Ordering the Circuits Configuration

Ordering the Circuit’s Configuration

Before you can begin to order your circuit, you need to know your circuit’s configuration. A dedicated circuit must be built the same on your phone system that it is on your carrier’s hardware in order to function properly. If your hardware is set up to speak B8ZS/ESF line coding and your carrier is set up for AMI/D4, you won’t be able to communicate. This is similar to a person who only speaks Swahili trying to talk to a person who only understands Russian. Until they are conversing in the same language, nobody will under-stand anything. In the telecom world, you have to worry about the protocol used to set up and tear down calls as well as the location of these messages to allow your hardware to communicate effectively with the network.

 Tip  To avoid having to understand this entire section on configuration, ask your hardware vendor to fill out the technical part of your carrier’s order form. It’s better to have someone who knows your hardware to fill out the form, than to try to figure out the techie stuff and do it yourself. If you are replacing existing circuits, ask your current carrier for your circuit’s configuration. If you’re coming to the end of your term with the carrier and you’ve had consistent problems for the past six months, the carrier might be suspicious that you’re thinking of switching services, but the carrier should still give you the information. Use the following sections to familiarize yourself with the information your new carrier will need. After you have a general grasp of what the new carrier needs, go through the technical worksheet and fill out what you can. When you find a section you’re unclear about, ask the carrier for clarification, and then pull the information from your hardware or your current carrier to fill out the section.

Every bit of the configuration is crucial when ordering a dedicated circuit. The greatest challenge is that there aren’t standard package configurations that everyone uses. The way your phone system is set up is uniquely tailored to meet your company’s needs and limitations, the features of your phone system, and the options available from your carrier.

 Tip  Speak with your hardware vendor whenever you think about changing your carrier or your phone system. The hardware vendor can be a tremendous help in determining the costs, timelines, and limitations of your system.

Time division multiplexing and clock source

Time division multiplexing, or TDM, is a technology that enables you to take a T-1 circuit that passes 1.54 Mbps and parse it out into 24 individual channels of 56 or 64 kbps. TDM does this in a methodical and controlled manner so that your carrier can decode and recombine data, using the same methodical and controlled manner. This technology is quite complex, and makes up the protocol for the vast majority of dedicated circuits used for voice calls in the world. Fortunately, you only need to know the basics to use it effectively.

TDM is based on sampling small sections of phone calls very often. The easiest analogy to imagine is that there are 12 sets of people who want to have conversations on the same phone line. In order to make sure everyone receives equal time, a stopwatch dictates when each pair is given its turn to speak. Because 12 pairs of people are speaking, each group can speak for only five seconds every minute. As the hand of the stopwatch passes from 12 to 1, the first couple has five seconds to speak. When the second hand passes from 1 to 2, the second couple has five seconds, and so on, down the line. After you speak for 5 seconds, your phone goes dead until everyone else has had their time, and then 55 seconds later, you can speak for another 5 seconds. Figure 8-1 shows the five seconds you have to speak, before you have to wait for the clock to return to the 12, at which point you get your time again.

Figure 8-1: This is an example of TDM based on a stopwatch dictating the allotted time for each conversation.

Asking people to speak for only five seconds per minute is impractical, so the world of telecom speeds the process up. Imagine the stopwatch moving faster. Instead of the second hand passing around once a minute, imagine the stopwatch sped up to complete an entire cycle once a second. Now you have to wait only a fraction of a second before you can speak again. The problem is that you still only have 1/12 of a second to say anything, so your conversation would be very choppy and generally incomprehensible. The beauty of TDM is that your conversation isn’t sampled only one time a second, but 3,000 times per second, so you have to wait only a few milliseconds before it’s your time to speak. Because what you’re saying is being sampled and transmitted so often, you don’t even know that 23 other conversations are going on around you on the same T-1 circuit. However, both phone systems still know who gets the correct timeslots; your hardware packages your call in the correct timeslot and sends the call down the line to your carrier, who uses the same system to unpackage and combine your call on the other end. This system allows everything to flow down the line without information being mixed up or lost.

