SONET

• A two dimensional frame structure or a Rectangular structure
• Consists of 810 bytes
• It is transmitted at 125 micro second intervals
• Three major parts:- Section, Line and Path
• Two fixed stuff Columns:- 30-th and 59-th.

• Section overhead is allocated to the first three rows of the transport overhead.  Subsequent slides describe each section overhead byte.
• Line overhead is allocated to the remaining six rows of the transport overhead.  The functions are described in the following slides with the exception of the payload pointer, which has its own section later on.

• The hex code F628 is transmitted in every frame of every STS-1.
• This allows a receiver to locate the alignment of the 125usec frame within the received serial bit stream.  Initially, the receiver scans the serial stream for the code F628.  Once it is detected, the receiver watches to verify that the pattern repeats after exactly 810 STS-1 bytes, after which it can declare the frame found.
• Once the frame alignment is found, the remaining signal can be descrambled and the various overhead bytes extracted and processed.

• Just prior to transmission, the entire SONET signal, with the exception of the framing bytes and the section trace byte, is scrambled.  Scrambling randomizes the bit stream is order to provide sufficient 0-1 and 1-0 transitions for the receiver to derive a clock with which to receive the digital information.
• Scrambling is performed by XORing the data signal with a pseudo-random bit sequence generated by the scrambler polynomial indicated above.  The scrambler is frame synchronous, which means that it starts every frame in the same state.
• Descrambling is performed by the receiver by XORing the received signal with the same pseudo random bit sequence.  Note that since the scrambler is frame synchronous, the receiver must have found the frame alignment before the signal can be descrambled.  That is why the frame bytes are not scrambled.

• Section trace allows individual fibers to be distinguished from each other.  This is particularly useful when systems are initially installed.  Without section trace, it can be difficult to detect when the fibers have been incorrectly connected.
• In earlier releases of the SONET standard, this byte was used for an STS-1 identifier.  All it did was count the timeslots in an STS-N multiplex.  The first STS-1 would have hex 01 in this byte, the second 02 and so on.
• In order to detect whether this byte is carrying the old STS-1 ID or the new section trace, the section trace does not allow hex 01 as a valid trace ID.
• In SONET, the section trace is defined as a one byte code, allowing for up to 255 identifiers (excluding hex 01).
• In SDH, the section trace is defined as a 16 byte string that is transmitted repeatedly.

• The section BIP-8 byte contains an 8-bit bit interleaved parity.  Bit 1 is even parity calculated over all the bit 1s in every byte of the previous STS-N frame.  Bits 2 through 8 are calculated similarly.
• Since there is only one section BIP-8 regardless of the value of N, the ratio of parity bits to data bits becomes less at higher line rates.  This ratio affects the error detection capability of the mechanism. Therefore, the section BIP-8 is more useful for fault isolation that it is for performance monitoring.

• Section orderwire is a channel reserved for voice communications between network provider maintenance personnel.  Since it is in the section layer overhead, it is accessible at a regenerator, such that a maintenance person working on the regenerator can talk to a colleague at the terminal site to coordinate maintenance activity.

• The section user channel is dedicated to network provider use.  SONET standards do not define a particular use, protocol or interface for this byte. However, a typical application would be to backhaul alarms from a regenerator site  to the central office.

The pointer overhead bytes carry a binary number that is an offset, in bytes, into the envelope capacity to the beginning of the next SPE frame.

• The content of the STS payload pointer bytes is shown above. 
• Bits 7 and 8 of H1 together with bits 1-8 of H2 carry the pointer offset.  If the first byte of the next SPE frame was the byte immediately following the H3 byte, the offset would contain the value 00 hex.
• The bits of the pointer offset are grouped into I bits and D bits.  These bits are used to flag increment operations and decrement operations respectively. A pointer processor monitoring an incoming pointer keeps track of the current offset and continuously compares the next received value with the current value. An increment operation is detected when the I bits of the received value are inverted with respect to the current value.  Similarly, a decrement operation is detected when the D bits of the received value are inverted with respect to the current value.
• Bits 1-4 of H1 contain the new data flag.  This flag is used to notify downstream pointer processors of a change in the pointer value to a new value unrelated to the previous value.  This would be used at a cross-connect system for example, when an outgoing STS-1 that was carrying an SPE from one particular incoming STS-1 suddenly starts carrying an SPE from a different incoming STS-1 due to a provisioning command.  Since there is no phase relationship between the two SPEs, the pointer value will change.  The new data flag is set to 1001 in the first STS-1 frame that contains the new pointer offset value, and then returns to its normal value of 0110.
• When an STS-1 is concatenated to another STS-1, its pointer offset contains the value 1001001111111111.  This is the concatenation flag and it effectively disables the pointer processing circuitry for this STS-1.
• When a failure is detected at the path layer, path AIS is generated downstream to provide a keep-alive signal. Path AIS  is detected by all ones in the H1-2 pointer bytes.
• Data rate at NE1 > NE2   overflow ie negative justification H3 is used
• Data rate at NE1 < NE2   underflow ie +ve justification or stuffing and H3 is not used

• A line terminal detecting an incoming line failure will send line RDI (110 in bits 6-8 of K2)  back to the other end of the line so that both ends are aware that a failure has been detected on the line. 
• 1:3  one is redundant path and 3 are used path.

