Every subscriber in the world is identified by a number, which is geographically tied to a physical location.4 This is the telephone number. The telephone number, as we used it above, is seven digits long. For example:
234–5678
The last four digits identify the subscriber line; the first three digits (i.e., 234) identify the serving switch (or exchange).
For a moment, let’s consider theoretical numbering capacity. The subscriber number, consisting of the last four digits, has a theoretical numbering capacity of 10,000. The first telephone number issued could be 0000; the second number, if it were assigned in sequence, would be 0001, the third would be 0002, and so on. At the point where the numbers ran out, the last number issued would be 9999.
The first three digits of the example above contain the exchange code (or central office code). These three digits identify the exchange or switch. The theoretical maximum capacity is 1000. If again we assign numbers in sequence, the first exchange would have 001, the next 002, then 003, and finally 999. However, particularly in the case of the exchange code, there are blocked numbers. Numbers starting with 0 may not be desirable because in North America 0 is used to dial the operator.
The numbering system for North America (United States, Canada, and Caribbean islands) is governed by the North American Numbering Plan (NANP). It states that central office codes (exchange codes) are in the form NXX, where N can be any number from 2 through 9 and X can be any number from 0 through 9. Numbers starting with 0 or 1 are blocked numbers in the case of the first digit N. This cuts the total exchange code capacity to 800 numbers. Inside these 800 numbers there are five blocked numbers such as 555 for directory assistance and 958/959 for local plant test.
When long-distance service becomes involved, we must turn to using still an additional three digits. Colloquially we call these area codes. In the official North American terminology used in the NANP is “NPA” for numbering plan area, and we call these area codes NPA codes. We try to assure that both exchange codes and NPA codes do not cross political/administrative boundaries. What is meant here are state, city, and county boundaries. We have seen exceptions to the county/city rule, but not to the state. For example, the exchange code 443 (in the 508 area code, middle Massachusetts) is exclusively for the use of the town of Sudbury, Massachusetts. Bordering towns, such as Framingham, shall not use that number. Of course, the 443 exchange code number is meant for Sudbury’s singular central office (local serving switch).
There is similar thinking for NPAs (area codes). In this case, these area codes may not cross state boundaries. For instance, 212 is for Manhattan and may not be used for northern New Jersey.
Return now to our example telephone call. Here the calling party wishes to speak to a called party that is served by a different exchange (central office5). We will assign the digits 234 for the calling party’s serving exchange; for the called party’s serving exchange we assign the digits 447. This connectivity is shown graphically in Figure 1.5. We described the functions required for the calling party to reach her/his exchange. This is the 234 exchange. It examines the dialed digits of the called subscriber, 447–8765. To route the call, the exchange will only work upon the first three digits. It accesses its local look-up table for the routing to the 447 exchange and takes action upon that information. An appropriate vacant trunk is selected for this route, and the signaling for the call advances to the 447 exchange. Here this exchange identifies the dialed number as its own and connects it to the correct subscriber loop, namely the one matching the 8765
number. Ringing current is applied to the loop to alert the called subscriber. The called subscriber takes her/his telephone off hook, and conversation can begin. Phases 2 and 3 of this telephone call are similar to our previous description.
