EE 422G - Signals and Systems LaboratoryLab 6 Digital TransmissionKevin D. DonohueDepartment of Electrical and Computer EngineeringUniversity of KentuckyLexington, KY 40506February 19, 2013EE 422G - Signals and Systems LaboratoryLab 6 Digital TransmissionKevin D. DonohueDepartment of Electrical and Computer EngineeringUniversity of KentuckyLexington, KY 40506February 19, 2013Objectives:- Understand how signal information can be encoded for transmission through base-band analog channels.- Observe the impact of typical sources of corruption that create errors in the communication process, namely channel noise and limited bandwidth. - Perform spectral estimation to analyze the relationship between encoded signal bandwidth requirements, channel noise, and bandwidth.1. Background:This lab considers encoding a digital data stream into analog signals for transmissionthrough a base-band analog channel. Various line coding methods for digital base-bandmodulation will be implemented and the spectral properties of the codes will beexamined. Simulation studies will be used to establish a relationship between bit errorrates and signal corruption.The classical model for corruption in communication channel is a band-limited filter(low-pass or band-pass filter) with additive Gaussian noise as shown in Fig. 1. Figure 1. Block diagram of communication system showing the key components that interfere with the information transfer, such as the band-limited filtering and the additive white Gaussian noise.bsBinary SourceBaseband ModulationBand-Limited Filter+Additive Gaussian Noise (AGN)Corrupted Data to DetectorChannelFigure 1 illustrates the main components of a communication source and transmission.The information to be transmitted is converted into a binary sequence (i.e. ASCII codesfor text or sequences of quantized signal amplitudes from a sampled analog waveform).The binary data source emits a series of 1’s and 0’s (bits) representing the communicationsystem source. The baseband modulation operation codes each bit into a continuouswaveform for sending the source sequence over a physical channel. The channel is amedium through which the signal propagates. Typical signals are formed by changes involtages and currents over a wire, light energy over an optical fiber, or electromagneticenergy through the atmosphere. No medium is totally free of noise (random perturbationson transmitted signal), therefore white or colored Gaussian noise is added to thetransmitted signal to simulate this corruption. Noise is an irreversible corruption (i.e.permanent loss of information). While the signals can be filtered to exploit signalredundancies and reduce the impact of noise, there will always be a level of uncertaintyin the received signal. Parameters extracted from signals denoting 1s and 0s must besufficiently separated (i.e. amplitude, frequency, phase …) to limit the impact of theambiguities introduced by the system noise.In addition to noise, channels have a limited bandwidth that can potentially distort thesignals by reducing the frequency content of the signal non-uniformly over its spectrum.So if the coded waveforms are not bandlimited to a value less than channel bandwidth,the waveform will be distorted. Distortion is a deterministic change in the signal and canbe reversed in some cases. If the distortion can be modeled as a linear filter, it can bereversed (the operation of undoing this distortion is referred to as deconvolution). In mostcases of nonlinear distortion, such as clipping or quantization, the distortion cannot bereversed. For the baseband channel, a low-pass filter is used to simulate its distortion. Basebandand low-pass essentially means the same thing when describing a signal. If signals aremodulated with a high frequency oscillator (baseband spectrum shifted up on thefrequency axis), such that they contained no significant DC energy, then the signals areconsidered passband (not baseband). This lab considers only baseband signals andchannels. In order to send information over the channel, the information must be encoded into a sequence of binary digits and these digits are converted into analog signals for transmission using a variety of signaling formats called line codes. To help examine their spectral properties, a Matlab function, modulb(), was written to create various line codes from bit sequences. The function syntax is:>> [y,t] = modulb(binary_sequence, Fd, Fs, line_code_name);where binary_sequence is a vector of 1’s and 0’s denoting the source binary sequence, Fdcorresponds to the binary data rate in bits per second (b/s), line_code_name is a specialstring indicating the particular line code to be used (see help file), and Fs is the samplingfrequency of the line code waveform used by the simulation. Note that sampling rate Fsis used to simulate the analog signal, so this sampling rate needs to be much higher thanthe bit rate (at least by a factor of 10, 100 is preferred if it does not excessively slowdown the system). The modulb function outputs the waveform as vector y andcorresponding time axis t. The command supports the following codes: ‘unipolar_nrz’,‘bipolar_nrz’, ‘bipolar_rz’, ‘ami’, ‘manchester’, ‘miller’, ‘unipolar_nyquist’, and‘bipolar_nyquist’.The type of line coding is selected to meet various system criteria such as powerrequirements, bit timings (additional transitions of the line code signals within the bitinterval can help in timing recovery), bandwidth efficiency (excessive transitions mayrequire more bandwidth than necessary), low frequency content (some channels blocklow frequency), error detection, and complexity. Figure 2 shows analog waveforms forvarious line coding examples. Figure 2. Line code waveform examples. Original binary sequence listed across the top is at a bit rate of 1, and applies to all codes. 2. Pre-Laboratory Assignment1. Use the modulb() function to plot waveforms representing the binary sequenceseq={0,1,0,0,1,1,0,0,0,1,1,1,0,0} using the following line codes at a bit rate ofRd=1kb/s (choose a reasonable sampling rate for potting the waveforms):a) unipolar NRZ (on-off signaling, NRZ= non-return to zero)b) bipolar NRZ (binary antipodal signaling)c) bipolar RZ (binary antipodal signaling RZ