TELECOM Digest Tue, 17 Jan 95 10:14:00 CST Volume 15 : Issue 34 Inside This Issue: Editor: Patrick A. Townson B8ZS, AMI, Bipolar Line Coding (William Wood) India Opens Doors to Foreign Telco's (Nikhil) Belgacom and Panaphone Greece (Viviane Engels) Antenna For Cellular Phone in Bangkok (Roland Peter Sauermann) GSM Information Wanted (Vincent Erwig) Question on Call-Back Operators (John Hacking) Anyone Have Experience With LDDS/Metromedia? (Chuck Lukaszewski) TELECOM Digest is an electronic journal devoted mostly but not exclusively to telecommunications topics. It is circulated anywhere there is email, in addition to various telecom forums on a variety of public service systems and networks including Compuserve and America On Line. It is also gatewayed to Usenet where it appears as the moderated newsgroup 'comp.dcom.telecom'. Subscriptions are available to qualified organizations and individual readers. Write and tell us how you qualify: * telecom-request@eecs.nwu.edu * The Digest is edited, published and compilation-copyrighted by Patrick Townson of Skokie, Illinois USA. You can reach us by postal mail, fax or phone at: 9457-D Niles Center Road Skokie, IL USA 60076 Phone: 708-329-0571 Fax: 708-329-0572 ** Article submission address only: telecom@eecs.nwu.edu ** Our archives are located at lcs.mit.edu and are available by using anonymous ftp. The archives can also be accessed using our email information service. For a copy of a helpful file explaining how to use the information service, just ask. ********************************************************************** *** * TELECOM Digest is partially funded by a grant from the * * International Telecommunication Union (ITU) in Geneva, Switzerland * * under the aegis of its Telecom Information Exchange Services (TIES) * * project. Views expressed herein should not be construed as represent-* * ing views of the ITU. * ********************************************************************** *** Additionally, the Digest is funded by gifts from generous readers such as yourself who provide funding in amounts deemed appropriate. Your help is important and appreciated. A suggested donation of twenty dollars per year per reader is considered appropriate. See our address above. All opinions expressed herein are deemed to be those of the author. Any organizations listed are for identification purposes only and messages should not be considered any official expression by the organization. ---------------------------------------------------------------------- Date: Tue, 17 Jan 1995 06:24:16 -0800 From: wewood@ix.netcom.com (William Wood) Subject: B8ZS, AMI, Bipolar Line Coding In telecom 15.29.20, Phillip Schuman asks the question which many people have asked about the issues of B8ZS, AMI and bipolar line coding and their relationship to bit robbing for voice. Though I am not an engineer and have limited understanding of some technical issues, I offer my experience on the matter for those who wish to listen. I hasten to add that it is my intent to convey concept level information in order to build mental constructs which can act as a framework for understanding technicalities. I have often been accused of oversimplification in my efforts. I take this as a compliment, flames notwithstanding, because the majority of people in telecommunications do not have the technical training necessary for understanding engineering level material; but were not stupid, and can understand a coherent explanation if one is offered. First the simple answer. There is no direct relationship between bit robbing and line coding. Even so, there are some indirect relationships if you skew you vision just a little. In order to understand, a bit of history is needed. Bit robbing is a process for transporting single line supervision and signaling between two individual digital carrier (T type) channel units. It developed in a more or less logical fashion from the necessity for converting battery and ground (DC) type supervision into something which could go thru a carrier system. The drop, loop, line, customer, user, access or phone side of a switch (in Plain Ol Telephone Service POTS) consists of two wires. Battery (-48v in USA, 50 v elsewhere) is applied to one of the wires by the switch and ground to the other. When the phone is taken off hook a loop closure is made (known as loop start signaling) which allows current flow between the two (tip and ring) wires. This is supervision to the switch to tell it to put dial tone on the loop pair toward the telephone. Next, the user signals what number they want by either rotary dialing or pushing keys on the DTMF (dual tone multifrequency) pad. In the case of the rotary dial, current flow (battery to ground) is interrupted in series intervals which in turn tell the switch where to connect. This kind of DC (direct current) signaling/supervision works fine as long as all the lines are physically attached to the same switch. Let us say though, that you want to connect to a line that is out of a switch which is in another city, state or country. The facilities between switches are shared on an as-needed basis and are know as trunks. The most significant thing to know at