The Protocol
Required Components
– Keypad
– Display
– Memory
– Logic
– Phase Locked Loop and Frequency Synthesis
Signalisation via the radio channel
The B-Netz
IMTS

Development of Automated Radio Telephony

What happened to car phones in the 1960s? First of all, the electronics for sending and receiving were improved. The tubes were gradually replaced by powerful transistors. As a result, phones became smaller and cheaper. But it didn’t change the fact that they were an appendage of the telephone network. An operator was needed, a lady in office who would arrange the connection. Establishing a radio connection with an operator took around 2 minutes. During this time the valuable radio channel was occupied. Assuming an average call duration of 2 minutes, the efficiency of the call was 50%. In addition, more and more flexibility was missing, for example when choosing radio channels. More capacity was needed. Capacity was the driving force of improvements, back in the sixties until today.

The Protocol

What was the difficulty of an “automatic connection”. Let’s take a look at what is necessary if a car telephone subscriber wants a connection to the telephone network.

First of all, the subscriber must enter the telephone number of the desired landline subscriber, which is typically a 10-digit number. To do this he has to interact with the telephone. This is a given today, but back then it was uncharted territory. Enter of a number like on a landline telephone with a rotary dial would have been anachronistic and also prone to errors. It is better to make an entry using 10 keys like on a pocket calculator. The phone now has to “remember” each telephone number, i.e. it has to store the number.

After each entry, there should be a confirmation from the telephone, for example by showing the number entered on a display.

Once the subscriber has entered the full number, he or she presses a dial button. Now it is the telephone in cooperation with the electronics of the base station that has to establish the connection.

First, the mobile telephone searches for an available channel by checking available channels. To do this, the mobile telephone must be able to set a channel independently and check whether it is free. If this is the case, the phone sends a signal to the base station to occupy that channel. The base station confirms this request and blocks the channel for all other users. Now the mobile phone can send information about what type of call he wants, e.g. whether he wants to receive information about the charges etc. The base station acknowledges the request by sending the same message back. Now the base station has to transmit the digits from its memory, digit by digit. If there are no more digits to be transmitted, this will be communicated with another signal. The base station now begins to call the subscriber in the landline network using the telephone number. It also switches the radio channel through and the mobile phone subscriber is connected to the landline subscriber. If the mobile phone subscriber finally hangs up the receiver, a signal is sent to the base station, which disconnects the radio connection.

Let’s summarize what we need technically:

• A keyboard for entering numbers and requirements
• A display of the number and the respective status
• A memory for the digits
• A control for the processes with logical decisions (protocol processing)
• Signaling to send information back and forth between the mobile telephone and the base station.
• A channel selection without mechanical control

Required Components

As shown, new car telephones need a keyboard, a display, memory and logic modules for sequence control. Digitally adjustable frequencies were also necessary. All of this was provided by new electronics components of the seventies.

Keypad

Every phone these days has the same layout of digits from 0 to 9. It consists of 4 rows and 3 columns. 1 2 3, 4 5 6, 7 8 9, *0#. This arrangement comes largely from the invention of DTMF dialing (Dual Tones Multiple Frequencies). The keys are arranged in a matrix which is connected to 2×4 audible tones. (See DTMF). This is how the first push-button telephones emerged in the USA. This button arrangement was also adopted for car telephones.

Display

Until the 1960s, there was little options in how information was displayed. The most common means was a light bulb or an analogue display, for example to show the reception strength of a station.

Red LEDs came onto the market in the 1970s. (see LED). This was used for 7 segment displays. These became primarily known in the first pocket calculators and digital watches.

Storage

4 bits are required to store a digit: 0000 = 0, 0001 = 1, 0010 = 2, 0011 = 3, 0100 = 4, 0101 = 5, 0110 = 6, 0111 = 7, 1000 = 8, 1001 = 9.

So you need 40 bits for 10 digits and a 64 bit memory would be sufficient. Memory chips appeared in the 1970s. Driven primarily by the computer industry. See: memory

Logic

Let’s look again at what a mobile telephone has to do to make a call.

When the telephone is turned on, it waits for “instructions” from the user. It is in a “state” that we can call “waiting for input”. Only “events” can cause the phone to exit this state. These events can be:

• A key is pressed
• A call is coming in

If a key is pressed, the phone checks which key was pressed and whether it is a number key. If it is a digit, it checks where the number should be stored. This is done with a pointer that can be incremented. It then stores the digit in a memory at the correct location and increments the digit pointer. The phone then returns to the state in which it waits for (further) input.

If a “Dial” key is pressed, the phone jumps to another state in order to search for a free communication channel.

Anyone who is already familiar with computer programming is familiar with such processes. You write and read data or values ​​and make decisions based on these values. What you basically need is:

• Memory
• Pointer or Counter
• Logic to make decisions

The new semiconductor industry provides all of this. We have already discussed storage above. Counters can be implemented by connecting many transistors in an IC and logic components are also implemented using transistors and are available as ICs e.g. from Fairchild or Texas Instrument. See logic.

PHASE LOCKED LOOP AND FREQUENCY SYNTHESIS

The new mobil telephones required an electronically switchable channel frequency. Until now, frequencies, for example in radio receivers, were set manually using a variable capacitor. Such technology was of course not applicable for automization. But the integrated circuits also led to improvements here. They provided an important element called the frequency divider. A frequency could be divided down using fast transistor circuits. So 100 MHz becomes a frequency of 1 MHz when divided by 100.

