Block Diagram Of A Communication System Computer Science Essay

The doorway to the nowadays wireless communication systems was opened by Guglielmo Marconi when he transmitted the three-dot Morse code for alphabet ‘ S ‘ by the use of electromagnetic waves over a 3-KM link in 1895. This laid the foundation of modern communication systems ranging from broadcasting, satellite transmission and radio eventually progressing to nowadays cell phones. It wouldn’t be wrong to say that wireless communication has indeed revolutionized our present society

A sudden increase has been observed in the expansion of radio systems during the last two decades. We have seen great evolution in Wireless communication systems from 1G narrowband analog systems in the 1980s to the 2G narrowband digital systems in the 1990s. Now the existing 3G wideband multimedia systems are being deployed. In the meantime, research and progress in the future-generation wideband multimedia radio systems is vigorously being pursued worldwide.

To connect mobile users to the public switched network the United States introduced first radiotelephone service by the end of the 1940s. Improved Mobile Telephone Service was launched by Bell Systems in 1960s due to which lots of improvements like direct dialing and increase in bandwidth took place. IMTS formed the bases of the first analog cellular systems. The term cellular was used due to the fact that coverage areas were split cells, they had a low power transmitter and receiver.

BLOCK DIAGRAM OF A COMMUNICATION SYSTEM

Figure 1. Block diagram of a general communication system.

ANALOG vs. DIGITAL COMMUNICATION SYSTEMS

Definition of Digital

A method of storing, processing and transmitting information through the use of distinct electronic or optical pulses that represent the binary digits 0 and 1.

Advantages of Digital

low-priced

reliable

Easy to manipulate

Flexible

Compatible with other digital systems

The information in digital form can only be transmitted without any degradation through a noisy channel

Incorporated networks

Disadvantages of Digital

Sampling Error

As compared to analogue, larger bandwidth is required in digital communications for the transmission of the same information.

Synchronization in the communications system is required to recognize the digital signals, but this is not the case with analogue systems.

Definition of Analogue

Analogue is a transmission standard that uses electrical impulses to emulate the audio waveform of sound. When you use a phone, the variations in your voice are transformed by a microphone into similar variations in an electrical signal and carried down the line to the exchange.

Advantages of Analogue

less bandwidth is required

More Accurate

Disadvantages of Analogue

Signal loss and distortion can be seen due to the effects of random noise which is impossible to recover

GENERATIONS OF CELLULAR SYSTEMS

The concept of cellular telephony was introduced in AMPS, short for Advanced Mobile Phone Systems. AMPS divided the total area into small regions called cells and this was from where the concept of cellular telephony started. Cellular Systems had many advantages such as they increased quality, capacity, reliability and availability of mobile telephone network. The generations of cellular systems are described below.

FIRST GENERATION CELLULAR SYSTEMS

First generation cellular telephone systems were introduced in 1980’s. They were based on Analog Frequency Modulation technique. Each channel was assigned a sole frequency.

First generation cellular systems offered only wireless voice services based on analog technology. Digital signals were only used for control information such as dialing a number etc. These systems were no able to cope with the increasing demands of users also they had very less capacity and provided poor voice quality.

Some first generations systems are

Advanced Mobile Telephone System, AMPS

NAMPS, AMPS

Total Access Cellular System (TACS)

Nordic Mobile Telephone System (NMT-900)

SECOND GENERATION CELLULAR SYSTEMS

Second Generation Cellular Systems provided larger capacity and provided much better services to users compared to first generation systems. They were based upon Digital Modulation technique which led to great enhancement in network’s capacity.

Second Generation Systems used multiple access techniques such as TDMA and FDMA.

The biggest draw back of Second Generation Systems was that its different systems were not compatible with each other. Therefore roaming between different systems was not possible.

Some of Second Generation Systems are

North American Digital Cellular, NADC

Global System for Mobile Communication, GSM

Pacific Digital Cellular, PDC

CDMAONE, IS-95 CDMA

In order to overcome Second Generation compatibility problem with increased data rates of modern internet applications, 2.5 Generation standards were developed. The best thing about them was that they allowed already existing Second Generation systems to be upgraded so that they can be used for higher data rate transmission.

2.5 Generation brought a new revolution in cellular telephony by supporting services like high speed internet and location based mobile services.

