T1 Training
- Kelli Higdon
The following is taken directly from our Reseller/Admin training classes.
T1 Basics
T1 is the physical delivery mechanism of DS1 signaling
DS1 is a T-Carrier signaling scheme created by Bell Labs
T1 generally means any 1.544 mbps service
T1 and DS1 are still often used interchangeably
DS1 (Digital Signal 1) is the transmission protocol used over a physical T1 line. A DS1 is made up of 24 64 kbps channels also known as timeslots or DS0’s.
Because DS1 is virtually the only standard used in North America, the terms DS1 and T1 are often used interchangeably.
DS1 – Information rate of 1.544 mbps
1 frame is 193 bits wide
Frames occur at 8000 frames per second
24 timeslots (DS0), each 8 bits wide
8 bits wide x 8000 frames per second = 64 kbps
1 bit for framing
1 bit x 8000 frames per second = 8 kbps
Maximum rate is 1.536 mbps
Framing overhead is 8kbps
1 framing bit x 8000 frames per second
24 DS0 x 64kbps = 1.536mbps
The information within a DS1 is sent in frames at a rate of 8,000 frames per second where each frame is 193 bits wide. The break down for each frame is as follows:
8 bits per channel x 24 channels per frame + 1 framing bit = 193 bits 193 bits per frame x 8000 frames per second = 1544000 bits per second 1544000 bits per second = 1.544 mbits per second
Due to a framing overhead of 8kbps (1bit per frame x 8000 frames per second), maximum data rate is 1.536 mbps which breaks down into 64kbps per timeslot.
T1 Line Encoding
Establishes how bits are put onto the wire
Alternate Mark Inversion (AMI)
0 = No Voltage, 1 = Voltage
Every other 1 is a different polarity
Prevents significant build up of DC
Long runs of 0’s still a problem
No voltage = No signal = Synchronization loss
Reserved 8th bit of each byte to meet “ones density”
00000000 = 00000001
7 of 8 bits = 56 kbps vs 64 kbps
Bipolar Violation, 1’s of the same polarity
Corrupted/Weak signal, 1 interpreted as 0 (visa versa)
Violations indication of transmission error
Line encoding establishes how binary bits are put onto the wire. On a T1, ones are sent by applying voltage to the wire, where a zero is sent by having no voltage on the wire. Alternate Mark Inversion, or AMI, is found on older multiplexing equipment and dictates the following three states: A zero is no voltage and a one (also called a mark) is positive and negative voltage.
The AMI system alternates polarity with each mark to help prevent significant build up of DC. This is important because DC voltage can be used to provide immediate power for repeater equipment. Alternating polarity also acts as a basic form of error checking in the form of bipolar violations (where two marks of the same polarity should not occur unless there is an error on the line) which can occur with corrupted or weak signals.
One major drawback to AMI is that it is susceptible to a long run of zeros. This is bad because repeating equipment can lose timing without any marks to count. As such, there is a general rule of “Ones Density” which is no more than 15 consecutive zeros in a row. With AMI, “ones density” was achieved by utilizing a form Zero Code Suppression where the 8th bit of every byte of every frame is set to a 1 during transmission.
Because the 8th bit of every byte is reserved, you now effectively only have 56 kbps per channel instead of 64 kbps.
Binary 8 Zero Substitution (B8ZS)
Avoids pitfalls of serial binary transmissions
Introduces “Bipolar Violation”
Meets “ones density” requirement
All 8 bits available = 64 kbps
B8ZS is the most common encoding type you will encounter in a North American PRI. It’s a method of line coding used in the T-carrier system which allows a full 64kbps per channel.
To recap: AMI is a standard of line encoding that specifies three states: no voltage is a zero, positive and negative voltage are ones (or marks). It uses bipolar violations to detect errors. B8ZS builds on AMI by using double bipolar violations to replace a pattern of eight zeros in a row.
In the image above, you have the original signal of eight zeros “00000000” The B8ZS encoded signal is now “00011011” The signal polarity (assuming the previous mark was negative) is (-)000-+0+-
Since this sort of violation will not be seen during normal transmission, the repeating and terminating equipment know exactly what this signal is and passes it along without considering it an error. At the destination, the bipolar violated 00011011 is converted into 0000000. With B8ZS, we have a full 64 kbps channel and meet the ones density requirement. It’s fairly obvious what the advantage of a B8ZS encoded line is over AMI.
Depending on where you live, carriers will offer you an ESF framed, B8Zs encoded PRI. However, if for some reason you come across a customer who wishes to use AMI/D4 because it’s slightly cheaper or they’ve received some sort of misinformation that it’s superior to ESF/B8ZS, you can produce a convincing counter-argument.
T1 Framing
Two framing formats, D4 and ESF, both use a Framing Bit to establish grouping
D4 (Superframe)
Groups of 12 Frames
6 Framing bits = Signaling bits
6 Framing bits = Alignment bits
Used to align equipment for framing
ESF (Extended Superframe)
Groups of 24 Frames
6 Framing bits = Alignment bits
6 Framing bits = Signaling bits
12 Framing bits = Data Link bits
Used to align equipment for framing
24 Frames = Additional overhead in framing bit
Frame synchronization is necessary to identify channels within each 192 bit frame. This synchronization takes place by allocating a framing bit. This results in an 8kbps channel for framing data. You will typical see this referred to as the 193rd bit, but the framing bit actually precedes the other 192 bits.