• The line bit interleaved parity is similar to the section BIP-8. However, each STS-1 has its own line BIP-8, and it is calculated over the entire STS-1 with the exception of the section overhead.
• Since there is a separate line BIP-8 for each STS-1, there is a constant ratio of parity bits to data bits regardless of line rate.  This means that the saturation point of the error checking mechanism is constant at about 10-4.  Therefore, the line BIP-8 is used for performance monitoring.

• In a synchronous network, all network elements receive a network clock that  is used to transmit all outgoing synchronous signals.  Often a network element receives its clock reference via an incoming synchronous signal.  A clock will be derived from the incoming signal and used to lock a local system clock.  In order to provide robust timing distribution, the network element will usually have a selection of references from which to choose in case the one that is being used fails.
• Synchronization messaging provides an indication of the quality of the SONET signal for use as a synchronization reference.  It allows a network element to evaluate all of its reference sources and choose the best one. 
• It also can be used to prevent the possibility of timing loops.

• Line REI  allows each end of a line system to evaluate the error performance of the signal that it is transmitting.
• Each end of the line system is checking the Line BIP-8 on a per frame basis on the received signal.  This check allows up to 8 x N errors to be detected in each frame of the STS-N.  The result of this check, that is the count of errors, is sent back to the transmitting end in the Line REI byte, so that the transmitting end can maintain the same error count as the receiving end.
• Thus, each end can provide performance monitoring statistics of the signal that is being received, using the Line BIP-8, as well as the signal that is being transmitted, using the Line REI.  This is referred to as single-ended performance monitoring.
• Note: Line REI previously known as Line FEBE (Far End Block Error).

• Line orderwire is similar to section orderwire, except that it is not accessible at regenerators, which, in general, do not have access to the line overhead.  Thus the line orderwire only allows communications between  maintenance personnel at line terminal sites.

• The SONET section and line data communications channels (DCC) are used to provide a communications path between a centralized operations system (OS) and the various network elements.
• The section DCC is carried in the D1, D2, and D3 bytes which combine to form a single 192Kb/s channel.  The line DCC is carried in the D4-D12 bytes which combine to form a single 576Kb/s channel.
• The OS connects to a gateway NE, possibly via an external data network.  From that point on, the messages are relayed from NE to NE  via the DCCs.  NEs are able to look at the address specified in received messages and decide whether they are for that NE.  If not, the NE knows how to route the message so that it reaches its ultimate destination.
• The overall data network created is called the Telecommunications Management Network (TMN).
• OSs monitor and control entire networks.  They provide operations, administration, maintenance and provisioning (OAM&P)  functions such as alarm surveillance and fault management, performance monitoring, configuration management, and connection set-up.
• Each network element is modeled using object-oriented techniques and the resulting information model is a standard.  Thus the OS views each NE as a set of objects that can be controlled and monitored, regardless of which vendor supplied the NE.  In general, all application layer messages in the TMN are from OS to NE and not directly from NE to NE.
• The section and line DCCs use the same protocol stack and were originally intended to carry the same messages.  The only distinction is that the line DCCs cannot be accessed at a regenerator.  The messages that the OS uses to operate on the NE objects are carried over this protocol.

• The first column of the STS synchronous payload envelope contains the STS path overhead.  STS path overhead allows for integrity verification of the end-to-end STS path.  In general, it is not modified at line terminations.

• Path trace provides a way of verifying that the path is being received is coming from the right place.  It does not provide any indication of the route that the path took; it only verifies the end points.
• The path trace byte carries a 64 byte string that repeats continuously. Each frame of the SPE contains one byte of the string, such that it takes 64 frames to transmit the entire string.
• The source of the path inserts a code in the path trace byte that uniquely identifies the source of the path.  Although this code is usually user provisionable, ANSI T1.105 recommends the Bellcore Common Language Location Identifier (CLLI) terminated with a carriage return (CR) and a line feed (LF) character.  The CLLI code is a string of ASCII characters that identify such things as the state, city, and office that contains the equipment.
• The termination of the path monitors the incoming path trace code and compares it to a locally provisioned value.  If the codes do not match, an alarm can be raised indicating a potential misconnection.
• In SDH, the path trace code is defined to be a 16 byte string, rather than 64.