Some of 2.5 Generation Mobile Systems are

General Packet Radio Service, GPRS

Enhanced Data Rate for GSM Evolution, EDGE

THIRD GENERTAION CELLULAR SYSTEMS

Designed to provide high quality and high capacity in data communication, Third Generation Systems require sophisticated spreading and modulation techniques.

Third Generation Systems are aimed to provide voice quality comparable to land line telephony and also to support high data rate.

These systems are compatible with circuit switched as well as packet switched data services. They are also compatible with the existing networks and use radio spectrum much more efficiently than before.

Some Third Generation Systems are

Wideband CDMA, WCDMA

Universal Mobile Telephone System, UMTS

CDMA 2000

BEYOND 3G

The highly developed version of the 3G mobile communication are the 4G mobile communication services. It is estimated that 4G mobile communication services will give increase in capacity, data transmission with high speed, broadband, HQ color video images for users, graphic animation games in 3D, audio services in 5.1 channels. For the system and architecture of 4G mobile communication many researches are done. Developments are made in the terminal protocol technology for high speed packet services, larger capacity, enabling downloading application programs by public software platform technology, multimode radio access platform technology, and high quality media coding technology over mobile networks.

Why 4G?

Services like wireless internet and teleconferencing can be carried by 4G.

Global mobility and service portability.

Wider bandwidths.

Increased bit rates.

Less expensive.

Mobile networks can easily be scaled.

CHAPTER # 02

Multiplexing is a process in which a single carrier is used to transmit several different signals. These several signals are transmitted all together by combining them and forming one signal that will effectively move through the carrier bandwidth. When one transmission is done and the signal reaches the destination point, the integrated signal re-assembles into its actual form and is then received.

Multiplexing is one of the most used techniques today in almost every communication system. Because of the technological advance multiplexing, we have seen major increase in efficiency of a wide range of telephony services and online applications.

Multiplexing has become an effective technique that assists in everything from video conferences and web conferences up to bulk data transmissions and even making a simple Point-to-Point phone call.

FDMA:

FDMA is the most usual technique used for multiple accessing. FDMA stands for frequency division multiple access. It is clear from its name that in this technique the frequency is divided among the users as the available spectrum is shared among them in the frequency domain. The message signals are transmitted onto carriers for different users using particular RF frequencies. Within FDMA structural design the Single Channel Per Carrier (SPSC) is the simplest method where each channel is provided with a separate carrier. This scheme finds its essence in the fact that the channels are assigned on the basis of demand. Within a cell all the channels are available to all users all the time, and the channels are assigned as soons as a message signal is received or a request is made.

Guard bands are used to reduce the chances of interference from adjacent channels. These guard bands are present between the bands allocated for various channels.

In the implementation of the first analog cellular systems, FDMA is the multiplexing technique that was used.

TDMA:

Time division multiple access techniques allots different time intervals to different users for the transmission of signals and storage of the data is carried out in one frequency channel not like FDMA which uses one frequency per channel.

Users are allowed to use the same frequency but the time slots are divided.

In TDMA techniques the available spectrum is divided into small frequency bands as in FDMA, which are further sub-divided into various time slots. The user can access the frequency channel only for time slot allotted to him. User can use periodically the particular duration of time.

In TDMA systems, guard bands are required between both frequency channels and time slots.

SDMA:

SDMA stands for Space-Division Multiple Access. It is a MIMO (Multiple-Input, Multiple-Output, a multiple antenna schematic architecture) based wireless communication network architecture. It enables access to a communication channel by the process of identifying the user location and establishing a one-on-one mapping between the network bandwidth allotment and the acknowledged spatial location that is why it’s mostly suitable for mobile ad-hoc networks.

For majority of the well known mobile communication architectures such as CDMA, TDMA and FDMA, SDMA architecture can be configured and implemented

CDMA:

CDMA stands for Code division multiple access. CDMA systems are based on the spread spectrum technique. In which transmissions by all the users are carried out simultaneously while operating at the same frequency and using the entire spectrum bandwidth.

For the identification and extraction of required transmission, each user is allotted with a unique code which cannot match with any other user. This issue of identification is due to the fact that all the users transmit simultaneously. To ensure this privacy, pseudo-random noise codes or PN codes are used.

