This article was first produced in Digital News - The Newsletter of the Digital Television Group November 98
|Digital Terrestrial - Signal Measurement|
A Low Cost Meter for Digital Terrestrial Signal Measurement.
C Swires, FSCTE, Swires Research
The DTG Reception Working Group recognised the need for a simple and cheap test unit to be used by installers to determine whether or not DTT reception is likely to be satisfactory at any particular aerial outlet, and how many of the six multiplexers will be receivable. The "blue skies" aim is for a box with six LEDs, which would light up green, amber or red, when the signal from an aerial is connected, and for the box to cost only a few pounds. The unit described below doesn't meet this perhaps impossible "holy grail", but is modestly priced and can give fairly accurate indications of DTT receivability.
The new IMdigital meter follows from the experience gained with the Swires IM96, widely used by the cable TV installers.
A key requirement was for a unit, which was simple to operate by non-technical personnel. However, at the same time giving accurate readings. The complete tuner was built in a screened compartment as an integral part of the main processor board. This reduced the interconnection cost. By using a microprocessor based design the linearity of the measuring circuits is corrected in the software, avoiding complex alignment procedures. New challenges have emerged with the introduction of digital television. Merely measuring the level of a digital signal is not sufficient. The quality of the signal must be shown together with an adequate margin before one can be confident of a 'no-recall' situation.
An accurate measurement of signal levels is very important to ensure that an adequate signal level is available on the aerial, it is also important that the digital receiver is not overloaded as this will cause distortion in the input stages of the tuner unit resulting in signal failure.
The overload problem is especially serious due to the presence of adjacent analogue signals on the tuner input at higher levels than the digital channel. This is frequently made worse by the use of pre-amplifiers,
which raise both the signal and the noise.
The most obvious way to measure the quality of a digital transmission is to measure its bit error rate, however in order to do this the signal must be de-modulated. Which at present requires expensive power hungry chip sets, which are not suited to low cost battery operated meters.
In practice, the main cause of degradation of bit error rate (BER) in the transmission medium is thermal noise. A large number of tests by a number of research groups have related noise to BER. The graph below shows the relationship between thermal noise and BER for a COFDM signal of 64 QAM.
The graph shows that the difference in signal to noise ratio between the maximum error rate of 1E-4 and 1E-9 is only a signal to noise ratio
difference of 4.2 dB, so even a good BER of 1E-9 only shows that we have a noise margin of greater than 4 dB. If signal to noise is measured, a much greater range can be obtained showing the total noise margin before failure will occur.
A great deal of research has shown that by far the primary cause of degradation of BER in transmission is due to noise and that the relationship between signal to noise ratio and BER is reliable.
Measuring the signal to noise ratio is the quickest and most meaningful way of knowing the margins available from the transmission.
The above considerations resulted in the choice of signal to noise ratio as the method used to define the quality of digital terrestrial signals in the IMdigital. In order to make this measurement easy the process is fully automated. The instrument selects a suitable noise floor by a scanning process. The signal to noise is presented but in addition, the digital readout gives the status as PASS, MARGINAL or FAIL.
Field tests carried out to date have shown the signal to noise measurement correctly predicts the point of failure when using digital signals.