What do you call the process of connection from your home to the central office

6.10.1 Theory of Operation and Requirements

In addition to studying the compression algorithm, we also need to learn a little about the operation of telephone systems.

The compression scheme we will use is known as adaptive differential pulse code modulation (ADPCM). Despite the long name, the technique is relatively simple but can yield 2× compression ratios on voice data.

The ADPCM coding scheme is illustrated in Figure 6.27. Unlike traditional sampling, in which each sample shows the magnitude of the signal at a particular time, ADPCM encodes changes in the signal. The samples are expressed in a coding alphabet, whose values are in a relatively small range that spans both negative and positive values. In this case, the value range is {−3,−2,−1,1,2,3}. Each sample is used to predict the value of the signal at the current instant from the previous value. At each point in time, the sample is chosen such that the error between the predicted value and the actual signal value is minimized.

An ADPCM compression system, including an encoder and decoder, is shown in Figure 6.28. The encoder is more complex, but both the encoder and decoder use an integrator to reconstruct the waveform from the samples. The integrator simply computes a running sum of the history of the samples; because the samples are differential, integration reconstructs the original signal. The encoder compares the incoming waveform to the predicted waveform (the waveform that will be generated in the decoder). The quantizer encodes this difference as the best predictor of the next waveform value. The inverse quantizer allows us to map bit-level symbols onto real numerical values; for example, the eight possible codes in a 3-bit code can be mapped onto floating-point numbers. The decoder simply uses an inverse quantizer and integrator to turn the differential samples into the waveform.

The answering machine will ultimately be connected to a telephone subscriber line (although for testing purposes we will construct a simulated line). At the other end of the subscriber line is the central office. All information is carried on the phone line in analog form over a pair of wires. In addition to analog/digital and digital/analog converters to send and receive voice data, we need to sense two other characteristics of the line.

•

Ringing: The central office sends a ringing signal to the telephone when a call is waiting. The ringing signal is in fact a 90 V RMS sinusoid, but we can use analog circuitry to produce 0 for no ringing and 1 for ringing.

Off-hook: The telephone industry term for answering a call is going off-hook; the technical term for hanging up is going on-hook. (This creates some initial confusion because off-hook means the telephone is active and on-hook means it is not in use, but the terminology starts to make sense after a few uses.) Our interface will send a digital signal to take the phone line off-hook, which will cause analog circuitry to make the necessary connection so that voice data can be sent and received during the call.

We can now write the requirements for the answering machine. We will assume that the interface is not to the actual phone line but to some circuitry that provides voice samples, off-hook commands, and so on. Such circuitry will let us test our system with a telephone line simulator and then build the analog circuitry necessary to connect to a real phone line. We'll use the term outgoing message (OGM) to refer to the message recorded by the owner of the machine and played at the start of every phone call.

We have made a few arbitrary decisions about the user interface in these requirements. The amount of voice data that can be saved by the machine should in fact be determined by two factors: the price per unit of DRAM at the time at which the device goes into manufacturing (because the cost will almost certainly drop from the start of design to manufacture) and the projected retail price at which the machine must sell. The protocol when the memory is full is also arbitrary—it would make at least as much sense to throw out old messages and replace them with new ones, and ideally the user could select which protocol to use. Extra features such as an indicator showing the number of messages or a save messages feature would also be nice to have in a real consumer product.

12.7.2 Analog

Enterprise analog connectivity to the CO is almost nonexistent in the 21st century. Many homes still have this class of service, but serious businesses will prefer a digital connection to the service provider. Where analog may have a play is on any station-side connection options. Many PBX vendors still support an analog connection option to desktop phones, and a UMC solution may take advantage of this option because of sunk costs. To take advantage of this situation, an analog⇔SIP gateway will be required. These gateway products are commercially available in four to 32 port versions.

Use of a UMC analog interconnect should be seriously questioned because of loss of key digital features and lack of future support. Features such as caller ID, call waiting, and MWI are typically not supported on an analog service. Additionally, having to add new analog ports on a PBX is often more expensive than adding digital ports, because the PBX vendors are trying to phase out support for this older technology and use pricing strategies to drive customers to digital connections.

10.4.1 Outline of Transmission Line System

Layer 1 architecture for access systems between customer and central office is shown in Figure 10.26. The section between T and V reference points is designated as the access digital section.

Basic access transmission line systems utilize existing two-wire metallic cables for economical and quick penetration of ISDN. Echo cancellation (ECH) schemes and time compression multiplexing (TCM) schemes are described as examples in the Appendices to ITU-T Recommendation G.961 [21]. Two line codes are described for the ECH scheme. Common requirements for these three schemes, such as function, operation and maintenance, transmission medium, system performance, activation/deactivation, and power feeding are specified in the main body of G.961.

For primary rate access, the access digital section at 2048 Kbit/s is specified in G.962 [6], and the access digital section at 1544 Kbit/s is specified in G.963 [7]. These are specifications for T and V reference points. The existing primary rate transmission systems used for trunk lines are applied to the access transmission line.

