Frame And The Layer 2 Sequence Numbers example essay topic

What is ISDN? ISDN, which stands for integrated services digital network, is a system of digitizing phone networks which has been in the works for over a decade. This system allows audio, video, and text data to be transmitted simultaneously across the world using end-to-end digital connectivity. The original telephone system used analog signals to transmit a signal across telephone wires. The voice was carried by modulating an electric current with a waveform from a microphone. The receiving end would then vibrate a speaker coil for the sound to travel back to the ear through the air.

Most telephones today still use this method. Computers, however, are digital machines. All information stored on them is represented by a bit, representing a zero or a one. Multiple bits are used to represent characters, which then can represent words, numbers, programs, etc. The analog signals are just varying voltages sent across the wires over time. Digital signals are represented and transmitted by pulses with limited number of discrete voltage levels.

[Hopkins] The modem was certainly a big breakthrough in computer technology. It allowed computers to communicate with each other by converting their digital communications into an analog format to travel through the public phone network. However, there is a limit to the amount of information that a common analog telephone line can hold. Currently, it is about 28.8 kb it / 's. [Hopkins] ISDN allows multiple digital channels to be operated simultaneously through the same regular phone jack in a home or office. The change comes about when the telephone company's switches are upgraded to handle digital calls.

Therefore, the same wiring can be used, but a different signal is transmitted across the line. [Hopkins] Previously, it was necessary to have a phone line for each device you wished to use simultaneously. For example, one line each for the phone, fax, computer, and live video conference. Transferring a file to someone while talking on the phone, and seeing their live picture on a video screen would require several expensive phone lines.

[Griffiths] Using multiplexing (a method of combining separate data signals together on one channel such that they may be decoded again at the destination), it is possible to combine many different digital data sources and have the information routed to the proper destination. Since the line is digital, it is easier to keep the noise and interference out while combining these signals. [Griffiths] ISDN technically refers to a specific set of services provided through a limited and standardized set of interfaces. This architecture provides a number of integrated services currently provided by separate networks. ISDN adds capabilities not found in standard phone service.

The main feature is that instead of the phone company sending a ring voltage signal to ring the bell in your phone, it sends a digital package that tells who is calling (if available), what type of call it is (data / voice ), and what number was dialed (if multiple numbers are used for a single line). ISDN phone equipment is then capable of making intelligent decisions on how to answer the call. In the case of a data call, baud rate and protocol information is also sent, making the connection instantaneous. [Griffiths] ISDN Concepts: With ISDN, voice and data are carried by bearer channels (B channels) occupying a bandwidth of 64 kb it /'s each.

A delta channel (D channel) handles signal ling at 16 kb it /'s or 64 kb it / 's. H channels are provided for user information at higher bit rates. [Stallings] There are two types of ISDN service: Basic Rate ISDN (BRI) and Primary Rate ISDN (PRI). BRI: consists of two 64 kb it /'s B channels and one 16 kb it /'s D channel for a total of 144 kb it / 's. The basic service is intended to meet the needs of most individual users.

PRI: intended for users with greater capacity requirements. Typically the channel structure is 23 B channels plus one 64 kb it /'s D channel for a total of 1.544 Mbit / 's. [Stallings] In this paper, I will concentrate on defining the specifics of Basic Rate ISDN for local loop transmission. I will provide an in depth view of ISDN as it relates to layer 1 to 3 of the seven layer OSI model.

I will also provide the specification for communication at the S / T customer interface. Basic Rate ISDN: Basic Rate Interface (BRI) - The BRI is the fundamental building block of an ISDN network. It is composed of a single 16 kb it /'s 'D-channel' which is used for call setup and control and two 64 kb it /'s 'B-channels'. The B-channels can be used to carry voice and both circuit mode and packet mode data traffic. The D-channel may also be used to carry X. 25 packet traffic if the network supports that option. [Griffiths] Basic Rate Interface D Channel - In the analog world, a telephone call is controlled in-band.

Tones and voltages are sent across lines for. ISDN does away with this. The D channel becomes the vehicle. This signal ling is called common channel since a separate channel for signal ling is used by two or more bearer channels.