These codes are actually the orthogonal codes and its advantage is that it reduces the chances of cross correlation among themselves. By using this PN code assigned to the specific user, modulation of the message signal from an individual user is done. Then we have the CDMA frequency channel through which all the modulated signals from different users are transmitted. At the receiver’s end, the desired signal is then recovered by de-spreading the signal with a replica of the PN code for the specific user. The signals whose PN codes are not matched with the desired signal and are assigned to different users are not de-spread and as a result are regarded as noise by the receiver.

CDMA differs from both TDMA and FDMA in a way that it allows users to transmit the signal at the same time and operate at the same nominal frequency so it requires less synchronization whereas in TDMA and FDMA frequency and time management is very critical so more dynamic synchronization is required. One more advantage of CDMA is that complete system’s spectrum is used by signals and hence no guard bands are required to protect against adjacent channel interference.

Intro to Spread Spectrum Communications

Following are the major elements that can clearly describe the Spread Spectrum communications:

By spread spectrum, bandwidth far in excess is available than that is necessary to send the information. Due to this characteristic the transmission can be protected against interference and jamming at the same time providing multiple access capability.

An independent code known as the Pseudo random code is used for signal spreading across the bandwidth. The distinct nature of this code separates spread spectrum communications from typical modulation techniques in which modulation always spread’s the spectrum somewhat.

For the recovery of the original signal the receiver is synchronized to the deterministic pseudo random code. Users can transmit the signal at the same time and operate at the same nominal frequency by using independent code and synchronous reception.

In order to protect the signal from interference a pseudo-random code is used. It appears to be random to anyone who does not have its pre-defined knowledge but in reality is deterministic, it is because of this fact that receiver is able to reconstruct the code needed for the recovery of the required data signal. This code used for synchronous detection is also called Pseudo noise sequence.

Types of Spread Spectrum Communications

Spreading of bandwidth of the signal can be achieved by three ways:

Frequency hopping

The signal is shuffled between different centre frequencies within the entire bandwidth available to the hopper pseudo-randomly, and the receiver used already knows where to look for the signal at a given time.

Time hopping

The signal is transmitted in short bursts pseudo-randomly, and the receiver knows when a burst is expected.

Direct sequence

Very high frequency is used to code the digital data. The code is pseudo-randomly generated. The same code is generated at the receiver end, and in order to extract the original data this code is multiplied to the received information stream.

CHAPTER # 03

SOURCE CODING AND DIGITAL MODULATION

3.0 INTRODUCTION

Digital Modulation is performed in order to represent digital data in a format that is compatible with our communication channel.

Why Digital Modulation? Digital modulation schemes have greater capacity to convey large amounts of information than analog modulation schemes.

3.1 DIGITAL DATA, DIGIITAL SIGNAL

Digital signal is binary data encoded into signal elements. Different encoding schemes for encoding digital data into digital signal are:

3.1.1 Non Return to Zero (NRZ)

In NRZ there are two different voltage levels for 0 and 1. There is no transition in the middle of the bit. The absence of signal denotes 0 and a positive voltage level denotes 1.

Figure 3.1, Non Return to Zero (NRZ)

The major drawback of NRZ scheme is that it adds a dc component to the signal.

3.1.2 Multilevel Binary (AMI)

In this encoding scheme there are more than two levels. No signal represents 0 and 1 is represented by some positive and negative voltage level. 1’s pulses are opposite in polarity.

Figure 3.2, Multilevel Binary (AMI)

There is no dc component in this scheme and also there is no loss of synchronization for consecutive 1’s.

3.1.3 Manchester Coding

There is transition in middle of each bit, which acts as a clock as well as data. The low to high transition represents 1 and high to low represents 0.

Figure 3.3, Manchester Coding

3.1.4 Differential Manchester

In this scheme transition at the middle of the bit represents only clocking while transition at start represents 0 and no transition at start represents 1.

Figure 3.4, Differential Manchester

3.2 ANALOG DATA, DIGITAL SIGNAL

Analog data is first converted into digital data by using analog to digital converters. These converters use different techniques to complete their task, some of them are:

3.2.1 Pulse Code Modulation

If a signal is sampled at regular intervals at a rate higher than twice the highest signal frequency, the samples contain all the information of the original signal. Each sample is assigned a digital value. Although its quality is comparable to that of analog transmission but still in this process some information is lost and the original signal can never be recovered.

