通信专业英4.pptVIP

通信专业英语 Unit Four Digital Communications Part One Digitizing a signal refers to sampling a signal at sufficient time intervals and coding these samples into one of N possible values. A coded signal should be decoded in order to reproduce the original signal. Wouldn’t the reproduced signal sound different from the original, unsampled one? It seems that sampling a signal chops it into pieces. Actually, a sampling signal is chopped into pieces. However, if a sufficient number of samples per second are taken, the decoded signal will reproduce an analog signal exactly like the original signal. Each sample can be assigned one of N possible values, dependent upon the magnitude of the sampled signal. For instance, a sample can take on one of 64 different values. Each of these values is assigned a binary number. The number 64 is so that each sample is coded into six bits of information. Thus, one sample can have one of 64 binary coded values ranging from 000000 to 111111. Then, how many samples per seconds are taken? For regular voice communications, such as telephone transmissions, 6600 samples per second are considered sufficient for adequate fidelity. Then with 6600 samples per second and 6 bits per sample, the data is transmitted at a rate of 39,600 bits per second. This is a quite low transmission rate. The circuitry for processing these signals is quite simple by today’s standards. All digital-to-analog and analog-to-digital converters for this kind of system use standard integrated circuits. This kind of signal processing is considered better than the old way of processing analog signals. Although encoding and decoding signals seems to complicate rather than simplify processing, we gain more flexibility in processing encoded signal. As each bit only takes on a value of zero or one, we can shape pulses any way we wish. Thus, we can use square pulses, sinusoids of two different frequencies, or any other variety we choose. There are