Digital is Analog

The basic idea of digital communication is to eliminate the unnecessary noises that are the inherent components of so called analog communications. The old AM receivers sound little awkward when an electric switch near to it is turned OFF/ON. The basic concept of digital communication is to represent anything in this world with two digits ‘1’ and ‘0’. But if we look closely at how these ‘1’s and ‘0’s are transmitted over the media, it will be very interesting. Before understanding how to maintain the integrity of signals, we got to know about what a signal is. After reading this article you will be able to understand that the digital signals are nothing but ensemble of analog signals. When you can deal any digital signal with analog approach, maintaining integrity of signals is made easy..!!!

Inside signals
Fourier was quite successful in defining the anatomy of any signal that had finite energy. He with his formula was able to identify the signals that collectively represented a totally new signal. This demands us to view the composed signal in a completely new domain called Frequency domain, for which mathematical transforms like Fourier Transform, Laplace Transforms are greatly useful. You can find the fourier transform of standard signals here. The signal that had different amplitudes, phases and more importantly frequencies when added up forms a new signal. Let us limit our scope of discussion to only digital signals. The following figure shows an example of ideal digital signal representing 1’s and 0’s.
Note that we are considering a typical digital pulse that has its levels continuously changing between 1 and 0. The frequency content of the above signal is shown in the following figure. Note that the digital signal that is shown looks like a SINC signal in the frequency domain. That means the digital signals are composed of infinitely large set of sine waves of different amplitudes and phases with frequencies progressively aligned. When these progressively aligned signals when viewed in time domain, represent a digital pulse.

Ideally the signal received in the other end of the transmission line must be very similar to the originally transmitted signal. The transmission line, when viewed as a system, must show a unit impulse response of H(ω) = 1, such that the signal is received at the receiver with no modification. But in reality all systems are band limited and so the transmission lines conduct only a certain band of signals that the signal falling out of its conduction band are lost in the line or reflected back to the receiver. Thus the signal shape of the received digital signal starts approaching that of a trapezoidal signal. A trapezoidal signal is seen at the receiver only if the conduction band has a flat response. If the conduction band is not flat enough the signals the received end loses its integrity and looks similar to the one shown below.

To maintain the signal shape at the receiver do,
• Have a second look at the transmission line
• not believe signals are digital
• Match the spectrum of signals to that of the transmission line

The next issue shall discuss the transmission line characteristics and reflection phenomenon using simple simulations.