Figure 3.5, Pulse Code Modulation

Delta Modulation

Analog input is approximated by a staircase function. Function moves up or down at each sample interval by one level (d).

Figure 3.6, Delta Modulation

Delta modulation is easier than PCM in implementation, but it exhibits worse signal to noise ratio for the same data rate. But it is good for data compression.

DIGITAL DATA, ANALOG SIGNAL

Different digital modulation techniques are:

Amplitude Shift Keying (ASK)

A modulation technique in which digital data is represented as variations in the amplitude of a carrier wave is called Amplitude-shift keying (ASK). One binary digit is represented by presence of carrier, at constant amplitude and the other binary digit represented by absence of carrier.

Figure 3.7, Amplitude Shift Keying (ASK)

3.3.2 Frequency Shift Keying (FSK)

In frequency shift keying different frequencies are used to represent incoming digital data. Say in case of Binary Frequency Shift Keying f1 is used to represent 0 while f2 is used to represent 1.

Figure 3.8, Frequency Shift Keying (FSK)

In MFSK more than two frequencies are used and hence bandwidth is more efficiently utilized.

3.3.3 Phase Shift Keying (PSK)

A digital modulation technique in which data is transmitted by modulating and changing the phase of the reference signal is called Phase-shift keying (PSK). In case of PSK, a finite number of phases are used. A unique pattern of binary bits is assigned to each of these phases. Generally, each phase encodes an equal number of bits. The symbol is formed by each pattern of bits that is represented by the particular phase.

Figure 3.9, Phase Shift Keying (PSK)

Figure 3.10, Constellation Diagram of BPSK

The bandwidth of ASK and PSK are specified as:

Whereas the bandwidth of FSK is given as:

Where,

R is the bit rate

DF = f2 – fc = fc – f1

CHAPTER # 04

CHANNEL CODING

4.0 INTRODUCTION

Why Channel Coding? In modern digital communication systems information is represented in bit streams, which are then modulated to analog waveforms before being transmitted onto a channel. At receiver this analog information is demodulated into bit streams, but because of the presence of interference and noise in the communication channel this bit stream may be corrupted. So to minimize occurrence of bits in error and protect digital data from channel noise and interference channel coding is used.

How Channel Coding is performed? Additional redundant bits are added to the message data stream to perform channel coding, these extra bits assist in error detection and correction at the receiver’s end.

Channel Coding at the cost of? Channel Coding is performed at the cost of bandwidth expansion and data rate reduction.

4.1 TYPES OF CHANNEL CODING TECHNIQUES

There are two main types of channel coding techniques,

Block Codes

Convolutional Codes.

Block Codes accepts k number of information bits and generate a block of n number of encoded bits, and thus are commonly known as (n.k) block codes. Some common examples of block codes are Hamming Codes and Reed Solomon Codes.

Convolutional Coding is forward error correction technique that is currently most widely used in modern communication systems, this particular technique is used for real-time error correction. Unlike block codes which append redundant bits at the end of original message signal, Convolutional coding form a new codeword using original data stream. The encoded bits are not solely dependent on k current input bits but at the same time on precedent input bits.

4.2 CONVOLUTIONAL CODES

In this project Convolutional Coding is implemented. Convolutional Codes are further classified as 1. Trellis Coded Modulation (TCM) 2.Turbo Codes.

Trellis Coded Modulation (TCM) is non recursive, non systematic and does not require an interleaver.

Turbo Codes on the other hand are recursive, systematic, parallel structured and they also require interleaver.

In Wideband CDMA systems TCM is used for all channels while Turbo Codes may be used for DCH and FACH channels. Turbo Codes are sometimes classified as separate branch of Channel Codes so from here onwards word Convolutional Code will only be used for TCM.

Types of Transmission Channels

Coding Schemes

Coding Rate

RACH

Convolutional Coding

1/2

BCH

PCH

DCH, FACH

1/2, 1/3

Turbo Coding

1/3

Table 4.1, WCDMA Specifications

4.3 CONVOLUTIONAL CODE REPRESENTATIONS

4.3.1 Polynomial Representation

No. of input information bits = k

No. of encoded bits = n

No. of stages (Constraint Length) = K

Code Rate = k/n

Encoded CodeWord = U

The following example shows how Convolutional Codes are represented.

