Analysis of Internet Protocol Networks - Coursework Example

DATA COMMUNICATION TCP/IP NETWORKS Introduction Protocols in data communication are the rules and procedures that govern communications between different devices on a network. They define the way in which devices communicate. They govern various aspects like how a modem and computer communicate, how files or e-mails are transferred over the internet r even how a telephone network connect to different telephones. Protocols can be in the different layers like the physical, data link, network or even transport layer protocols (Hudson KURT and Stewart Michael. 1998). TCP/IP protocol suite is one of the most important networking protocols being used today. Many protocols are members of the TCP/IP suite and some of them include: 1 Address Resolution Protocol Arpanet Best Effort Delivery Domain Dotted Decimal Notation Host Address IP Address Multicasting Port Request For Comment Sub-net Address Sub-net Mask History of TCP/IP In the past networks were isolated entities called local area networks. Protocol developers were not entirely concerned with the movement of data from network to another. After time the local area networks grew into multi-networks called inter-networks. ARPANET was a funded research mandated to come up with a prototype inter-network that was called the ARPANET in the 1960. Its goals were to encourage communication among research institutions and have a system that could withstand nuclear disasters. The first ARPANET connected four research institutions. This later developed into the internet that allowed many computers to be connected. This demonstrated the need to have rules that allowed routers to determine the destination of data as either for the local network or for the internet so that it could forward the data as required (Hunt 1992). In 1974 Vinton Cerf and Robert Khan an MIT professor developed a version of the TCP/IP protocols. This protocol was improved over the next years and until today it is the most used protocol. The OSI model has seven layers as compared to five for the tcp/ip protocol. OSI Model Layers 1 Application Layer Presentation Layer Session Layer Transport Layer Network Layer Data Link Layer Physical Layer TCP/IP Model 1 Application Layer Transport Layer Internet Layer Data Link Layer Physical Layer Application Layer This layer allows for the interoperability of the application programs that a user interacts with like email, downloading files. Support protocols in this layer include protocols that will provide booting, management services and host name mapping. Application Layer Protocols The most commonly protocols in this layer that are used to provide services in the internet include the following 1 Hyper Text Transfer Protocol(HTTP) This protocol transfers the Hyper Text Markup Language (HTML) used in the creation of web pages that people access through web browsers like Firefox, chrome, opera in the world wide web across the internet. HTML formats the web pages and its content and the various links used in navigation from one page to another. Simple Mail Transfer Protocol(SMTP This protocol is used in the transfer email that includes sending and receiving of emails across the internet. Email messages send through this protocol should be in the text format. File Transfer Protocol(FTP FTP is used to copy files from one computer to another over a network. It uses TCP to provide reliable services. FTP is mostly used by people when downloading files from the internet. Both the server and client computer that is transferring files between each other must have an FTP client application installed in their operating system for a connection to be established. FTP is not dependent on the operating system of either the server or the client computer (Liu 1994). Telnet This protocol allows for remote log-in into a host computer from another host computer. To effect a telnet connection a user must know the name or the IP address of the remote computer, the user-name and a password if authentication will be required to complete the connection. The function of telnet include remote control of a computer, connecting to a device in order to perform configurations on it, or to manage a remote web server. Domain Name Service This is a centralized service that equates a name with a hosts IP address. DNS names are used in sending emails(tcpip@protocol.com). They are also used in accessing web sites (www.tcpip.com) (Albitz 1992). Transport