Access and Transmission Facilities

Connections to PSTN switches can be divided into two basic categories: lines and trunks. Individual telephone lines connect subscribers to the Central Office (CO) by wire pairs, while trunks are used to interconnect PSTN switches. Trunks also provide access to corporate phone environments, which often use a Private Branch eXchange (PBX)—or in the case of some very large businesses, their own digital switch. Figure 5-4 illustrates a number of common interfaces to the Central Office.

Lines

Lines are used to connect the subscriber to the CO, providing the subscriber access into the PSTN. The following sections describe the facilities used for lines, and the access signaling between the subscriber and the CO.

• The Local Loop

• Analog Line Signaling

• Dialing

• Ringing and Answer

• Voice Encoding

• ISDN BRI

The Local Loop

The local loop consists of a pair of copper wires extending from the CO to a residence or business that connects to the phone, fax, modem, or other telephony device. The wire pair consists of a tip wire and a ring wire. The terms tip and ring are vestiges of the manual switchboards that were used a number of years ago; they refer to the tip and ring of the actual switchboard plug operators used to connect calls. The local loop allows a subscriber to access the PSTN through its connection to the CO. The local loop terminates on the Main Distribution Frame (MDF) at the CO, or on a remote line concentrator.

Remote line concentrators, also referred to as Subscriber Line Multiplexers or Subscriber Line Concentrators, extend the line interface from the CO toward the subscribers, thereby reducing the amount of wire pairs back to the CO and converting the signal from analog to digital closer to the subscriber access point. In some cases, remote switching centers are used instead of remote concentrators.

Remote switching centers provide local switching between subtending lines without using the resources of the CO. Remotes, as they are often generically referred to, are typically used for subscribers who are located far away from the CO. While terminating the physical loop, remotes transport the digitized voice stream back to the CO over a trunk circuit, in digital form.

Analog Line Signaling

Currently, most phone lines are analog phone lines. They are referred to as analog lines because they use an analog signal over the local loop, between the phone and the CO. The analog signal carries two components that comprise the communication between the phone and the CO: the voice component, and the signaling component.

The signaling that takes place between the analog phone and the CO is called in-band signaling. In-band signaling is primitive when compared to the out-of-band signaling used in access methods such as ISDN; see the "ISDN BRI" section in this chapter for more information. DC current from the CO powers the local loop between the phone and the CO. The voltage levels vary between different countries, but an on-hook voltage of –48 to –54 volts is common in North America and a number of other geographic regions, including the United Kingdom.

Dialing

When a subscriber dials a number, the number is signaled to the CO as either a series of pulses based on the number dialed, or by Dual Tone Multi-Frequency (DTMF) signals. The DTMF signal is a combination of two tones that are generated at different frequencies. A total of seven frequencies are combined to provide unique DTMF signals for the 12 keys (three columns by four rows) on the standard phone keypad. Usually, the dialing plan of the CO determines when all digits have been collected.

Ringing and Answer

To notify the called party of an incoming call, the CO sends AC ringing voltage over the local loop to the terminating line. The incoming voltage activates the ringing circuit within the phone to generate an audible ring signal. The CO also sends an audible ring-back tone over the originating local loop to indicate that the call is proceeding and the destination phone is ringing. When the destination phone is taken off-hook, the CO detects the change in loop current and stops generating the ringing voltage. This procedure is commonly referred to as ring trip. The off-hook signals the CO that the call has been answered; the conversation path is then completed between the two parties and other actions, such as billing, can be initiated, if necessary.

Voice Encoding

An analog voice signal must be encoded into digital information for transmission over the digital switching network. The conversion is completed using a codec (coder/decoder), which converts between analog and digital data. The ITU G.711 standard specifies the Pulse Coded Modulation (PCM) method used throughout most of the PSTN. An analog-to-digital converter samples the analog voice 8000 times per second and then assigns a quantization value based on 256 decision levels. The quantization value is then encoded into a binary number to represent the individual data point of the sample. Figure 5-5 illustrates the process of sampling and encoding the analog voice data.