Let g1(x) and g2(x) be encoder polynomials, where

g1(x) = 1 + x + x2

g2(x) = 1 + x2

Let input message bit stream be 101, therefore input message bit stream polynomial will be,

m(x) = 1 + x2

The encoded codeword ‘U’ will be combination of product of g1(x) with m(x) and g2(x) with m(x),

m(x) x g1(x) = 1 + 1.x + 0.x2 + 1.x3 + 1.x4

m(x) x g2(x) = 1 + 0.x + 0.x2 + 0.x3 + 1.x4

Therefore the codeword ‘U’, becomes

U = (1,1) + (1,0).x + (0,0).x2 + (1,0).x3 + (1,1).x4

U = 1 1 1 0 0 0 1 0 1 1

4.3.2 State Transition Diagram

Convolutional Coding can be represented using State Transition Diagram. Following are State Transition Diagram and State Transition Table for code rate 1/2.

Figure 4.1, State Transition Diagram for Code Rate ½

Table 4.2, State Transition Table for Code Rate ½

Again for the same input bit stream 10100, the codeword U = 11 10 00 10 11. In the input message last two ’00’ bits are tail bits.

4.3.2 Block Diagram Representation

The following diagram shows block diagram representation of Convolutional Coding with

Code Rate = 1/2

Constraint Length (No. of Stages) = 3

Figure 4.2, Block Diagram Representation of Convolutional Code with Code Rate = ½

The following example illustrates the process of Convolutional Coding using block diagram representation for input bit stream 101.

So the final codeword becomes, U = 11 10 00 10 11

4.3.2 Trellis Diagram Representation

For input bit stream 101, the following diagram shows how Convolutional Coding is performed using Trellis Diagram

Figure 4.3, Trellis Diagram Representation

CHAPTER # 05

PULSE SHAPING TECHNIQUES

3.0 INTRODUCTION

Why Pulse Shaping? It is done in order to reduce Inter Symbol Interference commonly known as ISI.

How Pulse Shaping is performed? In order to achieve zero-ISI the overall system response must be equal to Nyquist frequency response.

5.1 RAISED COSINE FILTER

Inter Symbol Interference significantly degrades the data detector ability to differentiate between a current symbol from diffused energy of adjacent symbol. This leads to the detection of error and increases BER. So in order to cater ISI, a real-time realization of Nyquist filter is applied in modern communication systems. Raised cosine filter is one of the realization of Nyquist filter.

where r = ‘roll-off factor’ = 1 ≤ r ≤ 0

and T = ‘symbol period’ = 1/R

Roll-off factor determines the filter bandwidth and represents a trade-off between the sharpness of the transition band of the filter and impulse response ringing magnitude of the filter.

A Nyquist filter has following properties:

Time response eventually goes to zero in a time period exactly equal to the symbol spacing.

By sampling the symbol sequence at a given symbol time point, present symbol is not affected by the energy spreading from the adjacent symbols.

The impulse response and the frequency response of the RC filter is

Figure 5.1, Impulse Response of RC Filter

Figure 5.2, Frequency Response of RC Filter

Time response of the RC filter goes to zero with a period that exactly equal to the symbol spacing. As the response equals zero at all symbol times except for the desired one none of the adjacent symbols interfere with each other.

5.2 ROOT RAISED COSINE FILTER

RC filter is divided into a root raised cosine (RRC) filter pair, with one at the transmitter end, which performs the pulse shaping in order to constrain the modulated signal bandwidth, and the other at the receiver end, that performs matched detection for optimizing the SNR of a known signal in AWGN presence.

The Root Raised Cosine filter is so named because its transfer function exactly is the square root of the transfer function of the Raised Cosine filter.

Where r = roll off factor and T is symbol period. The RRC filter bandwidth is equal to the root mean square (RMS) amplitude 2R.

The impulse response and the frequency response of the RRC filter is

Figure 5.3, Impulse Response of RRC Filter

Figure 5.4, Frequency Response of RRC Filter

Both RC and RRC have similar pulse shapes, but the RRC pulse makes slightly faster transitions, therefore the spectrum of RRC pulse decays more rapidly as compared to the RC pulse. Another important difference between both pulses is that the RRC pulse does not have zero Inter Symbol Interference. Because of the fact that RRC filter is used at transmitter and receiver both, the product of these transfer functions is a raised cosine, which will result in zero ISI output.