this point is that battery and ground signaling and supervision cannot now, nor could they ever, be sent over a carrier system. Carrier systems are used to connect switches by providing the physical facilities for the trunks. (The techies will insist that metallic [no carrier] trunks can send DC. True, but there are so few left in the USA, and the Dxing methods so numerous, I will not deal with them here.) Analog carrier systems can transport only audible (voice) tones so a conversion process is done to change battery and ground signaling into an audible tone. Each of the individual voice frequency channels in a carrier system are band limited to 4 KHz. The usable part (for transporting conversations) of this pancake of capacity is further limited to around 3 KHz (300 - 3400 Hz). This range is known as the VF Band to transmission folks. DC signaling/supervision can be converted into a tone that is either in-the-band or out-of-the-band. As an example, type N1 carrier channels internally can convert a battery on-off signal into a 3700 Hz on-off tone signal. This tone is out of voice band signaling because it's above 3500 Hz. Filters in the channel units block the signaling tones from the callers. A second option is to use an external tone generator such as an SF (signal/single frequency) unit. SF units typically convert DC supervision and signaling into 2600 Hz (USA) on-off tones. This is an in band (between 300 and 3400 Hz) signal type. Notch filters are supposed to keep this tone from the callers too. (Sometimes the filters dont kick in quick enough and you can hear the tone chirp for just an instant as the distant end hangs up on some long distance calls even today. Not much of the in-band left because a good whistler can put up bogus calls so the phone companies have disconnected most of these types of trunks.) With the advent of digital carrier systems (T carrier in early 60's) a decision was made to continue to use in-band (SF) signaling but also support a twist on the out-of-band idea. Because the signal leaving an individual channel unit was now digital, a single pulse (bit) out of the periodic series of eight bits belonging to that channel could be designated as a signaling/supervision bit. To illustrate; say that an off-hook, battery or seizure signal was presented to the channel unit by the phone or switch. This seizure would cause the channel unit's signaling function to stop applying an idle state mark bit onto the trunk facility. The receiving channel unit would detect the change from marking to spacing for this bit position and in turn apply a seizure toward its own switch or phone. The earliest T systems reserved this bit all the time for signaling (and sometimes even a second bit position) but this was soon replaced by the current system. Because each channel creates an eight bit unit (octet or byte), any single bit occurs 8 thousand times a second (the Nyquist rate for a 4 Hz channel). The system was originally designed to facilitate only interoffice dial pulse trunks. (In fact, in my area the telephone company didn't even originally consider T as a carrier system. It was an interoffice transport mechanism belonging to a different union group than that which maintained carrier systems.) The rate for dial pulses is only 10 to 12 pulses per second, so using a full 8 Kbps of channel bits is massive overkill. The current system still uses a single bit out of the eight, but now it only does it during the sixth appearance of the time slot into which the channel byte or octet is regularly dumped. The concept is that there is 64 Kbps signal (8 K Nyquist x 8 bit unit) ostensibly being created by the channel unit for payload transfer, but occasionally (every 6th byte or octet from the channel) a single bit is robbed and designated as a signaling/supervision indicator for a DC state on the drop side of the channel unit. This still results in creating a 1.3 Kbps out of (voice) slot signaling sub channel (sliced into A, B, and sometimes C & D sub-sub channels) which is only semi-massive overkill. Because dial pulsing is so slow for address transfer, some trunking over T carrier uses the bit robbing function for off /on hook supervision only and MF (multi-frequency which is not the same set of tones as DTMF) for address transfer. In some applications the bit rob is not used at all. SF is used for hook supervision and DTMF for address transfer. Lots of variations here on the main theme of using a carrier transmission bit on/off as a representation of a signaling/supervision battery on/off condition. Hold this thought. The Bipolar, AMI and B8ZS question relates to the composite, or multiplexed, signal leaving the digital (channel bank) mux. A standard DS1 digital stream consists of 24 DS0 time division multiplexed to run at a combined rate of 1.544 Mbps (24 x 64 Kbps + 8 Kbps framing) plus or minus a little . On twisted pair cables this is, even today, a significant number of bits. The digital square wave which is the DS1 stream is physically constructed of either +3, -3 or zero voltage pulse positions. The +3 and -3 volt pulses are known as marks and the zero voltage pulse or time positions are called spaces, (words