Extended Superframe, sometimes referred to as D5 framing, is the most common framing type in a North American PRI. ESF is preferred over Superframe also known as D4 because it includes a cyclic redundancy check and includes bandwidth for a data link channel which passes out-of-band data between carrier equipment.
Extended Superframe (ESF)
Includes a Cyclic Redundancy Check (CRC)
Includes bandwidth for a data link channel
1.536 Mbps after framing overhead
An ESF superframe is 24 frames long
8 bits per frame
Frames 4,8,12,16,20,24 used for alignment
Frames 1,3,5,7,9,11,13,15,17,19,21,23 for data link
Frames 2,4,6,10,14,18,22 pass CRC information
Inband T1
No D-Channel
Only use if carrier cannot provide a PRI
Analog lines bundled as a “T1”
24 Channels, FXS or E&M
No advanced features
No Caller ID (CID) by default
No Dialed Number Identification Service (DNIS) by default
No dedicated error checking
Simply put, avoid at all costs
One type of T1 service is an Inband T1. Although you get 24 channels for voice compared to 23 channels on a PRI, you do not have a dedicated channel for call signaling and control, thus signaling information is carried within each channel of the T1.
Inband T1s do not provide any built-in capability for Dialed Number Identification Service or Caller ID data; you need a PRI for that.
Since signaling data is sent along with each channel (inband), setup and destruction of a call typically takes longer than over a PRI with a dedicated channel.
ISDN
Two types of ISDN
Basic Rate Interface (BRI)
Primary Rate Interface (PRI)
Two types of channels
B-Channel (Bearer)
64 kbps per channel
Used to transport data (including voice)
D-Channel (Delta)
16 kbps / channel (BRI)
64 kbps / channel (PRI)
Intended for signaling and control
Common Channel Signaling (CCS)
There are two types of channels in ISDN. A Delta Channel which is typically referred to as the D-Channel and a Bearer Channel or B-Channel. The D-Channel is used for signaling and control; caller ID, automatic number identification, setup and destruction of a call, DNIS, channel requested, response requests are all handled over this channel.
The B-Channel is used to transmit in full duplex, either voice or data at a rate of 64kbps.
There are two ISDN implementations: a BRI (Basic Rate Interface) and PRI (Primary Rate Interface). A BRI consists of 3 channels (2 64k B-Channels and 1 16k D-Channel) that total 128kbits/s and is typically meant for home or small business use.
With CCS (common channel signaling), signaling information is only sent for a particular channel when necessary, unlike an inband T1 where the 8th bit of every 6th frame is reserved for signaling information.
ISDN T1 (PRI)
1 Delta Channel (D-Channel)
Setup and Destruction Information
Caller ID, Number Dialed
Information Services
Removes overhead from B-Channels
23 Bearer Channels (B-Channel)
Data transmissions (Audio)
24 Possible 64kbps Channels
1.536 Mbps Full Duplex (+8 kbps framing overhead)
Only available on B8ZS (North America)
A T1 PRI, which is the main focus, consists of 23 64 kbps B-channels and 1 64 kbps D-channel for a total of 1.536mbps which, after framing overhead, brings us to the total 1.544 mbits of a T1.
Keep in mind that a PRI does not automatically mean a T1. Most of the world uses E1 lines which have 30 B-Channels (2.048 Mbit/s total capacity).
On a PRI, you will always have only 1 D-Channel per T1 whether you have 1 B-Channel or 23. A D-Channel actually has the capacity to support up to 20 T1 circuits via Non-Facility Associated Signaling. The drawback to this is that you have 1 D-Channel that could impact 19 other T1 lines, so you could set a D-Channel backup. Why NFAS? Switches such as the Lucent 5ESS don’t handle “common channel signaling (CCS)” which is what an ISDN is on the same line card that terminated the T1 circuit, so the telephone company needs to buy a separate signaling card for each D-Channel. It’s unlikely for you to come across the need for NFAS, but this can give you an idea of associated cost.
On the other end of the scale, as mentioned earlier, B-Channels can be used for either data or voice communication in the form of a fractional or flex PRI.
PRI_CPE vs PRI_NET
Integration with 3rd Party Hardware
Channel Banks, Fax Servers, Recording Hardware, Predictive Dialers
Switchtype
national, ni1, 5ess, 4ess, dms100
Clearer Quality / Superior Equipment
Typically has HWEC at carrier level
Built-in CID and DID features
There are two signaling parameters for a PRI. PRI_CPE stands for Customer Premise Equipment and PRI_NET stands for Network, or essentially the provider.
You set signaling to correspond with what your Asterisk server sees itself as. For example, when receiving a PRI line from your provider, Asterisk sees itself as the CPE so we set signaling as PRI_CPE.
When connecting with channel banks, fax servers, recording hardware and predictive dialers, chan_zap would become a “provider” in a sense, and thus be set to PRI_NET.
There are different switchtypes that are available with a PRI. Depending on your location, national (also known as national 2) is the most common switch type.
PRI will sound clearer not only because you have clear channel (64 kbps) signaling, but the hardware used is newer in order to support ESF/B8ZS which has the added benefit of robust hardware echo cancellation.
This page has no content. Enrich Array by contributing.