Lecture 24

Source Coding COMMUNICATION SYSTEMS Lecture # 24 5th May 2007 Instructor

WASEEM KHAN

„

Source coding deals with the task of forming efficient description of information sources.

„

Efficient description permits a reduction in memory or bandwidth resources required to store or to transport the source data.

„

Depending on the type of source there are a number of categories of source coding.

„

If the source data is digital and it is intended to reduce the number of bits to represent it, we may call this process as data compression.

Centre for Advanced Studies in Engineering

Data Compression „

„

„

„

Some of the data compression techniques are loseless in which the original data can be recovered by decompressing. Files compression using WINZIP, WINRAR, etc. are examples of lossless compression. In lossy compression techniques not original but near to original data is recovered. JPEG, GIF, MPEG are examples of lossy compression techniques. Lossy compression techniques are acceptable as far as the recovered information is not much different from the original one. If the compressed speech, image or video is still comprehensible, it may be acceptable.

Source Coding in a Typical Digital Communication System

Block Diagram of a Typical Digital Communication System

Run-Length Codes* „

In many applications, lengthy runs of specific symbols are found.

„

Rather than code each symbol of a lengthy run, it makes sense to describe the run with an efficient substitution code.

„

The most important commercial example is facsimile coding, used for transmitting documents.

*see “Digital Communications” by Bernard Sklar for this topic (section 13.7.3)

„

The relative darkness or brightness of the scanned image at each pixel position is categorized as either black (B) or white (W) intensity levels.

„

Thus the signal observed during a scan line is a twolevel pattern representing the B and W image intensity.

„

A horizontal scan line across the image will exhibit a pattern consisting of long runs of B and W levels.

„

The standard CCITT run-length coding scheme to compress the runs of B and W levels is based on a modified variable-length Huffman code*.

„

For certain run-lengths of B and W, specific codewords are defined.

*Visit www.cs.duke.edu/csed/poop/huff/info/ to understand basic Huffman coding

Huffman Coding for Facsimile Transmission „

Facsimile transmission is the process of transmitting a two-dimensional image as a sequence of successive line scans.

„

The most common images are in fact documents containing text and figures.

„

The standard CCITT* document size is 11.7” x 8.27”.

„

The image of the document is divided into 1188 horizontal scan lines and each line is subdivided to 1728 pixels/line.

„

Hence there are 2,052,864 pixels per document.

*Comité Consultatif International Téléphonique et Télégraphique

„ „

„

„

In the coding scheme, there are two types of codewords. Makeup codewords represent runs of B and W having length integral multiple of 64 (up to 1728). Terminating codewords represent runs having length 0 to 63.

Example

Use the modified Huffman code to compress the line 200 W, 10 B, 10 W, 84 B, 1424 W consisting of 1728 pixel elements Solution „

192W

8W

10B

10W

64B

20B

Carrier Synchronization* „

In coherent systems it is essential to synchronize the phase of the local carrier with that of incoming carrier.

„

Phase synchronization can be achieved with a phaselocked loop (PLL).

„

Consider a normalized input signal r(t): r (t ) = sin[ω 0t + θ ] where ωo is the angular carrier frequency and θ is the phase offset.

EOL

010111 10011 0000100 00111 0000001111 00001101000 000000000001

Only 56 bits are required to transmit this sequence of 1188 bits. *see “Digital Communications” by Bernard Sklar for this topic(section 10.2.1)

„

The voltage-controlled oscillator (VCO) generates a normalized signal x(t):

„

The error signal comprises of two components; one is the sine of the phase difference while the other is a high frequency component.

„

The error signal is applied to a low-pass filter which removes the higher frequency component.

„

The output of the LPF depends on the difference of phases.

„

If this difference is positive, i.e. incoming carrier is leading in phase, VCO will try to catch it by increasing the frequency, and vice versa.

x ( t ) = 2 cos[ ω 0 t + θˆ ( t )] „

„

VCO is a component that oscillates at a nominal frequency if the input voltage is zero; a positive or negative input voltage leads it to oscillate at a higher or lower frequency respectively. The output of the VCO is multiplied with the input signal to get the error signal e(t): e (t ) = x (t ) r (t ) = 2 cos[ ω 0 t + θˆ ( t )] sin[ ω 0 t + θ ( t )] = sin[ θ ( t ) − θˆ ( t )] + sin[ 2ω t + θˆ ( t ) + θ ( t )] 0

Phase Synchronization with Suppressed Carrier „

„

„

„

„ „

The basic PLL works well if the incoming carrier is not modulated. In a suppressed carrier system, the modulated signal does not contain the carrier frequency. Therefore the basic PLL cannot be used in a suppressed carrier system. As an example, consider a binary PSK signal: r ( t ) = m ( t ) sin( ω 0 t + θ ) where m(t) = ±1 with equal probability. In order to track the phase of the carrier signal, the PSK signal is first applied to a squarer.

Wireless Communications

„

r 2 ( t ) = m 2 ( t ) sin = sin = „ „

„

A typical wireless communication system consists of the following basic units ‰ ‰ ‰ ‰ ‰ ‰ ‰

Information Source Modulator / Transmitter Power Amplifier Transmitting Antenna Receiving Antenna Receiver / Demodulator Information Sink

1 2

2

2

(ω 0 t + θ )

(ω 0 t + θ )

[1 − cos(

2 ω 0 t + 2θ ) ]

The second term contains the carrier related component. The phase lock of this frequency component can be achieved by the basic PLL. Input frequency to the basic PLL is 2ω0 , therefore the PLL will try to synchronize with cos(2 ω0 t ). The phase-locked carrier is obtained by the divide-by-2 circuit.

Some Basic Concepts „

„

The output of the squarer is:

„

„

„

„

Usually power amplifier is needed at the front-end of a transmitter. Usually a transmitter generates a low-power signal. The power amplifier stage enhances the signal power to an appropriate level according to the requirement. In radio communication systems the carrier wave is radiated from the transmitter by using a transmitting antenna. The transmitting antenna is a transducer that converts electrical signals into electromagnetic fields. At the receiver, a receiving antenna performs the reverse action; it converts EM fields into electrical signals.

Isotropic and Directional Antennas „

„

An isotropic radiator is a theoretical point source of waves which radiates uniformly in all directions over a sphere centered on the source. It is a reference radiator with which other sources are compared. If an antenna concentrates its radiations in some specific directions, it is called directional antenna. Radiation pattern of a directional antenna

EIRP „

Effective Isotropic Radiated Power (EIRP) is the amount of power that would have to be emitted by an isotropic antenna (that evenly distributes power in all directions and is a theoretical construct) to produce the peak power density observed in the direction of maximum antenna gain. EIRP = Pt Gt

Antenna Gain „

Antenna gain Gt is the ratio of the intensity of a directional antenna's radiation pattern in the direction of strongest radiation to that of a reference antenna, which is usually an isotropic antenna.

„

The antenna gain, Gt, is often expressed in units of dBi (decibels over isotropic).

„

A directional antenna doesn’t generate extra power but concentrates its power in specific directions; hence pointing less power in other directions.

Same power level observed at the two meters