Layer This layer is responsible for the end to end delivery of data, message segmentation and message reassembling. The protocols in this layer are User Datagram Protocol (UDP) and Transmission Control Protocol (TCP). User Data-gram Protocol This is a connectionless protocol and its key responsibilities are end to end delivery of data and making sure that the data is delivered error free. UDP does not offer facilities for enduring that there is free delivery of data and it does not also guarantee that the data will be arrive at the destination without duplication or loss. UDP was designed to be fast and as a low overhead protocol. The header packet in UDP has fewer bytes than the header in TCP. UDP is mostly used in higher level protocols that have their own error handling and flow control operations like the DNS. It is mostly used in video and audio casting and multimedia transmissions (Albitz 1992). Transmission Control Protocols It is also responsible are end to end delivery of data. This is a more complicated and robust protocol than UDP. It provides a connection oriented and to end delivery. TCP is very reliable as it provides services for error checking, reporting and complete delivery. TCP performs a check-sum of each packet as it arrives and gives a report of either success or failure back to the source of the message. When there is failure the source can resend the message again. TCP provides segmentation of data into smaller packets and inputs a sequence number in the headers as this will be used by the destination in ordering the message in the correct order. TCP also ensures that no packets are lost during the transfer phase and also makes sue there is no duplication of packets. In contrast to to UDP, TCP provides reliable delivery of data but there is slower delivery than UDP. It is used in accurate transfer of large data amounts. UDP/TCP Ports This protocol ensures that a message is delivered to the correct application in the computer. This is important because modern computers allow multiple applications to run at the same time. Each process in the computer has a unique port address. The port numbers range from 0 to 65535. Examples of application and their port numbers are: port 20 – FTP Data Channel. port 21 – FTP control channel. port 23 – Telnet. port 80 – HTTP on WWW and Web browsers. port 139 – Net-BIOS Session Services. Internet Layer This layer provides for routing and network addressing functionality. The protocols in this layer are The Internet Protocol (IP), Internet Control Message, and Internet Group Message Control Protocol. Internet Control Message Protocol – this protocol reports errors with network devices across the network. Internet Group Message Control Protocol – it provides the services of sending one message to a group of many hosts in a network. The process of doing this is called multicasting. Internet Protocol (IP) This protocol facilitates the movement of data between different networks in any data link layer and or physical layer protocols. IP does not guarantee reliable message delivery, it is also connectionless and does not offer error checking and delivery reporting (P.J. 2010). IP Addressing TCP/IP uses three basic addressing types in the movement of data though an inter-network. The type of addressing are physical, logical and port addressing. Physical Addressing – Is used to move data within a single LAN network. The physical address is the Medium Access Control (MAC) address that is always specified in the Data Link Layer frame. MAC address is expressed in 48 bit, Base 16 format (00-06-17-76-BC). Logical Addressing – is used to move data LAN to LAN in an inter-network. This address uniquely identifies a device in an inter-network. The addressing is always represented in Base 2 format (11001010.00001111.01010101.11100010). this is translated into decimal format t make it easily readable by people (142.110.227.1). Port Addressing – is used in end to end or process to process data movement. Addressing Rules The rules that are followed in creation of address are: - the bits used in the host portion of an internet address should not be all one unit. -bits used to define the portion network of an internet address should not be all zero bits. IP Classful Addressing The classes used in IP addressing are: Class A – range from 1.0.0.0 to 126.255.255.254. they were distributed to large corporations and governments and are no longer available for distribution. Class B – range