Zero-Substitution and nB/NB Translation in Data Communication Principles

DATA COMMUNICATION PRINCIPLES : Zero-substitution and nB/NB Translation

Zero-substitution and nB/NB Translation There are ways to inset signal transitions for clocking information in NRZ and AMI schemes without requiring the bandwidth needs of Manchester coding. These mechanisms increase the complexity of coding device. One of such mechanisms is called zero-substitution in which a long string of 0 s is coded such that there is a certain minimum number of pulses in the string. For example, in B8ZS (Bipolar with 8-zero substitution) eight zeros are coded together by inserting pulses in violation of the Bipolar AMI codes. The violation of AMI codes pertains to inserting pulses with the same polarity as the previous pulses occurrence. The violations are inserted in such as way as to avoid dc accumulation. Thus, if the polarity of the last pulse before the string of zeros was positive, then the eight zeros are coded as 0 0 0 + - 0 - +. Therefore, the two violations of the Bipolar AMI code help decode the correct string of zeros. If the last pulse before the occurrence of eight 0 s is negative, the polarity of the pulses representing the eight zeros is reversed making it 0 0 0 - + 0 + -. This is mostly used in USA for some long distance communications networks. Other coding schemes using zero- substitution are HDB3 (high density binary coding with 3 zeros), B3ZS and B6ZS. Figure 3- 11 shows plots of B8ZS and HDB3 encoding schemes.

The rules for HDB3 are given in Table 3-2 and for B8ZS in the the rules of encoding can be described as follows: If the polarity of the last pulse before the string of zeros is positive, then the eight zeros are coded as 0 0 0 + 0 - +. Therefore, the two violations of the Bipolar AMI code help decode the correct string of zeros. If the last pulse before the occurrence of eight 0 s is negative, the polarity of the pulses representing the eight zeros is reversed making it 0 0 0 - + 0 + -.

In another mechanism of coding with clocking information, a table is used, that has two columns. The first column consists of patterns of n bits. The second column consists of patterns of N bits. The value of N is chosen to be greater than n, so that the patterns of N bits have a specified number of both data bits. For example, in 4B/5B translation, all the patterns of the second column have at least two 1 s to guarantee clock information in a string of every 5 bits. The coder looks at the transmitting data 4 bits at a time. Then, using the table, it transmits the 5 bits pattern corresponding the 4 bits to be transmitted. In this way, 20% of the transmitted data constitutes the overhead for inserting clocking information.

REFERENCES Ahmad A. - Data Communication Principles. For Fixed and Wireless Networks Cornelius T. Leondes - Database and Data Communication Network Systems, Three-Volume Set_..-Academic Press