[Hopkins] User - Network protocols define how users interact with ISDN networks. Between the user equipment and network equipment is a set of defined interfaces. The Interface is between the central office and the customer premise. This interface carries information on the twisted pair of wires between the customer and the central office. At the S / T interface located at the customer location, two pairs of wires (one for transmitting, one for receiving) are used. The intermediate device between the U and the S / T interface is known as an NT 1.

Then 1 is a hybrid that converts from four wire to two wire and also transforms the 2 B+D signal into a different bit stream format. [Griffiths] ISDN and the OSI Model - The OSI (Open Systems Interconnect) seven layer protocol was developed to promote interoperability in the data world. ISDN, which followed OSI, was designed to be a network technology inhabiting the lower three layers of the OSI model. Consequently, an OSI end system that implements an OSI seven layer stack can contain ISDN at the lower layers. Also, services such as TCP / IP (Internet Transmission Control Protocol) can use the ISDN network.

[Griffiths] Layer 1 of User-Network Interface: Layer 1 protocols provide the details that describe how the signals (electrical or optical) are encoded onto the physical medium. These protocols describe how the user data and signal ling bits are transformed into line signals, then back again into user data bits. The ISDN layer 1 protocol supports the functions outlined below. [ITU-T, I. 430] (B Channel Transmission (D Channel Transmission (D Channel Access Procedure Channel Transmission - Layer 1 must support for each direction of transmission, two independent 64 kb it /'s B channels.

The B channels contain user data which is switched by the network to provide the end-to-end transmission source. There is no error correction provided by the network on these channels. [ITU-T, I. 430] D Channel Transmission - Layer 1 must support for each direction of transmission, a 16 kb it /'s channel for the signal ling information. In some networks user packet data may also be supported on the D channel. [ITU-T, I. 430] D Channel Access Procedure - This procedure ensures that in the case of two or more terminals, on a point to multi point configuration, attempting to access the Channel simultaneously, one terminal will always successfully complete the transmission of information.

[ITU-T, I. 430] Binary Organization of Layer 1 frame - The structures of Layer 1 frames across the interface are different in each direction of transmission. Both structures are shown in figure 1 below. [Griffiths] A frame is 48 bits long and lasts 250 (s. The bit rate is therefore 192 kb it / sand each bit is approximately 5.2 (s long.

Figure 1 also shows that there is a 2-bit offset between transmit and receive frames. This is the delay between frame start at the receiver of a terminal and the frame start of the transmitted signal. [Griffiths] Figure 1 also illustrates that the line coding used is AMI (Alternate Marks Inversion); a logical 1 is transmitted as zero volts and a logical 0 as a positive or negative pulse. Note that this convention is the inverse of that used on line transmission systems.

The nominal pulse amplitude is 750 mV. [Griffiths] A frame contains several L bits. These are balance bits to prevent a build up of DC on the line. For the direction TE to NT, where each-channel may come from a different terminal, each terminal's output contains anL bit to form a balanced block.

[ITU-T, I. 430] Examining the frame in the NT to TE direction, the first bits of the frame are the F / L pair, which is used in the frame alignment procedure. The start of a new frame is signalled by the F / Pair violating the AMI rules. Once a violation has occurred there must be a second violation to restore correct polarity before the next frame. This takes place with the first mark after the F / L pair. The FA bit ensures this second violation occurs should there not be a mark in the B 1, B 2, D, E, or A channels.

The E channel is an echo channel in which D-channel bits arriving at the NT a reechoed back to the TEs. There is a 10 bit offset between the D channel leaving a terminal, traveling to the NT and being echoed back in the E channel. [ITU-T, I. 430] The A bit is used in the activation procedure to indicate to the terminals that the system is in synchronization. Next is a byte of the B 2 channel, a bit of the E channel and a bit of the D channel, followed by an M bit. This is used for multi framing. The M bit identifies some FA bits which can be stolen to provide a management channel.

[ITU-T, I. 430] The B 1, B 2, D, and Channels are then repeated along with the S bit which is a spare bit. [ITU-T, I. 430] Layer 1 D Channel Contention Procedure - This procedure ensures that, even in the case of two or more terminals attempting to access the D channel simultaneously, one terminal will always successfully complete the transmission of information by first gaining control of the D channel and then retransmitting its information. The procedure relies on the fact that the information to be transmitted consists of layer 2 frames delimited by flags consisting of the binary pattern 01111110. Layer 2 applies a zero bit insertion algorithm to prevent flag imitation by a layer 2 frame. The inter frame time fill consists of binary 1's which are represented by zero volts. The zero volt line signal is generated by the TE transmitter going high impedance.