from 128.1.0.0 to 191.254.255.254. They are allocated to midsized organization and also no longer available. Class C – range from 192.0.1.0 t 223.255.254.254. They are allocated to small organizations but this addresses are shrinking. Class D and Class E are reserved for special addressing Data Link and Physical Layer Responsible for sending and receiving TCP/IP packets on the network medium. LAN technologies applicable in this layers are Ethernet, Token Ring, FDDI. WAN technologies applicable in this layers are Frame Relay, ATM, and x.25. The protocols of this layer are IEEE 802.3, IEEE 802.5, and IEEE 802.11, series of protocols (Aleksander 1985). TCP/IP OUTPUT FROM MATLAB/Simulink This simulation is used to show how simulink can communicate with remote applications using a simulink that has been developed. This block enables sending of data from the simulink model that has been developed into an application using TCP/IP. The block element is created with its base element being S-function block which uses a C MEX file. This simulation is performed through by writing the S-function source code and placing its name in an S-function block. A MEX S-function consists of a set of callback methods that will be invoked by the simulink engines to perform the tasks defined in the block. This function are implemented in c, c++, ada or fortan. The simulation Process writing the S-function source code and saving it in the s-function block. The s-function is composed of sets of callback methods that the simulink engine will invoke. The execution of the simulink model that proceeds in several stages. The stages include, initialization, simulation loop which is done on the several blocks until the simulation is complete. Comparing the output of the simulink against the expected results. Correction of errors if there is false output. This simulation follows the process of a TCP/IP server client streaming that has the following format. Server initialize the sockets create a socket. Bind the socket. Listen on the socket for availability of clients authenticate and accept a connection with a client receive and send data Disconnect the connection. Client initialize the socket create a socket connect to the server send and receive data disconnect from the server In this simulation the developed simulink block we created with the source code acts as a client. Source code This is the source code for: tcpipout.cpp #define S_FUNCTION_NAME tcpipout #define S_FUNCTION_LEVEL 2 #include "simstruc.h" #include "winsock.h" ssSetNumSFcnParams(S, 3); if(!ssSetNumInputPOrts(S, 1)) return; ssSetInputPortWidth(S, O, DYNAMICALLY_SIZED); ssSetOptions(S, SS_OPTION_WORKS_WITH_CODE_REUSE SS_OPTION_EXCEPTION_FREE_CODE SS_OPTION_USE_TLC_WITH_ACCELERATOR); ssSetSampleTime(S,0, mxGetScalar(ssGetSFcnParam(S, 2))); WSAStartup(wVersionRequested, &data); mxGetString(ssGetSFcnParam(S, 0),buf,buflen); host = gethostbyname(buf); port = (int) mxGetScalar(ssGetSFcnParam(S, 1)); mySocket = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP); serverSock.sin_family = AF_INET; serverSock.sin_port = htons(port); memcpy(&(serverSock.sin_addr), host->h_length); host->h_addr, connect(mySocket, (sockaddr *)&serverSock, si- zeof(serverSock); send(mySocket, data, strlen(data), 0); closesocket(mySocket); WSACleanup(); This is the output of the of the simulation DIGITAL COMMUNICATION Introduction Data communication that is sometimes known as digital transmission or digital communication is defined as the physical transfer of finite set of messages from one point to point or point to multiple points in a communication channel. The main objective of digital communication is to determine from a noise-petrubed signal which of the finite set of waveforms have been sent by the transmitter. This channels include optical fiber, copper wires, wireless communication channels, computer buses and storage media. The data that is being transmitted is represented in many formats that include electromagnetic signal such as electrical voltage, microwave, radio-wave, and infrared signal. Messages in digital transmission is represented in two formats which are either as a sequence of pulses by means of a line code which is base-band transmission or by a limited set of continuous varying wave forms which is called pass-band transmission using a digital modulation method. Modulation and demodulation is done by a modem (R.e 1990). Why not Analog Analog systems have existed for a long time and will