straight out of 19th century telegraph). For a number of technical reasons, the transmitter is required to reverse the DC voltage value as each mark is transmitted. That's where AMI gets its name. Each Alternation of a Mark must be Inverted in voltage value. Example with mark as 1, and space as 0: 101010111000001 could be transmitted (left to right) as +3,z,-3,z,+3,z,-3,+3,-3,z,z,z,z,z+3. This signal form came to be known as bipolar because it has two (bi) poles of voltage, although it is actually a ternary (three) form because zero volts also counts as a state. If you've stayed with me this long, you get the reward here. The B8ZS is what seems to put the twist in everyone's knickers. Since the earliest days of T carrier, there has been a ones density and consecutive zero requirement for the DS1 stream. For our purposes, the most significant is that there be no more than 15 consecutive spaces (zero voltage bit times) in the ongoing DS1 stream. On the face of it, this would not seem to be a problem because each channel generates only an 8 bit unit for transmission. Even if it sent an all zero octet, no harm would be done. Remember though the DS1 is a serial, time division multiplexed stream with one channel octet following another like ducks in an endless row. If two consecutive channel time positions contained all zero octets the total zeros in series would be 16, which is sufficient to cause clocking errors on the system. There are three solutions to this problem. The first, and original, is to restrict each individual channel from ever sending an all zero byte. This is the solution which has always restricted services like DDS from sending 64 Kbps. Each individual DDS channel unit allows only seven of the eight bits to be used for DDS payload and reserves the 8th bit to make it a mark in order to insure the channel can't send all zeros. Because the Nyquist rate in T carrier channels is always 8 k, and the DDS channel unit gives over only 7 bits to it, the result is 56 Kbps. It's important to note here that even though the results of restricting zeros and bit robbing for signaling/supervision may result in a channel payload of less than the full 64 Kbps DS0, the two things are not related. The second way to meet pulse density rules is to have the transmitter actively monitor the DS1 stream as it is applied to the transmission line. This is the B8ZS solution. In this process when the transmitter recognizes a series of eight or more zeros (and it doesn't need to know if they are all from one channel or a combination of two channels) it will substitute a different digital bit pattern for that series of eight. Instead of sending z,z,z,z,z,z,z,z it will send z,z,z,v,s,z,v,s. The z is zero volts, the v is a 3 volt pulse in the same polarity as the last valid mark signal which is technically a violation of the AMI rule. The s is a substitution bit for the original zero value at this time position. At the receiver, this string is converted back to z,z,z,z,z,z,z,z and given over to the channel receivers. Bipolar 8 Zero Substitution (binary 8 zero suppression?) permits any individual channel to send any combination of 8 bits with no restrictions. This allows interoffice systems to provide clear channel 64 Kbps service. Once again, notice that this has no direct relationship to bit robbing for supervision. It has an indirect (skew your vision a little) relationship because a robbed bit could be changed on the line to conform to B8ZS format, but it will be changed back at the receiver to its correct form with the process being transparent between the near and far channel units. European systems can also do this process but use a different substitution pattern. The third way is called ZBTSI for Zero Byte Time Slot Interchange. But, that's a story for another time. WE Wood Technotranslater Techish to English Translation Techtrans Animatics Group ------------------------------ From: Nikhil Subject: India Opens Doors to Foreign Telcos Organization: National Centre for Software Technology, India Date: Tue, 17 Jan 1995 06:52:35 GMT India has opened its doors to foreign telecom companies for providing the telephone service in the local loop. So far, the Indian Government held the monopoly for telecom for the last fifty years. This resulted in a long waiting list of up to ten years for getting a telephone connection. For the last three years India has been reforming its economic policies to speed up the development. Realizing that good telecommunications are a necessity in a liberal economy, it has opened it doors to foreign telcos and banned government companies from investing in the telecom sector. The highlights of the new guidlines anounced yesterday are: * Foreign telecom companies can bid for providing services as joint venture with a local company. * Companies presently having minimum 0.5 million lines in service can qualify to bid. * Foreign equity participation must be minimum 10% and maximum 49%. * Area divided for bidding into Telecom circles of class A,B and C where class A circle is high density city. * Minimum net worth of the joint venture company should be Rs.300 crores (US$ 100 million) for Class A, Rs.200 cr. for Class