论文部分内容阅读
DPCM图象传输系统中,对量化后的预测误差所采取的代码分配方案影响到系统对信道误码的敏感程度。本文利用一个实际的4bit量化器,对三种代码分配方案进行了比较。比较包括:误码引起的平均误差幅值、大误差出现的概率和出现概率超过5％的最大误差幅值。比较结果表明,当每码字中出现1～2个比特差错时,符号—幅值码对信道误码的敏感性最低,而当每码字中出现3～4个比特差错时,自然二进制码性能最好。所得结果可供结合实际信道类型优选代码分配方案。
In the DPCM image transmission system, the code allocation scheme adopted for the quantized prediction error affects the sensitivity of the system to the channel error. In this paper, an actual 4bit quantizer is used to compare the three code allocation schemes. Comparisons include: error caused by the average error amplitude, the probability of occurrence of large errors and the probability of occurrence of more than 5% of the maximum error amplitude. The comparison results show that when 1 ~ 2 bit errors occur in each codeword, the symbol-amplitude code has the lowest sensitivity to channel error. When 3 ~ 4 bit errors occur in each codeword, the natural binary code The best performance. The results are for optimal code allocation in conjunction with the actual channel type.