continue to exist but most communication of this age is being digitized this is because: fidelity The energy, that is the power that is used in transmission is low and the devices consume low power. There is bandwidth efficiency that has been achieved through coding chains. Moores Law is decreasing device costs for digital hardware. There is a continued need for digital information. People need more powerful digital information security. Channel and Media A Channel is a modeled as a linear filter with the addition of noise. Noise comes from a variety of sources. Thermal background noise that is due to the physics of living above 0 Kelvin. The other source of noise is interference from other transmitted signals. This are signals that cannot be completely canceled and they may result in non-Gaussian noise distribution or non-white noise spectral density. The properties of a media cannot be changed and examples of media include: - Acoustic ultrasound -EM Spectra that is anything above 0 Hz and include Radio-wave, light, mm-wave bands. - waveguides, transmission lines, fiber optic, coaxial cable, wire cables. - Disk that are used as data storage applications. The linear filtering of channel results from the EM and the physics of the medium. For example attenuation in telephone wire vary by frequency. Narrow-band wireless channels experience fading which also varies quickly a function of the frequency. Wide-band wireless channels are multipath, due to multiple time-delayed reflections, diffraction’s, and scattering of the signal off of the objects in the channeling environment. This filters may be constant or time invariant (haykin 1996). Signal Processing Steps Information source comes from the higher networking layers. This information may be continuous or may be in small packets. Source encoding This is the process of finding a compact digital representation of the data source. This process include sampling of continuous time signals, compression of sources and also quantization of continuous valued signals. Channel encoding This refers to redundancy added to the signals so that the bit errors can be corrected and this is done by a channel decoder. Modulation This is the process of converting data from digital to analog and this produces a continuous time signal that can be sent on the physical channel. Proper matching of a modulation allows for the optimum transfer of information. Demodulation is the process of analog to digital conversion and this is done by a modem. Some digital modulation formats include: - phase shift keying (PSK). -frequency shift keying (FSK). -amplitude shift keying (ASK). -hybrid combination of ASK and PSK that is called quadrature amplitude modulation (QAM). Multiplexing and multiple access This process combines signals that might have different characteristics or signals that originate from different sources. Spreading This process renders the signals less vulnerable to interference that is both natural and intentional and is used to afford privacy to the communicators. Encryption Encryption bars unauthorized clients from knowing messages and completely seals false messages into the system Synchronization Synchronization has different form that include carrier, symbol, frame and network synchronization. Advantages and disadvantages of digital communication systems Advantages They are reliable since they are less sensitive to changes in environment like temperature, humidity. Easy multiplexing is achieved in digital communication. There is easy processing of data like encryption decryption and compression. There is voice and data integration. easy signaling in terms of address digits, hook status, call progress information. The systems can be easily monitored to determine its performance for example quality of service monitoring. Integration of switching and transmission. Signal regeneration and operation at low SNR can be achieved leading to superior performance. There is integration of services leading to Integrated Services Digital Network (ISDN). Disadvantages There is need for precision timing and bit, character and frame synchronization. Increased band-with is needed, 64KB for a4KHz channel, without compression but is less with compression. There is higher complexity Analog to digital and digital to analog conversion and non-linear conversion is often used and this leads to performance degradation. Pass band Quadrature Amplitude Modulation Simulation Pass band quadrature amplitude modulation