is to insert multi-level error estimation bits that with different error estimation ability, which are random and uniformly distributed, into the sender’s packets. And the receivers estimate the BER by using the relation between the BER and parity check. Meanwhile, DEE optimizes the ability of error estimation bit of each level by making use of BER’s feature of non-uniform distribution, to enhance the estimate precision of BER with higher probability and lower the average estimation error. The experiments shows that, compared with the error estimating coding (EEC) method, the average estimation error decreases 44%, and the estimation error decreas-es as much as 68%when the redundancy is decreased.
The performance of interference detection in wireless sensor network depends on the performance of bit error rate (BER) estimation, however, the existed BER estimation methods are either too complicate to imple-ment or with low precision. To solve the problem, differentia error estimation (DEE) method is proposed to enhance the precision of BER. The main idea is to insert multi-level error estimation bits that with different error estimation ability, which are random and uniformly distributed, into the sender’s packets. And the receivers estimate the BER by using the relation between the BER and parity check. Meanwhile, DEE optimizes the ability of error estimation bit of each level by making use of BER’s feature of non-uniform distribution, to enhance the estimate precision of BER with higher probability and lower the average estimation error. The experiments shows that, compared with the error estimating coding (EEC) method, the average estimation error decreases 44%, and the estimation error decreas-es as much as 68%when the redundancy is decreased.