B and Rs.100 crores for Class C areas. * Assured business for next fifteen years with provision to extend for another ten years. However the long distance and international business still remain a monopoly of government companies for the next five years. Nikhil Thakkar email: INTERNET: nikhil@shakti.ernet.in X.400 : G=SYSTEMS S=NOVASOFT A=VSNB C=IN ------------------------------ From: Viviane.Engels@rug.ac.be (Viviane Engels) Subject: Belgacom Be-Panaphone Greece Date: 17 Jan 1995 05:59:36 GMT Organization: University of Ghent, Belgium I send this letter to protest for the services of Belgacom Proximus Cellular Telephone service and that of the Greek Panaphone. I am an owner of a Panaphone number and I am visiting Belgium since 10 December 1994. My telephone is not working in a certain area of Belgium even the Proximus signal is very stong. This area is established between the cities: Ghent, Antwerpen, Brugge, Kortrijk. In none of the above cities the phone is working (except Kortrijk). The Belgium coast has also a problem. The networks work normal at the rest of Belgium (as far as I checked). After my contact with both companies they just blame each other. Belgacom thinks the problem is for whole Belgium territory (so its a Greek problem) and the Greeks say there is a problem in a part of Belgium without being able to define where (Belgium problem). My remark is that all these people except doing their best to solve the problem there are also deaf (they have a client service department just to put nice music on the phone when someone is waiting for an answer). Finally except that they are losing money and credibility for their services, they are also against the law since in Greece is advertised use of Greek cellular phones in whole Belgium territory. If someone responsible is listening I whould appreciate an answer. P.S. I know Greek people doing business in Belgium that they were forced to buy a Belgian-Proximus number also just to work. [TELECOM Digest Editor's Note: Ah, readers ... does the complaint above sound familiar. Telco blames long distance carrier, and long distance carrier blames telco for whatever is wrong. Each convinced the other must correct the problem. Are there any European readers of the Digest who can make suggestions to this fellow? PAT] ------------------------------ From: roland@nwg.nectec.or.th (Roland Peter Sauermann) Subject: Antenna For Cellular Phone in Bangkok Date: 17 Jan 1995 11:31:54 GMT Organization: National Electronics and Computer Technology Center, Bangkok I am struggling with pretty poor AMPS 800 service in Bangkok on my three Watt Diamond-Tel cellular phone that I have permanently mounted in my car. I have noisy calls, call overlapping, and disconnected calls. The problems are probably not unique to any big city where the cells are overloaded, and there are lots of tall buildings and interference. My question is what is the optimum antenna for this environment? I currently have one of those antennas where the cable side is glued on one side of the windshield and the antenna proper is on the other side. My windshield is also very steep on my truck so the angle between the base and the antenna is very small. úÿ Would I do better with a Magnetic mounted antenna on the roof? What would suit me better a 3db or a 5db one? There seem to be a whole bunch of options, some cost three times others. One guy told me the material was different ... some antennas simply have squiggles in them others have a thick section of a ceramic or hard plastic. I know I need to be sure I get one for the correct phone system (we have five different types here from 470mhz, 800, 900 to the new Digital stuff). How do I evaluate these different antennas that run between $25-$60 and apparently have different properties? Thanks for any suggestions. Cheers, Roland P. Sauermann roland@nwg.nectec.or.th Bangkok, Thailand ------------------------------ From: v.erwig@stud.tue.nl (Vincent Erwig) Subject: GSM Information Wanted Date: Tue, 17 Jan 1995 13:32:37 Organization: Eindhoven University of Technology, The Netherlands Can anybody give me some information on where and how I can find information about GSM? I'm interested in the development of the GSM network, and the specific features that GSM / GSM telephone has, what new technologies have been used, and the advantages / disadvantages compared to other cellular phone systems. I need this information for a study project. Many thanks in advance. Vincent Erwig V.Erwig@stud.tue.nl ------------------------------ Date: Tue, 17 Jan 1995 20:55:00 +1000 From: JOHN.HACKING@telecom.telememo.au Subject: Question on Call-Back Operators There has been a great deal of discussion in the Australian media lately about Call-Back Operators and I'm interested in how they operate from a customer's point of view. From the little I know, it would appear that there are four possible ways of Call-Back Operator customers placing calls. These are: 1. Customer calls an international freecall number and talks to a live operator who then calls the customer back and presents the customer with dial-tone. The customer then dials the international number required. 