 Remember  With a TDM system, your hardware has a clock, and so does your carrier. The two clocks are perfectly synchronized (like two dancers) so that they can both know when to begin and end a call’s timeslot. The system functions like a synchronized waltz between the two pieces of hardware, and just as with dancing, someone has to lead. Your carrier should always be configured to lead, — to be the primary clock source — that dictates what the exact time is.

 Warning!  If you have multiple circuits from different carriers, you must confirm that your hardware is capable of assigning the primary clock source individually, on a per-circuit basis. Maybe your hardware is designed to handle multiple T-1 lines, but if it can’t identify more than one primary clock source for the group of circuits, you face some pitfalls. If all of your circuits are coming from the same carrier, this software configuration is fine, because all the circuits will probably use the same clock source from the group. If you have circuits from four carriers, each circuit needs to use the clock source from its own carrier. If your card handles four circuits, yet pulls clocking from one circuit and applies it to the remaining three, you will experience cumulative frame slips and errors on the other circuits. The frame slips represent your hardware failing to synch up with your carrier; this situation may cause echo or static on your calls, or it may have no effect until the circuit is so far behind the carrier clock source that the whole system crashes. After you restart your hardware, everything will work just fine — for a while, until your system falls behind and everything crashes again. The bottom line is that your hardware must be able to handle a primary clocking source for every carrier. If it doesn’t, and you want to use TDM, you’ll need to upgrade your hardware.

Line coding and framing

The line coding and framing of a T-1 define the overhead and useable bandwidth available for each of the 24 channels. The overhead of a channel works with the clocking and is responsible for the maintenance and housekeeping on a call, including the call setup and tear down, ringing, and those fun recordings that you get sometimes that tell you the number you have dialed is disconnected or is no longer in service. The useable bandwidth is everything left on the dedicated channel that isn’t overhead and is used to transmit the speech in the conversation. There are only two options when it comes to line coding and framing. You can have

• B8ZS/ESF (which stands for Binary 8 Zero Suppression/Extended Super Frame — say that five times fast). Traditionally, data transmissions are passed across B8ZS/ESF circuits. These circuits have more useable bandwidth and less overhead.

• AMI/SF (which stands for Alternative Mark Inversion/Super Frame, sometimes also called D4/SF). Voice transmissions are traditionally passed across AMI/D4 circuits. These circuits have less useable bandwidth and more overhead.

As you can see in Figure 8-2, both T-1s have 24 channels, with the top shaded portion identifying the overhead for the circuits.

 Tip  The increased overhead on the AMI circuit allows for more of the bandwidth to synchronize voice transmissions. Circuits that are used to transmit data need less overhead, because any problem in transmission can be overcome by resending the data.

Figure 8-2: Each of the 24 slots equals one DS-0 or channel on the T-1.

 Warning!  Ensure that you order the correct line coding and framing on your circuits. If you are ordering a long-distance T-1 and AMI/SF is delivered instead of B8ZS/ESF, correcting the mistake can easily take about two weeks. This is because the local carrier that delivers the circuit has to physically show up at your office to reconfigure their hardware. As long as the line coding and framing are correct, you can generally change the lower-level protocols, such as the outpulse signal and start on the fly when you activate your circuit. This is generally not the procedure for most carriers, but, as long as you get a warmhearted installation technician on the line, you’ll be well taken care of.

Outpulse signal and start

The outpulse signal and start are two protocols working together to identify how calls are initiated and terminated. The outpulse signal is used to set up and tear down calls. The start dictates the required sequence of events. For example, the start determines whether your carrier needs to wait to send users a dial tone, or whether the carrier needs to send you a signal first to acknowledge that users want to hear a dial tone. When you want to make a call from your home, you pick up your phone, at which point a signal is sent to your local carrier to send you a dial tone. A few milliseconds later, you hear a dial tone, press seven or ten digits, and your call is processed.

In the world of dedicated circuits, you have options about how you request a dial tone, and whether the overhead consists of a one-way or a two-way conversation. There are only three options for line coding and framing. See the following sections for more information.

Loopstart

Loopstart is a protocol whereby the carrier is always waiting for you to make a call. Loopstart is essentially a one-way conversation, but it’s an effective and simple protocol.