This means a binary 0 from a parallel terminal will overwrite as binary 1. Detection of collision is done by the terminal monitoring the E channel (D channel echoed from the NT). [ITU-T, I. 430] To access the D channel a terminal looks for the inter frame time fill by counting the number of consecutive binary 1's in the D channel. Should a binary 0 be received the count is reset.

When the number of consecutive 1's reaches a predetermined value (which is greater than the number of consecutive 1's possible in a frame because of the zero bit insertion algorithm) the counter is reset and the terminal may access the D channel. When a terminal has just completed transmitting a frame the value of the count needed to be reached before another frame may be transmitted is incremented by 1. This gives other terminals a chance to access the channel. Hence an access and priority mechanism is established. [ITU-T, I. 430] There is still the possibility of collision between two terminals of the same priority. This is detected and resolved by each terminal comparing its last transmitted bit with the next E bit.

If they are the same the terminal continues to transmit. If, however, they are different the terminal detecting the difference ceases transmission immediately and returns to the D channel monitoring state leaving the other terminal to continue transmission. [ITU-T, I. 430] Layer 1 Activation / Deactivation Procedure - This procedure permits activation of the interface from both the terminal and network side, but deactivation only from the network side. This is because of the multi-terminal capability of the interface. Activation and deactivation information is conveyed across the interface by the use of line signals called " Info signals'. [ITU-T, I. 430] Info 0 is the absence of any line signal; this is the idle state with neither terminals nor the NT working.

[ITU-T, I. 430] Info 1 is flags transmitted from a terminal to the NT to request activation. Note this signal is not synchronized to the network. [ITU-T, I. 430] Info 2 is transmitted from the NT to the TEs to request their activation or to indicate that the NTh as activated as a response to receiving an Info 1. An Info 2 consists of Layer 1 frames with a high density of binary zeros in the data channels which permits fast synchronization of the terminals. [ITU-T, I. 430] Info 2 and Info 4 are frames containing operational data transmitted from the TE and NT respectively. [ITU-T, I. 430] The principal activation sequence is commenced when a terminal transmits an Info 1.

The NT activates the local transmission system which indicates to the exchange that the customer is activating. The NT 1 responds to the terminals with an Info 2 to which the TEs synchronize. The TEs respond with an Info 3 containing operational data and the NT is then in a position to send Info 4 frames. Note that all terminals activate in parallel; it is not possible to have just one terminal activated in a multi-terminal configuration. The network activates the bus by the exchange activating the local network transmission system. Deactivation occurs when the exchange deactivates the local network transmission system.

[ITU-T, I. 430] Layer 2 of User-Network Interface: The Layer 2 recommendation describes the high level data link (HDLC) procedures commonly referred to as the Link Access Procedure for a D channel or LAP D. The objective of Layer 2 is to provide a secure, error-free connection between two endpoints connected by a physical medium. Layer 3 call control information is carried in the information elements of Layer 2 frames and it must be delivered in sequence and without error. Layer 2 also has the responsibility for detecting and retransmitting lost frames. LAP D was based originally on LAP B of the X. 25 Layer 2 recommendation. However, certain features of LAP D give it significant advantages.

The most striking difference is the possibility of frame multiplexing by having separate addresses at Layer 2 allowing many LAPs to exist on the same physical connection. It is this feature that allows up to eight terminals to share the signal ling channel in the passive bus arrangement. [ITU-T, Q. 920] Each Layer 2 connection is a separate LAP and the termination points for the LAPs are within the terminals atone end and at the periphery of the exchange at the other. Layer 2 operates asa series of frame exchanges between the two communicating, or peer entities. The frames consist of a sequence of eight bit elements and the elements in the sequence define their meaning as shown in Figure 2 below.