Frequency dividers are used in so-called phased locked loops (PLL). (See PLL). They allow the synthesis of (almost) any frequencies.

Signalization over a radio channel

How does the mobile telephone communicate with the base station? The following information must be exchanged:

• Channel number
• Telephone number
• Telephone identification
• Type of conversation
• Start of the dialing
• End of dialing

At the end of the 1960s, the TeKaDe company was again commissioned by the German Federal Post Office to develop a new system for public mobile land radio (ÖbL) which would enable automatic-dialing. TeKaDe developed a suitable data protocol. For this purpose, a series of 16 pulses each was sent. Each pulse had a length of 10 ms. So the whole sequence lasted 160 ms. A tone was sent in each pulse. If the tone was 2070 Hz, it corresponded to a logical 0. If the tone was 1959 Hz, it corresponded to a logical 1.

Already in the sixties, information was transmitted digitally using logical zeros and logical ones

Pos	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
	Synchronization					Information						Information mirrowed
Data

The transmitter and the receiver must therefore be able to send or recognize two tones. The first five pulses were always the same to ensure synchronization and to reliably identify the start of the transmission from position 6. Positions 6 to 10 contained the information. Position 11 was always at 0, positions 12 to 16 repeated the information mirrored. This means that five bits are available for information transfer.

How does a call from a mobile telephone work? Let’s assume the number is already stored and a call is to be made. The phone searches for a free channel. Once it recognizes a free channel, it sends a 2070 Hz tone to the base station. The base station blocks the channel for other participants and sends a 1950 tone. This causes the cell phone to send its own phone number using the data pulses described above. To check and acknowledge, the base station sends the same numbers back. If they are the same, the call continues; if not, the call is canceled. The mobile telephone now sends the telephone number, which is also sent back for checking. If this is succesful, the base station establishes a connection. The conversation can start.

But how does somebody call a mobile telephone? For this purpose, a special transmission channel channel „19“ was reserved for the base station. The landline subscriber needs to know where the mobile phone subscriber is located. If, for example, the mobile subscriber is in Bremen area, first the Bremen area code (0421) is dialed followed by 05 (mobile phone) and then the actual five-digit number of the mobile subscriber.

The base station then sends this number on channel 19 along with the channel to be used by the subscriber. Every cell phone has to constantly check channel 19 to see whether a number is coming and whether it is its own number. If this is the case, it reports to the base station via the specified channel.

The B-Network

The system that was developed by TeKaDe in and for Germany was called the B-Netz (B-Network) to show the succession to the A-Netz. It was put into operation in 1972 and initially ran parallel to the A-network.

Initially, 38 channels were provided in the lower 2 m band, around 150 MHz. The channel spacing was significantly smaller compared to the A network with 20 kHz, so that 28 channels could be accommodated in a relatively small band. The spacing of the duplex bands was 1.6 MHz. When major capacity bottlenecks occurred around 1980, the A network was shut down and a second B network band was made available in the same band at 160 MHz. This led to more complex transmit/receive electronics because two different duplex filters now had to be handled.

The telephones for the B network were all built by TeKaDe in Nuremberg. Other manufacturers used the TeKaDe hardware and only provided it with their own logo. A B-network telephone consisted of a control panel (as shown in the picture above), a separate handset and electronics in a box that were still very large and needed a lot of cooling. In the 1980s there were some improvements in electronics that allowed the system to shrink to shoebox size.

The identity of a B-network phone has been “hard-wired”. The numbers were programmed in using solder joints and the soldering was sealed so that the numbers could not be changed. However, this remained a weak point of the network as it was relatively easy to create a false identity.

In addition, the conversations were easy to listen to. All you needed was an ordinary FM receiver tuned to the appropriate channels and you could listen to a conversation.

Smaller and more features with Microprocessor

In the beginning of the eighties, the electronics of B-Netz telephones were significantly improved. The housing of the transmitting and receiving system became significantly smaller and cheaper. This is primarily due to the increased use of microcontrollers, which became available at the end of the 1970s.

IMTS

The American car telephone (Mobile Telephone System, MTS) was also further developed in the United States, using similar methods to the B-network. After all, the principles of signaling came from Bell Laboratories.

The new system introduced in 1964 was called the Improved Mobile Telephone System (IMTS). In contrast to the A and B networks, there were three bands in the USA.

• 35 MHz – 44 MHz         with 9 channels
• 152 MHz – 158 MHz     with 11 channels
• 454 MHz – 460 MHz     with 12 channels

Devices such as the Motorola TLD 1100 were not as advanced in appearance as the B-Netz telephone from TeKaDe. There was no display of phone numbers yet and dialing was done with a dial, later with buttons.

IMTS providers were the local telephone operators in the various cities. As with the A and B networks, capacity problems quickly came up. On the one hand, only one participant could make a telephone call per channel, and on the other hand, the possible number of telephones was technically limited. So there were waiting times to even get to a car telephone. Since the market was small, there were no great incentives for manufacturers like Motorola to build inexpensive devices. The use of mobile phones was only reserved for a small group of users.