9.3 System Backplane Interconnection Requirements

The backplane interconnection system requirements are implementation independent (optical or electronic) and provide the requirements for intrashelf (0- to 1-m distances) as well as intershelf (1-m to 1-km distances) PCB interconnections. Fundamentally, the interconnection technology used should look like a digital computer bus. This implies that the signal levels are quantized into two levels (digital) with sufficient noise margins such that optical, electrical, or thermal feedback are not necessary to readjust threshold levels in the transmitter or the receiver. This is radically different from traditional long-haul optical transmission systems where feedback is necessary in order to get the highest level of system performance. The problems that are being solved in switching are very different. They allow simplified transmitter and receiver designs at a low cost per interconnect.

The general system requirements (based on generic switching central office environments) are as follows, for an advanced high-performance system:

Cost/interconnection<$50 (includes: driver, receiver, connectors, and backplane media)

Reliability: 105 h < ρ

Bit error rate (BER): BERdata < 10− 14and BERVoice < 10− 12 (at 1 Gbit/s)

Temperature (ambient): 0 °C < T < 70 °C

Absolute humidity < 0.026 (5% < relative humidity < 95%)

Distance: 1 < length < 100 m

ECL I/O level compatible

Power supplies:+ 5, ground,-2, and-4.8

Bit rate (BR): 0 < BR < 622 Mbit/s

Average group interconnect delay variation (Δτgd):<200 ps

Skew (delay variation within a group (Δτd): across the total interconnect skew < 400 ps (driven by the desire not to have clock recovery circuitry nor allow for any jitter accumulation)

Fan-out: as much as possible (e.g., eight however is application dependent) because point-to-point connections (i.e., ring networks where the fan-out is one) are not as desirable as a bus architecture (where the fan-out is greater than 1) because of the added functionality a bus offers

Packaging size: 1 × 1 in. (integrated 32 channels)

Receiver sensitivity < – 20 dBm

Dynamic range > 16 dB

Low power dissipation (e.g., less than 2 W)

Total crosstalk < – 30 dB

Coupled power in the fiber > 1 mW (at 70 °C).

19.2.1.2 Gigabit time-division-multiplexing passive optical network Standards

TDM-PON is commonly specified with 20 km coverage between the CO and the end users with 1:32, 1:64, or 1:128 splitter as the RN. In a PON system. The upstream and downstream signals are separated by wavelengths using a wavelength diplexer inside the OLT and ONT optical transceivers.

The IEEE 802.3ah E-PON is a PON standard that offers 1 Gbps aggregated bandwidth in both upstream and downstream directions, whereas the ITU-T G-PON offers 2.5 Gbps and 1.25 Gbps in the downstream and upstream directions, respectively.

Both G-PON and E-PON have very similar physical layer characteristics. They both use 1310 nm wavelength for upstream transmission and 1490 nm for downstream transmission.

Miniature Jumper Cables

As local exchange carriers add fiber to their systems, the central office becomes more congested at the fiber distribution frame (FDF). Small jumper cables are needed to reduce the congestion not only at the frame but also inside cable troughs. A simple way to compare the various cordage products available is by cross-sectional area, as shown in Table 5.2.

Table 5.2. Comparison of Cable Cross-Sectional Areas

A 1.6-mm cordage takes up 55% less space than the 2.4-mm cordage and 35% less space than the 2.0-mm cordage. Figure 5.12 shows the jumper pileup as a function of the number of jumpers placed in a 4-in. wide cable trough.

It has been suggested that the pileup depth should not exceed 2 in. to reduce excessive stress to the jumpers during handling. Table 5.3 shows the calculated and empirical number of jumpers per square inch for various-diameter jumpers.

Table 5.3. Cable Density at 2-ln. Pileup Depth

Figure 5.13 shows 1.6-mm-diameter cable in both single-fiber and duplex designs.

1.1.8.3 Public Switched Telephone Network

The PSTN refers to the established copper wire connections between telephone companies' central offices and residences and buildings. Analogue phone lines to residences and businesses dominate energy consumption of the PSTN, accounting for roughly 165 million of the more than 190 million phone lines installed in the United States in 2000. Central office switches typically power analogue phone lines, but the analogue line components that account for much of the switch power draw only draw power on establishing a connection between two phones (i.e., when a call is completed). Other switch elements that continuously draw power include common equipment (one administrative module per switch) and line equipment (e.g., concentration modules, trunk ports). The sheer variety of switching equipment in vintage, scale, and complexity makes it difficult to characterize the installed base. Furthermore, there is a dearth of publicly available power draw data. Together, these two factors complicate development of an accurate estimate of total energy consumption. Nonetheless, because each phone draws an average of approximately 3 W per active line, the energy consumed by analogue phone lines appears to be quite small (on the order of 1 TWh per year).

Battery

A station battery voltage of 24–48 volts dc is supplied from the central office (the term battery is used regardless of the source of the voltage). The ring wire is negative with respect to the tip, but modern telephone instruments are insensitive to the polarity of the applied voltage. The battery voltage is used to power the telephone instrument and to provide current that indicates an off-hook condition. Current is limited to prevent wire heating and personal injury, and to protect the central office against shorts on a line. In the case of a cable telephony system, the network interface device provides this voltage.

5.4.2 Control Network Structure

In its basic form, the control network for a DCS consists of three basic layers, as shown in Fig. 7.107. These layers are various I/Os connected to the controller via an I/O bus, a controller/control system, the operator’s control console, and application processors. The data from the controller is presented to the operator via the data highway. On the data highway, there is one historian and application processor connected for carrying out miscellaneous functions such as trending, performance calculations, etc. There could be redundancy at various levels similar to what is shown in Fig. 7.46. As discussed earlier, there may be smart devices such as IEDs and transmitters connected to the system via a fieldbus, as shown in Fig. 7.108. Integration of various system buses into the systems is shown in Fig. 7.109. In this case, the industrial Ethernet created at the top is responsible for control, monitoring, and supervision because all MMIs are at that level.

In the drawing, a number of controllers have been shown, and these could be OLCS/CLCS or an intermix depending on the manufacturer’s system and the owner’s choice for the particular application. These controller divisions could be on a process or area basis also. The configuration in Fig. 7.109 could be for one unit, and similar such units could be integrated to form a plant network with a common server and/or application processor. One such system to form a plant-level network is shown in Fig. 7.110. Common offsite/systems to the units may interface at that common level. For fewer units, the industrial Ethernet may be extended to form a single network, and a common system may be interfaced via one controller of the Ethernet. So there exist a number of possibilities. In Fig. 7.110, three-level structures of the DCS are shown. The first level pertains to the unit where there are three operator stations, one engineering station, and a performance server for one unit (say, unit 1). The second level is the plant level. At the second level, various units have been connected at a common network. Also, common systems and offsite control systems interface at this level so that the same can be utilized by the operator of various units. Also, at this level, a data exchange between the two units is possible (for example, at the start up time, the unit may check the auxiliary steam header condition of the other unit and open the interconnecting valve to take the supply from the other unit to start).

The third level is the group level, where information exchange with the central office is possible or load distribution programs, etc., can be run. Also, various MIS and enterprise resource planning (ERP) programs, run by the group, can interface at this level. So, for a large network, there are three layers of control and monitoring.

MIS or group level layer: The top layer where all MIS services, ERP, interfaces, etc., are done through a remote network and/or the Internet (IoT/IIoT) so that from a central place, all managerial functions can be done, supervised, and/or advised. Also, units may get the load demand for the LDS through this level. This is the group/MIS level.

Supervisory information system SIS or plant level layer: This is the plant level where the production management at the plant level is possible. Also, here resource sharing between the units is possible. Hence this layer may be termed the SIS layer.

Control and monitoring of the unit layer: The bottom-most layer is for control and monitoring of individual units and/or local process monitoring and control.

32.13.1 Line Concentrating Modules (LCM)

There is a variety of these PMs which may be housed either at the Central Office (host site) or at a location closer to the customer premises being served (remote site). They support analogue lines and provide low level functions like line scanning and ringing.

A maximum of 640 lines can be connected to each LCM. Each LCM consists of two shelves, known as Line Concentrating Arrays (LCA). There can be up to 320 lines connected to each LCA.

A group of LCM connects to a Line Group Controller (LGC), a Line Trunk Controller (LTC), or a Remote Ouster Controller (RCC), which in turn connects to the Switching Network. There are a maximum of three DS30 links per LCA and a maximum of six DS30 links per LCM.

For reliability, each Line Concentrating Array is capable of taking over the lines of the mate LCA. The LCAs are connected by a serial data link that allows one LCA to checkpoint its data with the mate LCA. The data for each call in progress is sent to the mate LCA over this link. If a fault occurs in one LCA, the mate LCA can take over the calls in progress.

Between the two LCA, there can be one or more speech links (DS30). If all channels on an LCA are busy, but the mate LCA has free channels, a call originating on the busy LCA can be routed over one of the inter-LCA speech links to a free channel on the mate LCA. This capability provides access for all lines to all six DS30 links for traffic engineering purposes.

The LGC performs high level functions, such as call coordination and the provision of the different tones required. It is equipped with duplicated processors operating in hot worker/standby mode.

As Figure 32.14 illustrates, there is usually some concentration of lines, depending on the engineering of the particular exchange. The concentration can occur in two places: in the LCM or in the LGC.

Each LGC has a maximum of ten LCM and, therefore, a maximum of 6400 (10 × 640) lines. The LGC has a maximum of 20 ports available for DS30 links from the LCM and, therefore, also has a maximum of 600 (20 × 30) speech links from the LCM.

There are up to 480 (16 × 30) speech links between the Switching Network and the LGC.

Line concentration can be lessened either by reducing the number of Line Concentrating Modules or the number of lines per LCM.

Remote line modules can provide a limited local switching service should the remote site become isolated from the host.