[ITU-T, Q. 920] A fixed pattern called a flag is used to indicate both the beginning and end of frame. Two octets are needed for the Layer 2 address and carry a service identifier (SAPI), a terminal identifier (TEI) and a command / response bit. The control field is one or two octets depending on the frame type and carries information that identifies the frame and the Layer 2 sequence numbers used for link control. The information element is only present in frames that carry Layer 3 information and the Frame Check Sequence (FCS) is used for error detection. A detailed breakdown of the individual elements is given in Figures 3 and 4 below. [ITU-T, Q. 920] What cannot be shown in the diagrams is the procedure to avoid imitation of the flag by the data octets.

This is achieved by examining the serial stream between flags and inserting an extra 0 after any run of five 1 bits. The receiving Layer 2 entity discards a 0 bit if it is preceded by five 1's. [ITU-T, Q. 920] Layer 2 Addressing - Layer 2 multiplexing is achieved by employing a separate Layer 2 address for each LAP in the system. To carry the LAP identity the address is two octets long and identifies the intended receiver of a command frame and the transmitter of a response frame. The address has only local significance and is known only to the two end-points using the LAP. No use can be made of the address by the network for routing purposes and no information about its value will be held outside the Layer 2 entity.

[ITU-T, Q. 921] The Layer 2 address is constructed as shown in Figure 3. The Service Access Identifier (SAPI) is used to identify the service intended for the. An extension of the use of the D channel is to use it for access to a packet service as well as for signal ling. Consider the case of digital telephones sharing a passive bus with packet terminals.

The two terminal types will be accessing different services and possibly different networks. It is possible to identify the service being invoked by using a different SAPI for each service. This gives the network the option of handling the with different services in separate modules. In a multi-network ISDN it allows Layer 2 routing to the appropriate network. The value of the SAPI is fixed for a given service.

[ITU-T, Q. 921] The Terminal Endpoint Identifier (TEI) takes a range of values that are associated with terminals on the customer's line. In the simplest case each terminal will have a single unique TEI value. The combination of TEI and SAPI identify the LAP and provide a unique Layer 2 address. A terminal will use its Layer 2 address in all transmitted frames and only frames received carrying the correct address will be processed. [ITU-T, Q. 921] In practice a frame originating from telephony call control has a SAPI that identifies the frame as 'telephony' and all telephone equipment examine this frame. Only the terminal whose TEI agrees with that carried by the frame will pass it to the Layer 2 and Layer 3 entities for processing.

There is also a SAPI identified in standards for user data packet communication. [ITU-T, Q. 921] Since it is important that no two TEIs are the same, the network has a special TEI management entity which allocates TEI on request and ensures their correct use. The values that TEIs can take fall into the ranges: 0-63 Non-Automatic Assignment TEIs 64-126 Automatic Assignment TEIs 127 Global TEI [ITU-T, Q. 921] Non-Automatic TEIs are selected by the user; their allocation is the responsibility of the user. Automatic TEIs are selected by the network; their allocation is the responsibility of the network. The global TEI is permanently allocated and is referred to as the broadcast TEI.

[ITU-T, Q. 921] Terminals which use TEIs in the range of 0-63 need not negotiate with the network before establishing a Layer 2 connection. Terminals which use TEIs in the range 64-126 cannot establish a Layer 2 connection until they have requested a TEI from the network. In this case it is the responsibility of the network not to allocate the same TEI more than once at any given time. The global TEI is used to broadcast information to all terminals within a given SAPI; for example a broadcast message to all telephones, offering an incoming telephone call. [ITU-T, Q. 921] Layer 2 Operation - The function of Layer 2 is to deliver Layer 3 frames, across a Layer 1 interface, error free and in sequence. It is necessary for a Layer 2 entity to interface both Layer 1 and Layer 3.

To highlight the operation of Layer 2 we will consider the operation of a terminal as it attempts to signal with the network. [ITU-T, Q. 921] It is the action to establish a call that causes protocol exchange between terminal and network. If there has been no previous communication it is necessary to activate the interface in a controlled way. A request for service from the customer results in Layer 3 requesting a service from Layer 2.

Layer 2 cannot offer a service unless Layer 1 is available and so a request is made to Layer 1. Layer 1 then initiates its start-up procedure and the physical link becomes available for Layer 2 frames. Before Layer 2 is ready to offer its services to Layer 3 it must initiate the Layer 2 start-up procedure known as " establishing a LAP'. [ITU-T, Q. 921] LAP establishment is achieved by the exchange of Layer 2 frames between the Layer 2 handler in the terminal and the corresponding Layer 2 handler in the network. The purpose of this exchange is to align the state variables that will be used to ensure the correct sequencing of information frames. Before the LAP has been established the only frames that may be transmitted are unnumbered frames.