• The path BIP-8 is the same as the line and section BIP-8, except that it is calculated only over the SPE frame.  Since the content of the SPE  is not modified at intermediate points along the path (line terminations), the path BIP-8 allows end-to-end bit error integrity verification of the SPE.

• The path signal label provides an identifier of the contents of the SPE, since a variety of payloads can be mapped into it.
• This can be used to identify problems where the wrong mapper has been plugged into one end of the path.  For example, if a DS3 mapper is plugged in at one end and a DS1 mapper at the other end, the connection will obviously not work and the signal label will indicate why.
• The signal label can also be used as a way of automatically provisioning mappers that can handle more than one payload type.

• The path status byte provides a way for each end of the path to monitor the integrity of both directions of transmission.  The two functions carried in the path status byte are the path remote error indication (REI) and path remote defect indication (RDI).
• The path REI  is similar to the line REI.   Each end of the path  is checking the path BIP-8 on a per frame basis on the received signal.  This check allows up to 8 errors to be detected in each frame of the SPE.  The result of this check, that is  the count of errors, is sent back to the transmitting end in the path REI, so that the transmitting end can maintain the same error count as the receiving end.
• The path RDI allows the receiving end of the path to notify the transmitting end when failures have been detected, as opposed to just bit errors.  The various error possibilities are classified into 3 groups:
•   payload defects indicate that although the SPE seems to be OK, the payload is not getting through properly,
• •  connectivity defects indicate a provisioning problem, and
• •  server defects indicates a transmission failure of some sort.
• These conditions have different effects on the calculation of performance statistics and unavailability, therefore each end must know the kind of failure in order to come up with the same performance monitoring results.

• The path user channel allows network providers to connect information of their choosing across an STS path layer connection.  SONET standards do not define any use, protocol, or interface for this byte; that is left for the individual network providers to specify.

• Since the VT superframe will span four consecutive STS-1 SPE frames, a method is required to identify which VT subframe is present in the current STS-1 SPE frame.  That is the purpose of the H4 STS path overhead byte, which contains a simple modulo 4 counter.  All of the VT superframes are aligned such that the H4 byte simultaneously identifies the alignment of all of them

• Tandem connection monitoring uses the Z5 byte from the STS path overhead. 
• The path BIP-8 (B3) of the paths that are to become part of the tandem connection is monitored as they arrive from upstream.  The incoming error count detected is inserted into bits 1-4 of the tandem connection byte on a per path basis.
• At the other end of the tandem connection, the path BIP-8s are monitored again and the results are compared with the values received in the IECs.  Any differences are considered to be errors that occurred on the tandem connection.
• Bits 5-8 carry LAPD structured messages that are used to send far-end performance reports from each end of the tandem connection back to the other end so that each end can verify the integrity of the both directions of the connection independently. 
• When paths are bundled together on a tandem connection, the LAPD message is sent only on one path.  Paths may be grouped only in bundle sizes that correspond to the supported SONET line rates.
• Note that the insertion of the tandem connection byte involves the modification of the SPE at an intermediate point on the path.  In order to avoid corrupting the path BIP-8, it must be compensated.  When the TCM byte is inserted or removed, the value of the byte that was in location Z5 is subtracted from the BIP-8 and the new value added.  This preserves the end-to-end path integrity monitoring capability of the BIP-8.

• Automatic protection switching refers to the ability of SONET network elements to detect a failure and transfer  the affected traffic to another line.
• The most basic protection system is a linear 1+1 system.  The term linear differentiates it from ring systems and the “1+1” indicates that there is 1 working fiber and 1 standby fiber and that the traffic in both directions is permanently bridged onto both the working and the standby fiber.
• When one end of the system detects that the side from which it is currently selecting the traffic fails, it automatically switches to select the traffic from the other side after first verifying that the switch will restore the traffic and that there are no other reasons why the switch cannot be made.
• This will result in a switch in one direction of the traffic; a unidirectional switch.  If the system is provisioned for bidirectional switching, the other direction of traffic must be switched as well.  However, the failure may not be such that the other end of the system can detect it.  In order to ensure that both ends switch, the tail end (the end that initially detected the failure) sends an indication in the APS bytes (K1, K2) that it has switched and why and that the head end should do the same.
• A linear 1+1 system can be made survivable by diversely routing the protection channel as is shown in the slide.  In this way, a fiber cut on the working channel can still be restored on the protection channel.

• A linear 1:N (“1 for N”) protection system has one protection fiber providing a restoration path for up to N working fibers.  Such a system is not survivable since it is impractical to diversely route all of the fibers.  Therefore, a fiber cut, even if it does not affect the protection channel, is liable to affect more than one working channel, and the protection channel can restore only one working channel at a time.
• A 1:N system involves more complicated signaling between the two ends of the system to coordinate the switch.  This is because the failed channel must first be bridged onto the protection channel before the receiving end can do the switch, and the end that must do the bridge may not even be directly aware  that there is a failure.