2. Automatic version of the above -- customer calls an international freecall number, lets it ring three times and then hangs up. Then some sort of black box works out which customer called, calls the customer back and presents the customer with dial-tone. The customer then dials the international number required. The customer can only do this from one telephone number. 3. Customer calls an international freecall number, enters an account number PIN number and the number that they are calling from. A black box then validates the customer account details and then calls the customer back on the phone number they entered and presents the customer with dial-tone. The advantage of this method over two is that the customer can get the call back to any phone number. 4. A "bombardment" system -- the call back operator's black box polls the customer's phone line somehow and presents the customer with dial-tone without any call back. My questions are: a) Are there any other methods of operation available? Obviously method four is the most convenient from the customer point of view. b) How does the "bombardment" system work (method four)? c) Are there significant savings in using call-back operators or is it just a matter of a few cents per minute depending on time of day? d) Are there any hidden tricks or traps that customers should be aware of? I run a small business and make a significant number of calls to the USA, UK and Canada and I'm interested in saving money on my international traffic. Any information or advice would be appreciated. Thanks in anticipation. John Hacking Manager Queensland Training Systems [TELECOM Digest Editor's Note: Why would method four (polling) be the most convenient? I think it would be the least convenient since you never know when you are going to want to make a call versus when it the next time you are going to be 'polled'. The other thing to watch out for where callback systems are concerned are the *huge* number of misdialed calls (and/or telemarketer calls) to your 'callback number' which result in a call being made to you at all hours of the day and night (relative to your time of day) resulting in many cases in admin charges levied to your account for calls you did not make, to say nothing of the inconvenience of answering the phone at three in the morning your time to be greeted with callback dialtone you don't want merely because some fool in the USA accidentally dialed your callback number and let it ring a couple times before deciding he dialed in error. I sold a service called Telepassport for awhile, and wrong number/telemarketer calls to the numbers on their switch were an awful nuisance to their subscribers. They tried everything to reduce the instances of wrong numbers, i.e. they set their switch so if *more than one ring was received* before the caller disconnected, it was assumed to be a wrong number. This meant legitimate subscribers had to be instructed to dial their callback number and disconnect *IMMEDIATLY* when they heard a 'click' indicating they had connected and the switch was about to return an audible ringing signal. The assumption was telemarketers would let it ring at least three or four times before disconnecting, as would most people getting a wrong number. Then instead of using numbers in lower Manhattan -- apparently an area where incorrect dialing is legion -- they got blocks of numbers in New Jersey, from an area where it is assumed telemarketers are less interested due to the demographics of the area. You know what? Their callback subscribers still get awakened at odd hours by unsolicited callbacks due to someone in the USA triggering the switch in error (or on purpose, trying to sell something). There ought to be a flag which tells the CO receiving a call if the call is from within or without the USA, and to reject those calls which originate in the USA. ("I'm sorry, the number you dialed cannot be reached from within the USA"). I am pretty certain this can be done since not long ago I tried to call a place in Haiti (of all places!) and the response I got at the distant end after dialing the number was "the telephone you are calling does not accept calls from international points ... this is a Cable & Wireless recording, ". PAT] ------------------------------ From: clukas@mr.net (chuck lukaszewski) Subject: Anyone Have Experience With LDDS/Metromedia? Date: Tue, 17 Jan 1995 14:23:22 GMT Organization: Minnesota Regional Network I received some information from LDDS/Metromedia yesterday about their long distance service. The rates seem entirely too good to be true, and I'm wondering if anyone here has experience (good or bad) that they would share. At the moment, I'm spending a lot of time on the telephone to NYC. After a pretty exhaustive evaluation of AT&T, Sprint and MCI we found the lowest we could get was 23.7 cents per minute peak on AT&T. LDDS claims to charge 15 cents per minute with a one year commitment (includes a 90-day out clause). I talked to AT&T and they're pulling the "we're regulated and can't compete with those numbers" routine. chuck lukaszewski clukas@mr.net imp@krissy.msi.umn.edu ------------------------------ End of TELECOM Digest V15 #34 *****************************