The establishment procedure requires one end-point to transmit a Set Asynchronous Balanced Mode Extended (SABME) and the far end to acknowledge it with an Unnumbered Acknowledgment (UA). [ITU-T, Q. 921] Once the LAP is established Layer 2 is able to carry the Layer 3 information and is said to be the 'multiple frame established state'. In this state Layer 2 operates its frame protection mechanisms. Figure 5 below shows a normal Layer 2 frame exchange. [ITU-T, Q. 921] Once established the LAP operates an acknowledged service in which every information frame must be responded to by the peer entity. The most basic response is the Receiver Ready (RR) response frame.

Figure 5 shows the LAP establishment and the subsequent I frame RR exchanges. The number of I frames allowed to be outstanding without an acknowledgment is defined as the window size and can vary between 1 and 127. For telephony signal ling applications the window size is 1 and after transmitting an I frame the Layer 2 entity will await a response from the corresponding peer entity before attempting to transmit the next I frame. Providing there are no errors all that would be observed on the bus would be the exchange of I frames and RR responses. However Layer 2 is able to maintain the correct flow of information in the face of many different error types.

[ITU-T, Q. 921] Layer 2 Error Control - It is unlikely that a frame will disappear completely but it is possible for frames to be corrupted by noise at Layer 1. Corrupted frames will be received with invalid Frame Check Sequence (FCS) values and consequently discarded. [ITU-T, Q. 920] The frame check sequence is generated by dividing the bit sequence starting at the address up to (but not including) the start of the frame check sequence by the generator polynomial X 16 + X 12 + X 5 + 1. In practical terms this is done by shift register as shown in figure 6.

All registers are preset to 1 initially. At the end of the protected bits the shift register contains the remainder from the division. The 1's complement of the remainder is the FCS. At the receiver the same process is gone through, but this time the FCS is included in the division process.

In the absence of transmission errors the remainder should always be 1101 0000 1111. [ITU-T, Q. 920] The method for recovering from a lost frame is based on the expiration of a timer. A timer is started every time a command frame is transmitted and is stopped when the appropriate response is received. This single timer is thus able to protect both the command and response as the loss of either will cause it to expire.

[ITU-T, Q. 920] When the timer expires it is not possible to tell which of the two frames has been lost and the action taken is the same in both cases. Upon the timer expiring, Layer 2 transmits a command with the poll bit set. This frame forces the peer to transmit a response that indicates the value held by the state variables. It is possible to tell from the value carried by the response frame whether or not the original frame was received. If the first frame was received, the solicited response frame will be the same as the lost response frame and is an acceptable acknowledgment.

If however the original frame was lost, the solicited response will not be an appropriate acknowledgment and the Layer 2 entity will know that a retransmission is required. It is possible for the same frame to be lost more than once and Layer 2 the frame three times. If after three transmissions of the frame the correct response has not been received, Layer 2 will assume that the connection has failed and will attempt to re-establish the LAP. [ITU-T, Q. 921] Another possible protocol error is the arrival of an I frame with an invalids end sequence number N (S).

This error is more likely to occur when the LAP is operating with a window size greater than one. If, for example, the third frame in the sequence of four is lost the receiving Layer 2 entity will know that a frame has been lost from the discontinuity in the sequence numbers. The Layer 2 must not acknowledge the fourth frame as this will imply acknowledgment of the lost third frame. The correction operation is to send a Reject (REJ) frame with the receive sequence number N (R) equal to N (S) + 1 where N (S) is the send variable of the last correctly received I frame, in this case I frame 2.

This does two things; first it acknowledges all the outstanding I frames up to and including the second I frame, and secondly it causes the sending end to retransmit all outstanding I frames starting with the lost third frame. [ITU-T, Q. 920] The receipt of a frame with an out of sequence, or invalid, N (R) does not indicate a frame loss and cannot be corrected by retransmissions. It is necessary in this case to re-establish the LAP to realign the state variables a teach end of the link. [ITU-T, Q. 920] The Receiver Not Ready (RNR) frame is used to inhibit the peer Layer 2 from transmitting I frames. The reasons for wanting to do this are not detailed in the specification but it is possible to imagine a situation where Layer 3 is only one of many functions to be serviced by a microprocessor and a job of higher priority requires that no Layer 3 processing is performed. [ITU-T, Q. 920] Another frame specified in Layer 2 is the FRaMe Reject frame (FRM R).

