Plants and construction sites need a radio technology which is easy to install and maintain. If several square kilometres or parts of an operation in different locations do not need to be covered, then TETRA equipment seems quite expensive. This is where DMR (Digital Mobile Radio) comes into play. Originally developed by Motorola and branded MOTOTRBO, it has since become an ETSI standard and is being marketed by many suppliers at appropriately attractive prices.
DMR has come to fill the gap between commercial mobile phone technology (GSM), trunked radio (TETRA) and simple walkie-talkies. This technology is ideal for construction sites, small and medium-sized companies in need of short call setup times and group calls with a few users but on their own frequencies. Due to the call setup time, commercial mobile phones and cordless telephones are out of the question. Four channels as with TETRA are often over-dimensioned, and these users do not want to pay the overhead for the theoretical possibility of conducting calls into the fixed-line network. Their concern is the pragmatic possibility for short, quick communication across a manageable area, rather than encryption and permanent monitoring and recording of communications. Status messages and a manageable quantity of data transmission are also desired for some applications and possible with DMR.
Mode of operation
DMR offers several different operating modes. The common denominator is a system with two time slots of 30ms duration each, in which, 4,800 symbols per second can be sent in 4FSK modulation.
The channel bandwidth of 12.5KHz suits common European channel spacing below 1GHz, and thus there are applications on various licensed and free bands starting at 68MHz. The transmission power of the terminal devices is fixed and is typically one watt or less. With digital voice transmission and encryption on the air interface, listening in is usually made difficult.
This technical benchmark data forms the basis for different variants: the direct mode, communication via a repeater, andcommunication via a base station. In direct mode, the devices transmit and receive on a single frequency; in simplex mode, only one time slot is used in alternation; in duplex operation, both are used simultaneously. The timing on the radio interface is specified by the radio device currently transmitting. The channels are operated in TDD (Time Division Duplex), both time slots are 30ms long.
A DMR repeater can be used for difficult radio connections. The repeater and the terminal devices transmit on different frequencies: FDD is used, with 4.6 to 10MHz duplex spacing. The radio that wishes to talk puts the repeater into transmit mode through an activation burst; the temporal burst sequence is then specified by the continuous signal of the repeater. Regardless of whether a duplex conversation or communication in simplex mode is ongoing, the base station transmits in both time slots.
If the system reaches its limits even with a repeater, then it can be expanded to include base stations, which further expand the range of the radio communication. Here too, frequency duplex comes into play. Per channel pair, however, there are still only two time slots available on a carrier, and the number of simultaneous conversations remains limited. Communication between the base stations is not standardised; for small systems with few base stations, the customer will rely on a single supplier.
During development, in production and during the testing of the devices between missions, various types of transmitter and receiver measurements are performed. Parameters which are always important for the transmitter of mobile radio devices are transmit power, frequency, and modulation errors.
For the duration of the time slot (burst), during the user data transmission, the transmit power must remain constant; the information is transmitted by changes of the frequency. Due to the transmission in time slot, not only is the power in the steady state relevant, but also the transients while keying in the signal. If the increase or drop-off of the transmit power is too flat, for example, then this can result in interference of the signal in the neighbouring time slot. The graphic display of the burst profile facilitates a check.
There is a particularity for the power-related measurements: in the magnitude error measurements, deviations from the average output power are broken down according to symbol values, that is, according to the frequency deviation at the point of maximum effect of the symbol. Therefore, it is possible to establish a relationship between the respective frequency deviation and a deviation of the transmission power.
Potential interference with the neighbouring channels can be determined via the frequency error measurement. Modulation measurements serve to check the signal quality, which should fulfil the requirements in order to achieve coverage and immunity to interference. At an interval of 1/4,800 seconds, DMR-specific 4FSK modulation generates frequency deviations of theoretically ±0.648 or ±1.944KHz, depending on the symbol transmitted; a symbol, in turn, contains two bits of information. The modulation is regarded at the defined symbol times. The most significant measurement parameter here is the FSK error; at the symbol times the difference between the actual and theoretical frequency deviation is measured, divided by the nominal frequency deviation, and the square values are added up (RMS value). An FSK error of up to 5% is permissible. An optical, qualitative check is possible with the eye diagram (see Fig.) – the less the signal is dispersed in the symbol times, the better.
Another criterion for the modulation quality is the symbol clock error. Here, a trend in the temporal deviation of the symbol midpoint from the nominal value is regarded. The symbol clock error is measured in Millihertz; A significant value may be caused by an imprecise modulation frequency (target: 4,800Hz). Up to 48MHz is tolerable, however.
A technical particularity of DMR is its support of two essentially different duplex processes. In direct mode, the radio device is dependent on its own frequency stability and may not exceed a particular frequency and symbol clock error. In repeater mode, however, the radio must synchronise up to the frequency and timing of the receiver and may even have to switch cyclically between transmission and reception frequencies. For the measurements, this means essentially that the transmission quality must be tested in two different operating modes, that is, in the worst case, all measurements must be carried out in duplicate. The manufacturer’s specifications help to reduce the test time, for who knows better in which operating mode the radio design is most critical or which measurements do not have to be repeated. Experience in testing with various modes in commercial mobile phones also shows that the switching time between the modes (e.g. GSM – WCDMA – HSPA) has great influence on the overall test time.
For many measurements, no special test mode is required; nevertheless, the attempt of a measurement may sometimes fail. This is often due to the fact that the valid colour code for the simulated communication between the device under test and the radio test set is set incorrectly. The colour code must be the same for all devices belonging to a system, so the value in the test set must be changed to the value programmed in the device under test.
Measuring the receiver
As for all digital communication systems, so for DMR, the quality of the receiver is determined using statistical measurements of the bit error rate. In direct mode (simplex operation), for example, the measuring device repeatedly sends a longer, defined bit sequence. For sensitivity measurements on the lower limit, the test set’s output level is set to a low value. The radio synchronises itself to the (known) bit sequence, counts the incorrectly received bits and outputs the bit error rate calculated from this. For this, the radio must be put into a test mode. In contrast to most standards, the DMR standard does not define loopback of the signal with measurement of the bit error rate in the measuring device. The reason for this is that there are radios which are designed only for simplex operation, so they cannot send the signal received directly back to the measuring device.
Its master’s voice
The DMR standard supports a special test mode for audio measurements for which a tone of 1031Hz is transmitted. Thus, only the audio reception part can be tested. Therefore, a radio test set should also have the capability of looping the audio signal received back to the terminal. In combination with the test mode for audio measurements, errors such as signal distortions can then not only be determined, but their cause can be localised.
The repeater should also be tested from time to time, especially when the whole DMR is of mobile nature, i.e. even the repeater is transported between alternating locations. A test after the setup is recommended in order to be able to expose transport damage before use.
Since the repeater sends continuously, only an activation signal is needed for the transmitter measurements. The repetition (i.e. the loopback) of the received signal is all but the actual task of the repeater, so this can be truly useful for receiver measurements.
Measurements on digital communication systems such as DMR cannot be carried out with any cobbled-together signal generators and analyzers; rather, a radio test set is required that is familiar with the system parameters (channels, colour code) and speaks the necessary signalling protocol in order to establish a connection for receiver measurements. In contrast to analogue radios, the receiver cannot be tested quantitatively via an audio measurement, but via the bit error rate measurement. The number of different measurements is manageable for the average test task; deeper-reaching measurements combined with the corresponding experience with the radio device type enable researching and eliminating the cause.