Groundstart

Groundstart is another one-way conversation, but from the opposite end of the circuit. In this case, you tell the network when you want to make a call. It’s more of a command from your hardware telling the carrier “Hey, I’m making a call, process it!”

E&M Wink or E&M Immediate

E&M is an old telecom protocol that stands for ear and mouth. It’s the only bidirectional protocol of the group and has its niche in the telecom world. Instead of receiving orders like groundstart or waiting for calls like loopstart, the E&M protocol is like a constant conversation between your hardware and the hardware of your carrier. It looks something like this:

Your hardware: “I’m going to make a call; give me dial tone.”

Carrier: “Okay, here is a dial tone.”

Your hardware: “I’m sending the phone number; do you have it?”

Carrier: “I got the phone number and I am processing it.”

Carrier: “The call reached the final office; I am sending you a ring tone now.”

Your hardware: “Okay, I got the ring tone.”

Carrier: “Someone answered the call; I will stop sending the ring.”

Your hardware: “Okay, the ring has stopped.”

Carrier: “I am sending you the conversation now . . .”

E&M protocol comes in two flavors, E&M Wink and E&M Immediate.

• E&M Wink: You have the full conversation with E&M Wink. This might seem chatty, but if you have toll-free numbers that are riding on your circuit, you might need E&M Wink in order to use all the information and features sent in the call’s overhead.

• E&M Immediate: If you aren’t using toll-free numbers, E&M Immediate might be a faster and more efficient protocol.

   Remember  If your circuit will be processing calls for toll-free numbers that require DNIS or ANI delivery, you must use E&M Wink. Groundstart, Loopstart, and E&M Immediate don’t provide the bidirectional protocol required to use these dedicated toll-free features. Other out-of-band signaling protocols provide DNIS and ANI delivery, but they have to be offered as options on your hardware. Skip ahead to the section called “Feature Group D” if you are interested in these other protocols. If you aren’t sure what DNIS or ANI delivery is, Chapter 5 gives you all the info you need.

Introducing trunk groups

Trunk groups are the contiguous channels within your dedicated circuits that are partitioned off from the rest of your T-1. The trunk groups can split a single T-1 into smaller sections, or they can combine several T-1s into one large group. If you combine three circuits into one trunk group, you can then receive calls on your toll-free number on all 72 channels just as you would with one large circuit. If each T-1 has its own trunk group, your toll-free calls can come in on only 24 channels; otherwise, you have to set up overflow to the other trunk groups, which might cost more.

Trunk groups only affect how your carrier sends calls to your circuit. If your carrier splits your T-1 into 4 trunk groups of 6 channels each, your phone system can still dial out on all 24 lines, in any order you want, without any concerns. Because your phone system determines the channel you dial out on, it isn’t bound by trunking parameters set up by your carrier. It simply seizes whatever channel it wants and dials.

Because some trunk groups are large and cover multiple T-1s, there are standard ways you receive the calls within a trunk group. Your options are as follows:

• Least Idle: In this case, the channel that has been active the longest receives the next call. Call centers typically use this feature because their fastest employees get more of the calls. Generally, this option can overload a single port on your hardware and might result in failures, so I don’t recommend it.

• Most Idle: This is a better scenario; the system sends the call to the phone line that has been idle for the longest time. This type of routing distributes calls evenly across all your channels, without burning out one specific channel.

• Ascending: Calls are sent into your system in sequential order, starting at channel 1, and going up to channel 24 (or the end of the trunk group). The first channel on your circuit will probably always be active, because this option continues to backfill the lower channels when they become available.

• Descending: The opposite of ascending, calls are sent to your circuit in sequential order, but this time starting with channel 24 (or the top end of your trunk group) and on down.

 Tip  Try setting up your inbound calls by using descending and configuring your phone system to use ascending for outbound calls. In this scenario, your inbound calls are working down from channel 24 and your outbound calls are hunting up from channel 1. This reduces your chances of glare, which is an undesirable condition that occurs when an inbound call and an outbound call attempt to seize the same channel.