This frame may be received by a Layer 2 entity but may not be transmitted. It is included in the recommendation to preserve alignment between LAP D and LAP B. After the detection of a frame reject condition the data link is reset. [ITU-T, Q. 920] Disconnecting the LAP - After Layer 3 has released the call it informs Layer 2 that it no longer requires a service. Layer 2 then performs its own disconnection procedures so that ultimately Layer 1 can disconnect and the transmission systems associated with the local line and the customer's bus can be deactivated. [ITU-T, Q. 921] Layer 2 disconnection is achieved when the frames disconnect (DISC) and UA are exchanged between peers. At this point the LAP can no longer support the exchange of I frames and supervisory frames.

[ITU-T, Q. 921] The last frame type to be considered is the Disconnect Mode (DM) frame. This frame is an unnumbered acknowledgment and may be used in the same way as a UA frame. It is used as a response to a SABME if the Layer 2 entity is unable to establish the LAP, and a response to a DISC if the Layer 2 entity has already disconnected the LAP. [ITU-T, Q. 921] TEI Allocation - Because each terminal must operate using a unique TEI, procedures have been defined in a Layer 2 management entity to control their use. The TEI manager has the ability to allocate, remove, check, and verify TEIs that are in use on the customer's bus. As the management entity is a separate service point all messages associated with TEI management are transmitted with a management SAPI.

[ITU-T, Q. 921] TEI management procedures must operate regardless of the Layer 2 state and so the unnumbered information frame (UI) is used for all management messages. The UI frames have no Layer 2 response and protection of the frame content is achieved by multiple transmissions of the frame. In order to communicate with terminals which have not yet been allocated TEIs a global TEI is used. All management frames are transmitted on a broadcast TEI which is associated with a LAP that is always available.

All terminals can transmit and receive on the broadcast TEI as well as their own unique TEI. All terminals on the customer's line will process all management frames. To ensure that only one terminal acts upon a frame a unique reference number is passed between the terminal and the network. This reference number is contained within an element in the UI frame and is either a number randomly generated by the terminal, or 0 is the TEI of the terminal, depending on the exact situation. Figure 7 below shows the frame exchange required for a terminal to be allocated a TEI and establish its data link connection. [ITU-T, Q. 921] Layer 3 of User-Network Interface: This layer effects the establishment and control of connections.

It is carried in Layer 2 frames as can be seen in figure 8. [ITU-T, Q. 930] The first octet contains a protocol discriminator which gives the D channel the capability of simultaneously supporting additional communications protocols in the future. The bits shown in figure 8 are the standard for user-network call control messages. [ITU-T, Q. 930] The call reference value in the third octet is used to identify the call with which a particular message is associated. Thus a call can be identified independently of the communications channel on which its supported. The message type coded in the fourth octet describes the intention of the message (e.g. a SETUP message to request call establishment).

These are listed in Table 1 at the end of this paper. A number of other information elements maybe included following the message type code in the fourth octet. The exact contents of a message are dependent on the message type. [ITU-T, Q. 931] The message sequence for call establishment is shown in figure 9. In order to make an outgoing call request, a user must send all of the necessary call information to the network. Furthermore, the user must specify the particular bearer service required for the call (i.e. Speech, 64 kb it /'s /s unrestricted, or 3.1 k Hz Audio) and any terminal compatibility information which must be checked at the destination.

[ITU-T, Q. 931] The initial outgoing call request may be made in an en bloc or overlap manner. Figure 9 illustrates the call establishment procedures. If overlap sending is used then the SETUP message must contain the bearer service request but the facility requests and called party number information may be segmented and conveyed in a sequence of INFORMATION messages as shown. Furthermore if a speech bearer service is requested and no call information is contained in the SETUP message, then the network will return in-band dial tone to the user until the first INFORMATION message has been received. [ITU-T, Q. 931] Following the receipt of sufficient information for call establishment, the network returns a call PROCEEDING.