Figure 5-5. Voice Encoding Process

[View full size image]

Two variations of encoding schemes are used for the actual quantization values: A-law and m-Law encoding. North America uses m-Law encoding, and European countries use A-law encoding. When voice is transmitted from the digital switch over the analog loop, the digital voice data is decoded and converted back into an analog signal before transmitting over the loop.

The emergence of voice over IP (VoIP) has prompted the use of other voice-encoding standards, such as ITU G.723, G.726, and ITU G.729. These encoding methods use algorithms that produce more efficient and compressed data, making them more suitable for use in packet networks. Each encoding method involves trade-offs between bandwidth, processing power required for the encoding/decoding function, and voice quality. For example, G.711 encoding/decoding requires little processing and produces high quality speech, but consumes more bandwidth. In contrast, G.723.1 consumes little bandwidth, but requires more processing power and results in lower quality speech.

ISDN BRI

Although Integrated Services Digital Network (ISDN) deployment began in the 1980s, it has been a relatively slow-moving technology in terms of number of installations. ISDN moves the point of digital encoding to the customer premises. Combining ISDN on the access portion of the network with digital trunks on the core network provides total end-to-end digital connectivity. ISDN also provides out-of-band signaling over the local loop. ISDN access signaling coupled with SS7 signaling in the core network achieves end-to-end out-of-band signaling. ISDN access signaling is designed to complement SS7 signaling in the core network.

There are two ISDN interface types: Basic Rate Interface (BRI) for lines, and Primary Rate Interface (PRI) for trunks. BRI multiplexes two bearer (2B) channels and one signaling (D) channel over the local loop between the subscriber and the CO; this is commonly referred to as 2B+D. The two B channels each operate at 64 kb/s and can be used for voice or data communication. The D channel operates at 16 kb/s and is used for call control signaling for the two B channels. The D channel can also be used for very low speed data transmission. Within the context of ISDN reference points, the local loop is referred to as the U-loop. It uses different electrical characteristics than those of an analog loop.

Voice quantization is performed within the ISDN phone (or a Terminal Adapter, if an analog phone is used) and sent to a local bus: the S/T bus. The S/T bus is a four-wire bus that connects local ISDN devices at the customer premises to a Network Termination 1 (NT1) device. The NT1 provides the interface between the Customer Premises Equipment (CPE) and the U-loop.

The NT1 provides the proper termination for the local S/T bus to individual devices and multiplexes the digital information from the devices into the 2B+D format for transmission over the U-loop. Figure 5-6 illustrates the BRI interface to the CO. Only ISDN devices connect directly to the S/T bus. The PC uses an ISDN Terminal Adapter (TA) card to provide the proper interface to the bus.

The ISDN U-Loop terminates at the CO on a line card that is specifically designed to handle the 2B+D transmission format. The call control signaling messages from the D channel are designed to map to SS7 messages easily for outbound calls over SS7 signaled trunks.

Trunks

Trunks carry traffic between telephony switching nodes. While analog trunks still exist, most trunks in use today are digital trunks, which are the focus of this section. Digital trunks may be either four-wire (twisted pairs) or fiber optic medium for higher capacity. T1 and E1 are the most common trunk types for connecting to End Offices. North American networks use T1, and European networks use E1.

On the T1/E1 facility, voice channels are multiplexed into digital bit streams using Time Division Multiplexing (TDM). TDM allocates one timeslot from each digital data stream's frame to transmit a voice sample from a conversation. Each frame carries a total of 24 multiplexed voice channels for T1 and 30 channels for E1. The T1 frame uses a single bit for framing, while E1 uses a byte. Figure 5-7 shows the formats for T1 and E1 framing.

Figure 5-7. T1/E1 Framing Formats

[View full size image]

The E1 format also contains a channel dedicated to signaling when using in-band signaling. The T1 format uses "robbed bit" signaling when using in-band signaling. The term "robbed bit" comes from the fact that bits are taken from the PCM data to convey trunk supervisory signals, such as on/off-hook status and winks. This is also referred to as A/B bit signaling. In every sixth frame, the least significant bits from each PCM sample are used as signaling bits. In the case of Extended Superframe trunks (ESF), A/B/C/D bits are used to indicate trunk supervision signals. A/B bit signaling has been widely replaced by SS7 signaling, but it still exists in some areas.