• The line layer APS bytes (K1, K2) are used to carry the APS protocol between the two ends of the system.  The K1 bytes on the protection fiber are used.
• The K1 byte acts as the switch request channel.  When one end of  the system detects a failure on a channel, it indicates the kind of failure and identifies the channel to the other end of the system using the K1 byte.  In addition, K1 is used to signal switch requests for reasons other than failures, such as manual and forced switches that are used by maintenance personnel.  The different request types are organized in a hierarchy or priority scheme.  If the node detects a signal degradation on one channel, but a signal failure on another, it will try to restore the channel affected by the signal failure because the signal failure has a greater impact on traffic.
• When the other end of the system receives the switch request, it looks at the priority of the request and compares it to local conditions.  If the far end is requesting a switch on one channel due to a manual switch, but it sees a signal failure on another channel on its end, it will ignore the manual switch request and signal back to the far end a switch request for the channel affected by the signal failure and the far end is obliged to comply.
• Even after a switch has been established, if a higher priority condition arises, the old switch will be dropped in favor of the new one.  This is referred to as switch preemption.
• K2 is used to signal which channel has been bridged onto the protection fiber.  It also includes some provisioning status related to APS. 
• When a node receiving a switch request agrees to the request, it will bridge the requested channel onto the protection fiber.  It will then signal back to the other end, using the K2 byte, that the channel has been bridged.  This will indicate to the other end that it is now safe to complete the switch by taking the traffic off of the protection fiber.  A node wanting to switch (that is,. select traffic from the protection fiber) will wait until K2 indicates that the traffic has been bridged so that it knows it is selecting the proper traffic.
• When the failure has been corrected, the node requesting the switch will drop its request by signaling no request in byte K1.  This will result in the bridges and switches being dropped.

• The K bytes for ring signaling are similar to those used for linear protection signaling. 
• The K1 byte is used as a switch request channel as in linear and there is a prioritized selection of switch requests available. However, there are separate requests for ring switches and span switches.  The span switch requests are only used in 4-fiber BLSRs.  Rather than indicate a channel number, which is not relevant in a ring, the K1 byte identifies a destination node.  Each node on a ring has an identifier and all nodes know that ID as well as the location of the node on the ring.  A node wishing to initiate a switch needs to be able to identify which node the signaling is destined for.
• The K2 byte carries the ID of the source of the signaling.  Since ring signaling is generally propagated in both directions around the ring, the K2 byte also indicates whether the received code is arriving on the short path (that is, across the span affected by the failure) or on the long path (that is, the long way around the ring).  Finally, the K2 byte also indicates the status of bridges and switches at the source node.

• The following introduces the layers bottoms up, that is, at the receiver of the end-to-end connection.  Each layer builds on the services provided by the lower layers.
•  The photonic layer provides optical transmission at some bit rate, such as OC1, OC12, or OC48...
• · The section layer verifies and obtains frame synchronization, and also descrambles the received bitstream
• · The line layer performs line maintenance and automatic protection as well as demultiplexes STS-1 signal into lower rate SONET signals
• · The path layer de-maps Sonet signals into various services, in this example, DS3.
• Note that all layers can be implemented at one time or they can be broken into separate steps.  A slightly more detailed descriptions of the major functions of each of the four optical interface layers is listed/described as follows:
• A.  Path Layer
• · It deals with the transport of services between path terminating equipment (PTE); examples of such services are DS1s, CEPT-1s, DS1Cs, DS2s, DS3s, video signals, and so on
• · Its main function is to map the services into a format required by the line layer; examples of those formats are STS-1, STS-3, STS-12, STS-48 and STS-192
• · It communicates end to end via the path over head
• · The overhead defined for this layer is read, interpreted, and modified by all equipment that terminates this layer; the network elements are Sonet Path Terminating Equipment (PTE)
• An example of system equipment that communicates at this level is DS3 to STS-1 mapping circuit.

• Total SPE= 87N
• No. of Columns of Fixed bytes = N/3-1
• Column of POH = 1
• Total number of unusable Columns = N/3-1+1 = N/3
• Therefore No. of Columns for Information = 87N-(N/3-1+1) or = 87N-N/3

• There are two types of STS-N signals: STS-N and STS-Nc (N = 3, 12, 48, or 192).  An STS-Nc frame is 125 m seconds in length too. However, the difference between STS-N and STS-Nc frames is that the “Concatenated” frame has only one POH column.
• Because of this one column assignment of POH, an STS-3c signal has two more columns of payload capacity (º 1.152 Mbps) that is available for information (data) transport even though STS-3c and STS-3 have the same signal rate of 155.52Mbps. For any SONET signal STS-Nc, the savings from using only one column of path overhead bytes over an STS-N is (N-1)columns: