How to choose an ADSL modem. Description of DSL technology. Encoding methods in DSL technology
ADSL - what is it?
Let's start with the name: ADSL stands for Asymmetric Digital Subscriber Line. This standard is part of a whole group of high-speed data transmission technologies under the general name xDSL, where x is a letter characterizing the speed of the channel, and DSL is the abbreviation already known to us Digital Subscriber Line - digital subscriber line. The name DSL was first heard back in 1989, it was then that the idea itself first arose digital communications using a pair of copper telephone wires instead of specialized cables. The imagination of the developers of this standard is clearly lame, so the names of the technologies included in the xDSL group are quite monotonous, for example HDSL (High data rate Digital Subscriber Line - high-speed digital subscriber line) or VDSL (Very high data rate Digital Subscriber Line - very high-speed digital subscriber line). All other technologies in this group are much faster than ADSL, but require the use of special cables, while ADSL can work on ordinary copper pair, which is widely used when laying telephone networks. Development ADSL technologies started in the early 90s. Already in 1993, the first standard for this technology was proposed, which began to be implemented in telephone networks in the USA and Canada, and since 1998, ADSL technology has gone into the world, as they say.
In general, in my opinion, it is still premature to bury the copper subscriber line, which consists of two wires. Its cross-section is quite sufficient to ensure the passage of digital information over quite significant distances. Just imagine how many millions of kilometers of such wire have been laid throughout the Earth since the appearance of the first telephones. Yes, no one has lifted distance restrictions; the higher the speed of information transmission, the shorter the distance it can be sent, but the problem of the “last mile” has already been solved! Thanks to the use of high-tech DSL, adapted to a copper pair, on the subscriber telephone line, it became possible to use these millions of kilometers of analog lines to organize cost-effective high-speed data transfer from the provider, who owns a thick digital channel, to the end user. The wire, once intended exclusively for providing analog telephone communication, with a slight movement of the hand turns into a broadband digital channel, while maintaining its original responsibilities, since owners of ADSL modems can use the subscriber line for traditional telephone communication while simultaneously sending digital information. This is achieved due to the fact that when using ADSL technology on the subscriber line to organize high-speed data transmission, information is transmitted in the form of digital signals with significantly higher frequency modulation than that usually used for traditional analogue telephone communications, which significantly expands the communication capabilities of existing telephone lines.
ADSL - how does it all work?
How does ADSL work? What ADSL technologies make it possible to turn a pair of telephone wires into a broadband data transmission channel? Let's talk about it (;)).
To create an ADSL connection, two ADSL modems are required - one at the provider and one at the end user. Between these two modems there is a regular telephone wire. The connection speed may vary depending on the length of the “last mile” - the further you are from the provider, the lower the maximum data transfer speed.
Data exchange between ADSL modems takes place at three frequency modulations sharply spaced apart from each other.
As can be seen from the figure, voice frequencies (1) are not involved at all in receiving/transmitting data, and are used exclusively for telephone communications. The data reception frequency band (3) is clearly demarcated from the transmitting band (2). Thus, three information channels are organized on each telephone line - an outgoing data transmission stream, an incoming data transmission stream and a regular telephone communication channel. ADSL technology reserves a 4 KHz frequency band for the use of regular telephone service or POTS - Plain Old Telephone Service (plain old telephone service - sounds like good old England). Thanks to this, a telephone conversation can actually be carried out simultaneously by reception/transmission without reducing the speed of data transfer. And if there is a power outage, telephone communication will not disappear anywhere, as happens when using ISDN on a dedicated channel, which, of course, is an advantage of ADSL. It must be said that such a service was included in the very first specification of the ADSL standard, being the original highlight of this technology.
To increase the reliability of telephone communications, special filters are installed that extremely effectively separate the analog and digital components of communication from each other, without excluding joint simultaneous operation on one pair of wires.
ADSL technology is asymmetrical, like Dial Up modems. The speed of the incoming data flow is many times higher than the speed of the outgoing data flow, which is logical, since the user is always more information uploads than transmits. Both the transmission and reception speeds of ADSL technology are significantly higher than those of its closest competitor ISDN. Why? It would seem that the ADSL system does not work with expensive special cables, which are ideal channels for data transmission, but with ordinary telephone cable, which is as perfect as walking to the moon. But ADSL manages to create high-speed data transmission channels over a regular telephone cable, while showing better results than ISDN with its own dedicated line. This is where it turns out that the engineers of Hi-Tech corporations do not eat their bread in vain.
High reception/transmission speed is achieved by the following technological methods. First, the transmission in each of the modulation zones shown in Figure 2 is in turn divided into several more frequency bands - the so-called bandwidth sharing method, which allows several signals to be transmitted on one line simultaneously. It turns out that information is transmitted or received simultaneously through several modulation zones, which are called carrier frequency bands - a method that has long been used in cable television and allows you to watch several channels over one cable using special converters. The technique has been known for twenty years, but only now are we seeing its application in practice to create high-speed digital highways. This process is also called frequency division multiplexing (FDM). When using FDM, the reception and transmission ranges are divided into many low-speed channels, which provide data reception/transmission in parallel mode.
Oddly enough, when considering the method of dividing bandwidth, a widespread class of programs such as Download manager comes to mind as an analogy - they use the method of splitting them into parts to download files and simultaneously downloading all these parts, which makes it possible to use more efficiently link. As you can see, the analogy is direct and differs only in implementation; in the case of ADSL, we have a hardware option not only for downloading, but also for sending data.
The second way to speed up data transfer, especially when receiving/sending large volumes of the same type of information, is to use special hardware-implemented compression algorithms with error correction. Highly efficient hardware codecs that allow on-the-fly compression/decompression of large amounts of information are one of the secrets of ADSL speeds.
Thirdly, ADSL uses an order of magnitude larger frequency range compared to ISDN, which makes it possible to create a significantly larger number of parallel information transmission channels. For ISDN technology, the standard frequency range is 100 KHz, while ADSL uses a range of about 1.5 MHz. Of course, long-distance telephone lines, especially domestic ones, attenuate the reception/transmission signal modulated in such a high-frequency range quite significantly. So at a distance of 5 kilometers, which is the limit for this technology, the high-frequency signal is attenuated by up to 90 dB, but at the same time continues to be reliably received by ADSL equipment, which is required by the specification. This forces manufacturers to equip ADSL modems with high-quality analog-to-digital converters and high-tech filters that could catch a digital signal in the jumble of chaotic waves that the modem receives. The analog part of the ADSL modem must have a large dynamic range reception/transmission and low noise level during operation. All this undoubtedly affects the final cost of ADSL modems, but still, compared to competitors, the costs of ADSL hardware for end users are significantly lower.
How fast is ASDL technology?
Everything is learned by comparison; you cannot evaluate the speed of a technology without comparing it with others. But before that, you need to take into account several features of ADSL.
First of all, ADSL is an asynchronous technology, that is, the speed of receiving information is much higher than the speed of transmitting it from the user. Therefore, two data rates must be taken into account. Another feature of ADSL technology is the use of high-frequency signal modulation and the use of several lower-speed channels lying in a common field of receive and transmit frequencies for simultaneous parallel transfer of large volumes of data. Accordingly, the “thickness” of the ADSL channel begins to be influenced by such a parameter as the distance from the provider to the end user. The greater the distance, the more interference and the greater the attenuation of the high-frequency signal. The frequency spectrum used is narrowed, the maximum number of parallel channels is reduced, and the speed decreases accordingly. The table shows the change in the capacity of data reception and transmission channels when the distance to the provider changes.
Receive channel | Transmission channel | Distance |
8.160 Mbps | 1.216 Mbps | 1.8 km |
7.872 Mbps | 1.088 Mbps | 2.7 km |
3.648 Mbps | 864 Kbps | 3.7 km |
1.984 Mbps | 640 Kbps | 4.3 km |
1.408 Mbps | 544 Kbps | 4.6 km |
960 Kbps | 416 Kbps | 4.9 km |
576 Kbps | 320 Kbps | 5.2 km |
320 Kbps | 224 Kbps | 5.5 km |
128 Kbps | 128 Kbps | 5.8 km |
In addition to distance, the data transfer speed is greatly influenced by the quality of the telephone line, in particular the cross-section of the copper wire (the larger the better) and the presence of cable outlets. Our telephone networks are traditionally of poor quality with a wire cross-section of 0.5 square meters. mm and an ever-distant provider, the most common connection speeds will be 128 Kbit/s - 1.5 Mbit/s for receiving data going to the user and 128 Kbit/s - 640 Kbit/s for sending data from the user at distances of 5 kilometers. However, as telephone lines improve, ADSL speed will increase.
For comparison, let's look at other technologies.
Dial Up modems, as you know, are limited to a maximum data reception speed of 56 Kbps, a speed that I, for example, have never achieved on analog modems. For data transfer, the speed is a maximum of 44 Kbps for modems using the v.92 protocol, provided that the provider also supports this protocol. The usual data sending speed is 33.6 Kbps.
The maximum ISDN speed in dual-channel mode is 128 Kbit/s, or, as you can easily calculate, 64 Kbit/s per channel. If the user calls on an ISDN phone, which is usually supplied with the ISDN service, then the speed drops to 64 Kbps, since one of the channels is busy. Data is sent at the same speeds.
Cable modems can provide data transfer rates from 500 Kbps to 10 Mbps. This difference is explained by the fact that the cable bandwidth is simultaneously distributed among all connected users on the network, therefore, the more people, the narrower the channel for each user. When using ADSL technology, the entire channel bandwidth belongs to the end user, making the connection speed more stable compared to cable modems.
And finally, dedicated digital lines E1 and E3 can show data transfer speeds in synchronous mode of 2 Mbit/s and 34 Mbit/s, respectively. The performance is very good, but the prices for wiring and maintaining these lines are exorbitant.
Today, data transmission technology, called ADSL, has become quite widespread. It is part of a group of technologies called xDSL and is widely used to organize an inexpensive and fairly high-quality connection to global network. The abbreviation stands for Asymmetric Digital Subscriber Line. Using this technology, high-speed Internet access, information transfer, and work with interactive services are provided.
The essence of this method of data transmission, as well as all xDSL technologies, is that ordinary two-wire telephone network wire is used as a high-speed path. Previously, telephone networks were already used to transmit information, for which dial-up modems were used. However, the connection speed organized in this way did not exceed 56 kbit/s. In the case of ADSL, data transfer speeds can reach very high rates, on average up to 8 Mbit/s for download and up to 2 Mbit/s for upload (theoretically, it is possible to receive 24 Mbit/s in the downstream and 3.5 Mbit/s in the upstream ).
Such high results are achieved through full use of line resources and the use of two modems, one on the user side and the other on the telephone exchange. In addition, three transmission streams are organized in the telephone wire: information: incoming and outgoing streams, as well as a stream for voice communication. The operation of the telephone and the network connection are independent of each other, which makes it possible to use telephone communication even while connected to the Internet.
Mutual interference is excluded - the telephone channel is reliably protected by a filtering system; it uses a frequency range of 0.3-3.4 kHz, and the lower limit of the information transmission range is 26 kHz. The high data transfer speed is due to the use of special algorithms for compressing and converting information. Converting a digital signal to analog, transmitting it over a telephone cable and decoding is extremely fast. After this, the resulting digital signal goes to the provider that provides access to the Internet.
The main advantages of ADSL technology are:
- high data exchange speed (exceeds the speed of dial-up connections by at least 50 times);
- always-ready connection that does not require dialing a telephone number;
- full use of line resources;
- simultaneous use of a telephone and a network connection;
- the ability to change the connection speed if necessary without replacing equipment;
- cost-effective connection organization model.
However, ADSL technology also has disadvantages:
- first of all - dependence on the state of the telephone line, cable resistance, signal attenuation, noise level;
- maximum distance to the automatic telephone exchange is 6 km (as a rule, the actual distance of the end user to the port is from 3.5 km to 5.5 km);
- the need to modernize lines (telephone distribution wire, used in most cases, must be replaced with twisted pair);
- the need to purchase subscriber equipment (modem);
- inability to directly connect several end subscribers to the network via the ADSL infrastructure;
- the word Asymmetric in the name of the technology implies the predominance of the volume of incoming traffic over outgoing traffic (it makes it difficult to use peer-to-peer networks, video communications and high-quality online television).
So, today ADSL is relatively cheap and in a simple way organizing mass connections to the Internet. Of course, this technology is already somewhat outdated and is secondary, because it is being rapidly replaced by the infrastructure of networks organized using fiber optic cables using Ethernet technology and WiMax wireless networks (Wi-Fi).
However, so far, despite all the shortcomings and limitations, there is no real alternative to xDSL in general and ADSL in particular. Of course, over time, in the medium term, fiber-optic networks will completely displace xDSL from the access market, and in the long term, obviously, the future of the Internet lies in wireless networks. In the meantime, more than half of connections to the Internet have been made using xDSL technologies, and right now these technologies are not going to lose ground, changing and evolving. Vivid examples of such development are the new ADSL 2 and ADSL 2+ technologies...
Technology Reviews
Dell R630 – an excellent server from a leading company
If the consumer electronics market is ruled by completely different companies, like Lenovo with its laptops, then when it comes to serious work, Dell is still out of competition. Of course, this does not always happen, it all depends on what...Review of the Xbox 360 gaming console
Xbox 360 belongs to the seventh generation of gaming consoles (or game consoles - these are synonyms), its release to the wide market marks its beginning. The development of the console began in February 2003 under the leadership, not much, not less, of...ADSL development
ADSL was developed in the early 1990s and first appeared on US telephone networks in 1993. But widespread worldwide this development received only in 1998.
The abbreviation ADSL stands for Asymmetric Digital Subscriber Line“Asymmetric digital subscriber line”, which initially implies in this technology a difference in data transfer rates from the subscriber and back.
ADSL development assumes the following principles and standards:
ADSL asymmetrical assumes a large volume of data transmitted to the subscriber (video, sound, programs, etc.) and a small volume of data emanating from the subscriber (commands and requests). Therefore, ADSL development is a more suitable option for working with Internet networks.
ADSL modems transmit data much faster than ordinary analog modems, using the same telephone bands. ADSL provides high-speed broadcast access. The data transfer speed reaches 8 Mbit/s (but the standard defines the highest speed as 2 Mbit/s).
ADSL works over existing telephone wiring and does not occupy the telephone channel. Consequently, it allows you to use a telephone and access the Internet.
In practice, the ADSL modem forms three channels:
- high-speed data transmission channel from the network to the computer (speed - from 32 Kbit/s to 8 Mbit/s),
- high-speed data transmission channel from a computer to the network (speed - from 32 Kbit/s to 1 Mbit/s),
- an ordinary telephone communication channel through which telephone discussions are conducted.
The data transfer speed depends on the equipment used, the length and properties of the telephone line.
ADSL does not require, like an analog modem, dialing a number to establish a connection to the network.
Typically, telephone services use the smallest part of the telephone bandwidth, ADSL takes up the rest to implement high-speed transmission.
General topology of a network built on ADSL technology.
When accessing the network from the PBX, an ADSL access multiplexer must be located - DSLAM. It separates subchannels from the general channel. They are respectively divided into voice channel and high-speed channel. The voice subchannel goes to the PBX, and the high-speed channel goes to the router.
In addition to the ADSL modem, it is installed on the subscriber side additional device called an ADSL splitter, designed to separate data and voice frequencies.
The quality of communication does not depend on the class of the telephone exchange. ADSL equipment does not affect operation similar devices like telephones, faxes and modems. In the absence of interference, any devices that do not use frequency multiplexing can operate on the bands (used by security alarm devices, interlocks, AVU, LF/HF).
In terms of cost/quality, ADSL modems are superior to conventional analog modems. The price of ADSL modems is actually comparable to a high-quality analog modem. And in terms of communication quality and transmission speed, they are much ahead of analog modems.
Source of material: http://www.technodesign.ru/kms/info/technologies/adsl/
More detailed information It is possible to obtain it on the websites:
D-LINK- general description ADS technologies
IXBT.com - ADSL Development
DSL technology
DSL technology. Any technology, first of all, provides a specific physical model of the transport environment. One of the promising technologies that allows the transfer of digital information over copper wires (copper wires usually refers to the public telephone network - PSTN or POTS - Plain Old Telephone Service in English abbreviation) is DSL technologies (Digital Subscriber Line - digital subscriber line) .
When using DSL technology (often abbreviated xDSL, where the letter “x” means one of the possible subtechnologies, i.e. variant of the basic technology) there is no need to build a new transport network, because the existing POTS network is used. This is precisely the main economic advantage of DSL technology.
The origins of DSL can be traced back to the early 80s, when Bellcore Corporation developed high-data-rate DSL (HDSL) technology. Channel HDSL was designed to expand the capabilities of T1 technology by replacing the interleaved element coding based on the representation of two bits in one quaternary code (2 binary 1 quaternary - 2B1Q).
The development of Internet services that require high bandwidth (such as video) has created a demand for higher bandwidth connections. Observations show that most of the traffic received from the Internet is intended for the end user (downstream), and only a small percentage is traffic that is actually supplied by the user (upstream). As a result, the channel was developed ADSL(A - Asymmetric - asymmetric digital user line), used in traditional public telephone networks (PSTN - Public Switched Telephone Network).
ADSL technology uses a method that allows the same telephone line to be used simultaneously for both voice and data without increasing the switching requirements of the PSTN telephone network. To reserve a POTS channel with frequencies up to 4 kHz (in telephony, the voice bandwidth is set to 4 kHz), frequency division multiplexing (FDM - Frequency - Division Multiplexing) is additionally used. In this case, digital streams (data) are transmitted at frequencies above 4 kHz (usually starting from 25 kHz).
Due to the continuing decline in distance limitations in DSL technology and the increase in available bandwidth, interest in DSL media has increased in recent years. Before we talk about DSL, let's look at the main types of DSL technology.
- ADSL is the most common DSL technology because it is asymmetrical. This means that the speed of downloading data to the user's computer (modem) is higher than the speed of downloading data to the remote computer. To encode data in ADSL technology, CAP methods (Carrier less Amplitude and Phase modulation - amplitude and phase modulation without a carrier) are used. The CAP method is not a standardized method for a DSL channel, but DMT has been standardized by the ANSI Institute (ANSI T1.413) and the ITU International Union (ITU G.992.1).
- EtherLoop – patented technology of the Elastic Network company – abbreviation for Ethernet local loop – subscriber channel of the Ethernet network. EtherLoop technology uses an advanced signal modulation technique that combines the half-duplex packetization characteristic of an Ethernet network. EtherLoop modems guarantee RF signals only for the duration of the transmission. The rest of the time they use low-frequency control signals. Due to the half-duplex nature of EtherLoop technology, constant throughput can be maintained on either the downstream only or the upstream only. Nortel's system was originally planned for speeds in the range of 1.5 to 10 Mbps, depending on link quality and distance limitations.
- G.L.te – ADSL version with low data transfer speed. It is an addition to ANSI T 1.413. Within the ITU standards committee it is known as G.992.2. It, like ADSL, uses DMT modulation, but a POTS network splitter is not installed in the subscriber's building (usually signal splitting is carried out using the local exchange).
- G.SHDSL – this channel was defined in the ITU standard G.991.2 as a high-speed digital subscriber line on a single twisted pair of wires. G.SHDSL technology is symmetrical, which allows data to be transmitted at the same speed in forward and reverse streams, which is very important because it is intended to replace older telecommunication technologies such as T1, E1, HDSL, HDSL2, circuit-based DSL (SDSL), ISDN and ISDN-based DSL (IDSL).
- HDSL – this channel operates at a speed of 1.54 Mbit/s and has a range of about 2750 m on a wire with a cross-section of 0.5 mm 2. HDSL technology uses 2B1Q line-coded modulation.
- GDSL 2 – this technology was developed in order to ensure the transmission of the T1 signal over the wires of one pair. The technology was created to operate at a speed of 1.544 Mbit/s. It can provide all services that are offered by HDSL technology.
- TDSL – This DSL service, based on ISDN technology, uses 2B1Q line coding and typically supports a data rate of 128 kbit/s. The IDSL service operates on a single pair of wires, and the channel itself can be up to 5800 m long.
- RADSL - used in all RADSL modems, but it in a special way associated with a patented modulation standard developed by Globespan Semiconductor. It uses DMT modems of the CAP.T1.413 standard. The uplink speed depends on the downlink speed, which in turn depends on the line condition and the S/N (signal to noise ratio).
- SDSL – the technology provides a constant data transfer rate and does not have existing standards, which is why it is rarely used.
- VDSL – ultra-high-speed DSL channel for data transmission (Very - high - data - rate DSL) – relatively new technology, designed to increase available data transfer rates (up to 52 Mbps). VDSL technology takes advantage of fiber optic communications and benefits from placing the end equipment closer to the subscriber. By placing end equipment in offices and multi-apartment buildings, it is possible to reduce the length of the local communication line (i.e. subscriber channel), which will increase speed. VDSL technology assumes operation in both asymmetric and symmetric modes.
Table 1 provides a comparison of some types of DSL technologies and shows their most important characteristics, comparable.
Encoding methods in DSL technology
In DSL technology, three main encoding methods are most widely used, briefly discussed below.
Table 1 Comparison of different DSL technologiesTechnology | Max. upstream data rate (Mbit/s) | Max. downstream data rate (Mbit/s) | Wire diameter standard | Maximum distance (meters) | Coding | Standards |
ADSL | 0,8 | 8 | some | 5200 | ATS or DMT | ANSI T1.413 and ITU G.992.1 |
EtherLoop | 6 | 6 | some | 6400 | QPSK, 16QAM, 64QAM |
Patented technology from Elastic Networks |
G.Lite | 0,512 | 1,5 | some | 6700 | DMT | ITU G.992.2 |
G.SHDSL | 2,304 | 2,304 | some | 6100 | TC PAM | ITU G.992.1 |
HDSL | 1,544 T1 2 E1 | 1,544 T1 2.0 E1 | 26 AWG*) 24 AWG*) | 2750 3650 | 2B1Q | ITU G.992.1 |
HDSL2 | 1,544 T1 2 E1 | 1,544 T1 2.0 E1 | 26 AWG*) 24 AWG*) | 2750 3650 | TS RAM | ITU G.992.1 |
IDSL | 0,144 | 0,144 | some | 5800 | 2B1Q | ANSI T1.601 and TR-393 |
RADSL | 1,088 | 7,168 | some | 5500 | ATS or DMT | ANSI T1.413 and ITU G.992.1 |
SDSL | 0,768 | 0,768 | some | 3050 | 2B1Q | ITU G.992.1 |
VDSL | 20 | 52 | some | 910 | CAP/DMT/ DWMT/SLC |
TBD |
1) Quadrature Amplitude Modulation (QAM) corresponds to a change (fixed offset) in the amplitude and phase of the signal to different bit values. Name quadrature amplitude modulation(i.e. QAM) arose because the signals differ in phase by 90 o, and 4 such phases (hence quadrature) together make up 360 o, or a full cycle. Figure 1 (QAM constellation) shows QAM encoding with three bits per baud (signal states are described by different amplitudes and phases). In each direction (0°, 90°, 180° and 270°) there are two points corresponding to two possible amplitude values, resulting in eight different states. If there are eight unique states, then 3 bits can be transmitted in each of them (2 3 = 8).
table 2
|
Table 2 shows the possible values for 8 QAM encoding (8 possible bit patterns). The more different phase offsets and amplitude levels used, the more bits of information can be included in each point or symbol. Problems arise when constellation points are so close that noise on the line or in the receiving equipment makes it impossible to distinguish one point from another.
2) ATS coding – it's adaptive form of QAM code. This method allows the symbol values to be adjusted based on the line condition (eg noise) at the start of the connection. When coding with this method The carrier frequency is removed from the output wave. In the CAP method, frequency division multiplexing (FDM) provides support for three subchannels—POTS, downstream, and upstream.
![](https://i2.wp.com/konturm.ru/tech/img/dsl2.gif)
Voice signals occupy a standard frequency band of 0...4 kHz (see Fig. 2). The CAP method adapts the transmission rate based on the channel state by modifying the bit number or frame (i.e. constellation size + carrier bit rate in baud). This is indicated by various pairs carrier frequencies(for example, 17 kHz and 136 kHz).
Figure 2 shows the frequency spectrum of ACS modulation. Access is supported in two frequency ranges: 25-160 kHz for upstream and 240-1100 kHz (up to 1.5 MHz) for downstream.
3) DMT coding (Discreate Multi-Tone modulation) is a signal transmission method in which the full bandwidth is divided between 255 subcarriers or subchannels with a bandwidth of 4 kHz each. The first subcarrier channel is used for traditional voice and POTS network transmission. Upstream data is typically transmitted on channels 7-32 (26-128 kHz), and downstream data is typically transmitted on channels 33-250 (138-1100 kHz). In reality, the DMT method is a variation of FDM compaction. The incoming data stream is divided into N channels having the same bandwidth, but different average frequency carrier. Using multiple channels with a narrow bandwidth provides the following advantages:
- whatever the line characteristics, all channels remain independent, so they can be decoded separately;
- when using DMT, the transmission coefficient is selected in such a way that each channel can function independently in the presence of noise; this method changes the number of bits per subchannel or tone. The result is a reduction in the overall noise impact of pulsed noise at a constant frequency.
The main characteristics of the DMT method are:
![](https://i2.wp.com/konturm.ru/tech/img/dsl3.gif)
Figure 3 shows the frequency spectrum for DMT modulation.
Typical activation of subscriber equipment for simultaneous viewing of TV programs and access to the Internet is shown in Fig. 4.
![](https://i0.wp.com/konturm.ru/tech/img/dsl4.gif)
A crossover filter (the crossover frequency is usually in the range of 6...8 MHz) is sometimes unreasonably called a splitter. Essentially, this is a frequency diplexer, which includes a low-pass filter (low-pass filter) and a high-pass filter (high-pass filter) in parallel. In particular, such a wiring scheme is carried out by the Stream-TV company.
![](https://i1.wp.com/konturm.ru/tech/img/dsl5.gif)
Figures 5 and 6 illustrate the general possible layouts of physical wiring in the client’s premises. In Fig. 5, the Customer Premises Equipment (CPE) has integrated POTS network splitters, and Fig. 6 shows the line that is branched at the NID (Network Interface Device) device, which is usually the entry point into the subscriber's building. At this point the local communication line becomes the building wiring). In the latter case, the signal (see Fig. 6) supplied to a regular telephone passes through a low-pass filter, and the data elements supplied to the branches pass through the high-pass filter. This approach ensures that in both cases the necessary signals are received. Both topologies are used depending on where the line should branch and where the wires will be physically placed.
DSL Noise Immunity assessed by the criterion of error occurrence rate (BER – Bit Error Rate) BER≤10 -7. When S/N (Signal - to - Noise) is lowered, an excessive number of errors appear in the data stream. The noise margin is understood as the difference in S/N (in dB) for a real line and for BER =10 -7. When S/N (Signal - to - Noise) is lowered, an excessive number of errors appear in the data stream. The noise margin is understood as the difference in S/N (in dB) for a real line and for BER =10 -7.
At any moment in time, both the signal level and the noise level in the line can change, as a result of which the realized S/N value will also change. Note that the higher the DSL link speed, the lower the S/N, and the lower the DSL link speed, the higher the S/N. Consequently, the noise immunity limit will be lower in longer cables (reduced signal strength and increased noise) or at higher transmission speeds on the DSL link.
Rate adaptive DSL (RADSL) technology is a technology in which the transmission rate is adjusted so that the required noise immunity can be maintained, thereby maintaining a BER value below 10 -7. Tests show that the optimal noise margin for DMT services is 6 dB for both downstream and upstream. You should not configure a DSL service with a noise margin that exceeds the optimal value because the system will be prepared for a very low data rate connection over the DSL channel to meet the specified limit. You should also not set the noise immunity limit value too low (for example, 1 dB), because A slight increase in noise will result in excessive errors and a re-training process to establish a connection at a lower bit rate over the DSL link.
The noise immunity of a DSL channel increases as the distance decreases (the noise level decreases) and the wire diameter increases (losses decrease). Of course, increasing the power level on the link will also increase the S/N, but may result in interference with signals from other services on the same cable.
Forward Error Correction(FEC - Forward Error Correction) is carried out mathematically at the receiving end of the transmission channel without a request for retransmission of erroneous data, which allows efficient use of bandwidth for user data. However, we note that even in a situation where no error occurs during transmission, using the FEC method leads to some reduction in throughput, because this adds unnecessary overhead. The ratio of the number of corrected to uncorrected errors shows the efficiency of the error correction algorithm or the relative intensity of errors. There are two main techniques associated with FEC: FEC byte appending and interleaving.
FEC bytes also called control bytes or redundant bytes. FEC bytes are added to the user data stream, thereby providing a means of detecting the presence of erroneous data. On many systems, you can select the number of FEC bytes: 0 (none), 2, 4, 8, 12, or 16. Obviously, the more FEC bytes, the greater the error correction efficiency. However, it should be taken into account that the larger the number of FEC bytes, the more O Most of the communication channel bandwidth will be occupied only by service signals, which is very ineffective for low-noise channels. It can be added that 16 bytes per frame (204 – 16 = 188 bytes of useful information) at a transfer rate of 256 kbit/s takes up a percentage of O more bandwidth than the same number of FEC bytes at 8 Mbps.
In most systems, the FEC overhead is isolated and subtracted from the overall flow before reporting the bit rate on the DSL link. Thus, the observed bit rate on a DSL link is actually the bandwidth available to the user.
Interleaving is the process of rearranging user data in a specific sequence, used to minimize the occurrence of sequential errors in the Reed-Solomon - RS FEC algorithm at the receiving end of the channel. The efficiency of using the RS algorithm when single or time-spaced errors (not occurring sequentially) occurs is higher.
If a noise spike occurs on a copper transmission line, it can affect multiple sequential data bits, resulting in sequential error bits. Since the data in the transmitter is interleaved, de-interleaving the data in the receiver not only restores the original sequence of bits, but also spreads out the erroneous bits over time (the erroneous bits appear in different bytes). Consequently, the erroneous bits are no longer sequential, and the FEC process with the RS algorithm works more efficiently.
Signal power levels on DSL channels significantly higher than those used when transmitting voice data. This is explained by the fact that the linear attenuation of a telephone line increases very quickly with increasing frequency. So, for example, in order to normally receive a signal at the end of a line 5...6 km long, a power of about 15...20 dBm (dBmW) will be required - the number of decibels (dB or dB) measured from a power equal to one milliwatt, calculated at a resistance of 600 Ohms .
Power levels of wideband signals are usually measured in dBm/Hz (dBm/Hz). This value is called power spectral density (PSD - Power Spectral Density):
PSD = P - 60 | (1) |
Formula (1) is valid for a channel bandwidth of 1 MHz, i.e. Applies only to ADSL channel.
Without going into technical details, we note that the following factors play a role in the performance of DSL channels:
Cable losses increase with frequency, primarily due to capacitance distributed along the transmission line ( Y C = j ω WITH).
Note also that the resistance of a copper wire changes significantly with fluctuations in ambient temperature, especially when laying cables along telegraph poles when they are in the sun. Consequently, under some topological conditions, the characteristics of a DSL communication link can vary greatly depending on the time of day. As the temperature increases, the resistance of the wire increases. Losses are also growing. And with increasing resistance (and associated losses), the S/N value decreases due to a decrease in the signal level.
Conclusion
DSL technology can be considered a full-fledged technology that can be used in the last mile of broadband networks. Different flavors of DSL technology may be used in different scenarios, depending primarily on distance and bandwidth requirements. There are many factors that affect connection quality, and many parameters need to be adjusted to improve DSL link speeds and S/N margin. The solution lies in understanding the technology and what factors play what role in the connection.
DSL network topologies can vary greatly among different service providers, so don't assume that if user equipment(CPE) for a DSL network works on one carrier, it will work on another. Different topologies have their advantages and disadvantages, but all topologies are still widely used.
0 Users and 2 Guests are viewing this topic.