Trunks are multiplexed onto higher capacity transport facilities as traffic is aggregated toward tandems and transit switches. The higher up in the switching hierarchy, the more likely optical fiber will be used for trunk facilities for its increased bandwidth capacity. In North America, Synchronous Optical Network (SONET) is the standard specification for transmission over optical fiber. SONET defines the physical interface, frame format, optical line rates, and an OAM&P protocol. In countries outside of North America, Synchronous Digital Hierarchy (SDH) is the equivalent optical standard. Fiber can accommodate a much higher bandwidth than copper transmission facilities, making it the medium of choice for high-density trunking.

Standard designations describe trunk bandwidth in terms of its capacity in bits/second. The basic unit of transmission is Digital Signal 0 (DS0), representing a single 64 kb/s channel that occupies one timeslot of a Time Division Multiplex (TDM) trunk. Transmission rates are calculated in multiples of DS0 rates. For example, a T1 uses 24 voice channels at 64 kb/s per channel to produce a DS1 transmission rate of 1.544 mb/s, calculated as follows:

24 x 64 kb/s = 1.536 kb/s + 8000 b/s framing bits = 1.544 mb/s

The optical transmission rates in the SONET transport hierarchy are designated in Optical Carrier (OC) units. OC-1 is equivalent to T3. Higher OC units are multiples of OC-1; for example, OC-3 is simply three times the rate of OC-1. In North America, the electrical equivalent signals are designated as Synchronous Transport Signal (STS) levels. The ITU SDH standard uses the STM to designate the hierarchical level of transmission. Table 5-2 summarizes the electrical transmission rates, and Table 5-3 summarizes the SONET/SDH transmission rates.

Table 5-2. Electrical Transmission Rates

Designation	Voice Channels	Transmission Rate mb/s
T1 (North America)	24	1.544
E1 (Europe)	30	2.048
E3 (Europe)	480	34.368
T3 (North America)	672	44.736

Table 5-3. SONET/SDH Transmission Rates

SONET Optical Level	SONET Electrical Level	SDH Level	Voice Channels	Transmission Rate mb/s
OC-1	STS-1	—	672	51.840
OC-3	STS-3	STM-1	2016	155.520
OC-12	STS-12	STM-4	8064	622.080
OC-48	STS-48	STM-16	32,256	2488.320
OC-96	STS-96	STM-32	64,512	4976.64
OC-192	STS-192	STM-64	129,024	9953.280
OC-768	STS-768	STM-256	516,096	39,813.120

In additionto copper and fiber transmission mediums, microwave stations and satellites are also used to communicate using radio signals between offices. This is particularly useful where it is geographically difficult to install copper and fiber into the ground or across rivers.

ISDN PRI

Primary Rate Interface (PRI) provides ISDN access signaling over trunks and is primarily used to connect PBXs to the CO. As with BRI, PRI converts all data at the customer premises into digital format before transmitting it over the PRI interface. In the United States, PRI uses 23 bearer channels for voice/data and one signaling channel for call control. The single signaling channel handles the signaling for calls on the other 23 channels. This scheme is commonly referred to as 23B+D. Each channel operates at a rate of 64 kb/s. Figure 5-8 illustrates a PBX connected to the CO through a PRI trunk.

Other variations of this scheme use a single D channel to control more than 23 bearer channels. You can also designate a channel as a backup D channel to provide redundancy in case of a primary D channel failure. In the United States, U-Loop for PRI is a four-wire interface that operates at 1.544 mb/s. The U-Loop terminates to an NT1, which is typically integrated into the PBX at the customer premises.

In Europe, PRI is based on 32 channels at a transmission rate of 2.048 mb/s. There are 30 bearer channels and two signaling channels, which are referred to as 30B+2D.