signaling uses wave forms that have different amplitudes and phases depending on the kind of data that is being carried, this therefore can be viewed as an amplitude-phase keying, which always combines phase and amplitude modulation (haykin 1996). The objective of this simulation is to simulate pass band M = 2 b = 2 4 -ary QAM signaling depicted in figure 2 below and plot the bit error probability versus SNRdB r , b = 10 log 10 ( E b /( N 0 / 2)) for checking the validity of the theoretical derivation results found below. Theoretical Derivations Results P e , s ( M = L N ) = 1 − P ( probability of correct detection ) = 1 − (1 − P e , s ( L ))(1 − P e , s ( N )) P e, b = 1/b Pe, s Source code PercentagepassbandQAM.m Percentage simulates a digital communication system in Figure 3 Clear, clf b=4; M=2^b; L=2^ (b/2); % # of bits per symbol and the modulation order SNRbdBt=0:0.1:15; SNRbt=10. ^ (SNRbdBt/10); Pm=2*(1-1/L)*Q (sqrt (3/2*b*SNRbt/ (M-1))); Pobet= (1-(1-Pm). ^2)/b; Tb=1; Ts=b*Tb; % Bit/Symbol time Nb=16; Ns=b*Nb; % # of sample times in Tb and Ts T=Ts/Ns; LB=4*Ns; LBN1=LB-Ns+1; % Sample time and Buffer size Ssc= [0 0; 0 1; 1 1; 1 0]; sss=ssc; WC=8*pi/Ts; wcT=WC*T; t= [0: Ns-1]*T; Su=sqrt (2/Ts)*[cos (WC*t); -sin (WC*t)]; suT=su*T; % Basis signals Sum= 0; % 16-QAM signal waveforms corresponding to rectangular constellation For i=1: L For l=1: L S (i, l, 1) =2*i-L-1; s (i, l, 2) =2*l-L-1; %In-phase/quadrature amplitude Sum= Sum +s (i, l, 1) ^2 +s (i, l, 2) ^2; ss(L*(l-1)+i,:)=[ssc(i,:) sss(l,:)]; sw(L*(l-1)+i,:)=s(i,l,1)*su(1,:)+s(i,l,2)*su(2,:); end end Eav=Esum/M, Es_av=2*(M-1)/3 % Average signal energy (A=1) Es=2; % Energy of signal waveform A=sqrt(Es/Eav); sw=A*sw; levels=A*[-(L-1):2:L-1]; SNRdBs=[1:15]; MaxIter=10000; % Range of SNRbdB and # of iterations For iter=1: length (SNRdBs) SNRbdB= SNRdBs (iter); SNR=10^ (SNRbdB/10); sigma2= (Es/b)/SNR; sgmsT=sqrt (sigma2/T); Yr= zeros (2, LB); nobe= 0; % Number of bit errors to be accumulated For k=1: MaxIter Im= ceil (rand*L); in= ceil (rand*L); Imn= (in-1)*L+im; % Index of signal to transmit s=ss (imn, :); % Data bits to transmit For n=1: Ns Percentage Operation per symbol time wct= wcT*(n-1); bp_noise= randn*cos(wct)-randn*sin(wct); rn= sw(imn,n) + sgmsT*bp_noise; yr= [yr(:,2:LB) suT(:,n)*rn]; % Multiplier end ycsk=sum(yr(:,LBN1:LB)); % Sampled correlator output - DTR input %Detector(DTR) [dmin_i,mi]= min(abs(ycsk(1)-levels)); [dmin_l,ml]= min(abs(ycsk(2)-levels)); d= ss((ml-1)*L+mi,:); % Detected data bits nobe = nobe+sum(s~=d); if nobe>100; break; end end pobe(iter)= nobe/(k*b); end subplot(222), semilogy(SNRbdBt, pobet, k-, SNRdBs, pobe, b*) title(Probability of Bit Error for 16-ary QAM Signaling) The results of the simulation which is the plot of bit error probability versus SNRdB r , b = 10 log 10 ( E b /( N 0 / 2)). References AHITUV, Niv (1987). A Metamodel of Information Flow: A Tool to Support Information Systems Theory. Communications of the ACM, 30 (9), 781--791. ALBITZ, Paul, and Liu, Cricket (1992). DNS and BIND. vol.2. 2 ed., Havard, havard university press. albitz, 1. ALEKSANDER, Igor (1985). Advanced Digital Information Systems. vol.3. 4 ed., New Jersey, wiley and sons inc. igor, 5. ANDRE, F. et al (1986). Distributed Computing Systems. vol.2. 4 ed., Los angeles, Pearson. Distributed Computing Systems, 1. GRAHAM, Ian (1995). HTML Sourcebook. vol.1. 2 ed., New York, John Wiley & Sons. graham, 1. HALL, Devra (1995). Build A Web Site. vol.1. 2 ed., London, Prima Publishing. prima, 2. HAYKIN (1996). adaptive filter theory. vol.2. 5 ed., London, prentice hall. haykin, 2. HUDSON KURT AND STEWART MICHAEL. (1998). TCP/IP. vol.2. 2 ed., CHICAGO, the coriolosis group. coriolis group, 1. HUNT, Craig (1992). TCP/IP Network Administration. vol.1. 2 ed., washington, OReilly & Associates. OReilly & Associates, 2. KANELLAKIS, Paris (2007). Elemements of a relational database. technical report, london, Brown university (CS-89-39). LIU (1994). Managing Internet Information Services. vol.3. 3 ed., illinois, OReilly & Associates. oreilly, 4. MINASI (1995). Mastering Windows NT Server 3.5. vol.1. 2 ed., new hermshire, Sybex. 1, 2. P.J., Nelson (2010). Mastering Communication. Mastering Roadblocks, 12-20. R.E, Blahut (1990). Digital transmission of information. vol.4. 3 ed., new york, addison-wesley. wesley, 2. STAJANO, Frank (2000). Gentle introduction to relational and object oriented databases. ORL technical report, chicago, brookline university (Tr-98-2). Read More