5.3 ROLL OFF FACTOR

The roll-off factor, r, is a measure of the excess bandwidth of the filter, i.e. the bandwidth occupied beyond the Nyquist bandwidth of 1/2T.

Where ∆f is excess bandwidth and Rs is symbol rate.

CHAPTER # 06

SPREAD SPECTRUM

Spread spectrum is a type of modulation where the data is spread across the entire frequency spectrum. This process of spreading the data across the entire spectrum helps signal against noise and interference. These techniques are mostly employed in cell phones and also with wireless LAN’s.

To qualify as a spread spectrum signal, two criterions must be met

The transmitted signal bandwidth must be in excess of the information bandwidth.

Some function other than the data being transmitted is used to establish the bandwidth of the resultant transmission.

Why Spread Spectrum ?

Due to its exclusive and peculiar properties spread spectrum is preferred over other modulation schemes. Some of these properties are characterized as advantages and disadvantages of a basic spread spectrum system below.

Advantages

• It reduces the effects of multipath interference and at times removes them entirely.

• Frequency band is shared simultaneously with other users.

• Pseudo random codes ensure protection of transmission and privacy.

• As the signal is spread over an entire spectrum it has a low power spectral density.

Disadvantages

• Due to spreading operation it consumes more bandwidth.

• It is at times difficult to implement.

Types of Spread Spectrum Techniques

Most commonly used techniques in a spread spectrum systems are

Direct Sequence Spread Spectrum

Frequency Hopping Spread Spectrum

Frequency Hopping Spread Spectrum

A frequency hopping spread spectrum hops from one narrow band to another all within a wider band. In general the frequency hopper transmitter sends data packets at one carrier frequency and then jumps to another carrier frequency before sending ore packets and continues the same routine throughout the period of transmission. The pattern that emerges seems to be random but is in fact periodic and easily traceable by pre configured transmitter and receiver. These systems can be vulnerable to noise at a particular hop but usually are able to send packets during the next hop.

Direct Sequence Spread Spectrum

Most widely used technique of spread spectrum is the Direct Sequence Spread Spectrum. A Direct Sequence Transmitter receives the incoming data stream which is to be transmitted and then converts it into a symbol stream where the size of a symbol can be one or more bits. Using any of the modulation schemes for digital systems such as Quadrature Amplitude Modulation (QAM) or Quadrature Phase Shift Keying (QPSK) this symbol stream is multiplied to a noise like sequence known as pseudo random sequence. It is also know as a chip sequence. As a result of this multiplication the bandwidth of the transmission is significantly increased.

Figure 3. Direct Sequence Spread Spectrum System

Figure 3. shows the working of a basic Direct Sequence Spread Spectrum system. For clarity purposes, one channel is shown working in one direction only.

Transmission

For each channel a distinct and different Pseudo random code is generated.

In order to spread the data the data stream is multiplied with the previously generated Pseudo random code.

The signal obtained as a result of this multiplication is then modulated onto a carrier.

This modulated carrier waveform is then amplified before broadcasting.

Reception

The carrier wave is amplified as soon as it is received by the receiver.

The signal received is then multiplied with a locally generated carrier which gives the spreaded signal.

Again a Pseudo random code is generated on the basis of the signal expected.

The process of correlation is carried out on the received signal and the generated code which gives the original message signal.

Pseudo-Random Noise

The spread spectrum systems are constructed very similar to other conventional systems. The difference being the addition of pseudo random generators both at the transmitter and the receiver which generate the Pseudo noise sequences required for the functionality of Direct Sequence spread spectrum. These pseudo random noise sequences are used for spreading the signal at the transmitter side and dispreading at the receiver side. A pseudo noise sequence is made up of a number of symbols which are called chips. It is imperative that a good code is selected for its type and length directly affects the system capability.

A unique code is generated for each channel. All the transmissions for a particular user are then added together. At the receiver each user generates its own matching code in order to recover the original signals.

In order to qualify as a pseudo random code and proper functionality a code must have the following properties.

The pseudo random code must be deterministic so that the receiver can generate the exact code as used by the transmitter

It should be random for any listener who does not have the knowledge of the code being used so that it appears as noise to any interferer or eavesdropper.

Two codes must be identical and have no or very small relevance to each other.

The code should not repeat shortly there must be a long time before that happens.

Maximum length sequences

Maximum Length sequences can be generated by the help of shift registers with feedback applied on them. These sequences meet all the conditions for spreading sequences very strictly. The cross correlation between this sequence and noise is very low, which helps in detecting signal out of noise in the receiver. These sequences are very useful for encryption as they also have a very low cross correlation with each other.

The randomness properties of maximal length sequences can be seen here.

Gold Sequence

In order to create two gold sequences, two maximum length sequences are to be combined. They have a very low auto-correlation which enables CDMA systems to transmit asynchronously. Gold sequences are constructed by modulo-2 addition of two maximal length sequences of same length chosen from pairs of preferred m-sequences.

Walsh Codes

Walsh codes have to be created from hadamard matrices. All generated Walsh codes would be orthogonal to each other. The basic hadamard matrix is shown below. These sequences provide low cross-correlation between each other. Secondly, the number of 1’s is same as the number of 1’s in each codeword.

By looking at the matrix above, Walsh codes with different lengths can be generated with the help of recursion. For a clear understanding Walsh codes with length equal to 4 are illustrated below.

=

Spreading the Signal

Signal spreading is achieved when the signal is multiplied bitwise with the pseudo random sequence. Figure 4. Below gives an insight into how two bits are multipliied to a pseudo random sequence having a length of 8 and the its corresponding effect on the frequency spectrum. Spreading process of two bits with the help of an 8-chip PN sequence is shown on the left side and the effect of this spreading on the signal’s frequency spectrum is shown on the right side of figure. 4

Figure 4. The principle of signal spreading

De-spreading the Signal

De-spreading of the signal is carried out by multiplying each bit of the signal to its correspond pseudo random code and the result is summed bitwise. Figure 5. describes the process of de-spreading when a pseudo random signal of length 8 is used for transmission and the de-spreaded signal is equal to the spread signal. De-spreading process of two bits with a PN sequence of 8-bit is shown on left side and the result of the de-spreading on the signal’s frequency spectrum is shown on the right.

Figure 5. The principle of signal de-spreading

Effect of Spreading on Bandwidth

Figure 1. As a result of spreading operation the signal’s frequency is spread over a wider bandwidth

Spread spectrum modulation is applied in tandem with some other digitial modulation schemes such as QAM or QPSK.

Effect of Despreading on Bandwidth

Figure 2. The despreading operation recovers the original signal.

A spread spectrum modulation is applied in coherence with a conventional demodulation procedure in order to recover the original signal.

Multiple Access Capability

The received signal s1 is detected by the receiver summed with the other signal s2. The signal s1 which is for user 1 is only extracted at the receiver for user 1 whereas the signal s2 appears to be as noise in comparison to the user 1 signal s1. This is shown in Figure 6. below

For efficient recovery of the data from user 1 it is must that the signal appearing as noise has adequately low energy in the information bandwidth of signal s1.

Spreaded data multi-users

De-spreaded data multi-users

Spreaded data multi-users

De-spreaded data multi-usersFigure 6. Multiple Access Capibility when user 1 recovers signal s1

CHAPTER # 07

DESIGN OF A CDMA BASED COMMUNICATION SYSTEM

7.1 THE TRANSMITTER

7.1.1 Source Encoding

In order to make incoming message compatible with the communication system source coding is used. In this project NRZ encoding scheme is used so that the incoming message signal becomes compatible with the walsh codes which are also in NRZ form. NRZ encoding provides a better utilization of the bandwidth, also if we compare it with AMI scheme it is more efficient because in that receiver has to distinguish between three different levels and it requires more signal power. While the Manchester and differential Manchester schemes have modulation rate twice or more than that of NRZ so hence they require greater bandwidth.

7.1.2 Channel Encoding

Channel encoding is performed so that bit errors can be reduced at the receiver; also it protects the data from channel noise and interference. In this project we have used Convolutional Encoding technique and then have tried to upgrade our system using Low Density Parity Check (LDPC). LDPC improves channel capacity as it approaches to Shannon capacity limit. Both Convolutional encoder and LDPC encoder form a codeword based on the incoming message signal. We have used ½ rate encoders that means that we have two input bits and the output is going to be four bits. For Convolutional encoder inbuilt matlab trellis structure along with convenc command has been used while in case of LDPC fecldpc.enc command is used to make LDPC object using sparse H matrix and then message is encoded using LDPC object by encode command.

7.1.3 Spreading

The fundamental criterion of spreading is cross correlation and auto correlation of message symbols. In case of a CDMA based communication system as there are many users that simultaneously transmit data so the system must have good cross correlation properties. This generates the need of orthogonal codes so Walsh codes are used. Walsh codes have good cross correlation properties. In this project direct sequence spread spectrum (DSSS) technique has been employed to spread the signal across the entire bandwidth.

7.1.4 Digital Modulation

Modulation is a technique that facilitates the message signal to be transmitted over the medium. In case of digital modulation, digital signal is modulated using amplitude shift keying, frequency shift keying and phase shift keying etc. We have implemented Binary PSK, BPSK uses two symbols that have opposite phase to each other to modulate 0 and 1.Sampling frequency of 44 kHz and a data rate of 2 kHz has been used in this system. In case of LDPC modulate command has been used to modulate the signal.

7.1.5 Pulse Shaping

The effect of inter symbol interference (ISI) is minimized by the help of pulse shaping process. In this project Root Raised Cosine Filter has been used at the transmitter side to shape the pulse so as to limit the bandwidth of the modulated signal. By using this technique bandwidth is limited but the signal spreads in so there must be a threshold in order to counter this problem, this threshold is known as Nyquist limit. It is basically a low pass filter with a roll off factor 0.5 which has been used in this project.

7.1.6 Symbol Packing

In symbol packing, a training sequence is appended at the start and end of the signal. Also in addition to this about thousand zeros are inserted at the end and start of the signal. These are added so that even if the receiver fails to receive initial bits these bits are additional zeros and not the message signal.

7.2 THE CHANNEL

In a communication system channel could be wired or wireless. Different channels have different characteristics. When ever we talk about a channel we talk about noise, it is basically the unwanted electrical signal that confines the receiver ability to take correct decisions. Channel noise degrades the output very quickly. Small channel doesn’t make much difference to the output signal but large channel noise considerably degrades the signal. In case of small channel noise the only noise that is present in the system is quantization noise. Another important factor in a communication system is Inter symbol interference (ISI). When the channel bandwidth is not much greater than signal bandwidth the spreading of the signal causes ISI.

Normally, the wireless channel is modeled with AWGN channel. A thermal noise source is used to spread an equal amount of noise power per unit bandwidth over all frequencies that is the reason why a simple model for thermal noise presumes that its power spectral density Gn(f) is flat for all frequencies, as is denoted as:

Gn(f) = N0/2 watts/hertz

7.3 THE RECEIVER

7.3.1 Synchronization

In this system, it is assumed that the receiver is on all the time. Therefore whenever the transmitter will send the message signal, there should be some way through which the receiver can determine where its signal of interest is. This can be done by correlating the training sequence with the received signal. The part where the maximum correlation occurs would be the start of the signal and the second maximum in the correlation would give the end of the message signal.

7.3.2 Matched Filtering

The received signal is passed through root raised cosine filter in order to perform matched filtering and down sampling at the receiver end. Usually RC filter is divided into a root raised cosine (RRC) filter pair, with one at the transmitter end, which performs the pulse shaping in order to constrain the modulated signal bandwidth, and the other at the receiver end, that performs matched detection for optimizing SNR of a known signal in AWGN presence.

7.3.3 Demodulation

The signal is demodulated by multiplying it by the carrier signal. The output would be demodulated signal, but this demodulated signal would be sum of all the spreaded signals.

7.3.4 Despreading

In order to extract the individual message signal of each user, the spreaded signal is then again multiplied by the Walsh code and is then integrated over a symbol time.

7.3.5 Channel Decoding

If the channel coding was not employed by the transmitter, then the despreaded messages would be the received signals. In the channel coding case, the signal will have to be decoded by using the viterbi decoding function in matlab.

7.3.6 Source Decoding

While performing source coding we had used NRZ encoding scheme, so now at the receiver this encoded message is again decoded back to its original form.