Key RF Measurements in Digital TV Systems

The secret to maintaining reliable, high-quality service across various digital TV transmission systems is to focus on key factors that can compromise the integrity of the system. This article describes key RF measurements that can help detect problems before viewers completely lose digital TV service and picture.

Compared to traditional analog TV, modern digital cable, satellite and terrestrial systems are very different, and their signals are susceptible to noise, distortion and interference in the line. Today's consumers have become accustomed to watching analog TV easily. If the picture quality deteriorates, people will usually adjust the indoor antenna to get a better picture. Even if the picture quality is still poor, if the program is attractive enough, viewers will usually continue to watch as long as they can still hear the sound.

Digital TV is not that simple. Once reception is lost, it is not always obvious how to recover. The problem may be caused by an MPEG SI or PSIP table error, or simply by the RF power being too low to reach the digital operating threshold or "spike" point. RF problems may include any of the following: satellite dish or low noise block converter (LNB) problems, terrestrial RF signal reflections, poor noise performance or channel interference, in addition to a faulty cable amplifier or modulator.

There are several approaches to solving digital TV reception problems. One solution is to reduce the sensitivity of the set-top box receiver to signal quality, but the more fundamental solution is for operators to maintain clean and high-quality RF signals. To ensure this, Tektronix provides key RF measurement capabilities, integrating real-time MPEG monitoring and recording functions in a single MTM400 instrument. These instruments can be economically deployed at various locations throughout the transmission chain from downlink and decoding to multiplexing and remultiplexing, and finally program distribution through uplink, headend and transmitter sites. With the MTM400, operators can make critical RF measurements at a fraction of the cost of dedicated RF test equipment. Web-based remote control allows the correct measurements to be made at the appropriate signal layer in the transmission chain, ensuring cost-effective results.

Bit Error Rate (BER)

The bit error rate is the ratio of the number of bits that have errors to the total number of bits transmitted. Early digital TV monitoring receivers provided a bit error rate indication as the only measure of the quality of the digital signal. This was easy to do because the data was usually provided by the tuner demodulator chipset and was easily processed. However, the tuner may often output the BER after performing forward error correction (FEC). A better approach is to measure the BER before FEC, which gives an indication of how well the FEC is working. After the Viterbi deinterleaving process, Reed-Solomon (RS) decoding will correct the error bits to give a quasi-error-free signal at the output.

This approach is feasible when the transmission system operates far from the peak point, when only few data errors occur and the bit error rate before Viterbi is close to zero. As the system approaches the peak, the bit error rate before Viterbi increases gradually, the bit error rate after Viterbi increases rapidly, and the bit error rate after FEC (after RS) increases sharply. Therefore, FEC has the effect of sharpening the corners of the peak. As a result, very sensitive bit error rate measurements can give an alarm, but it is usually too late to take any corrective measures. Nevertheless, it is still useful to display the BER to record or quantify the quality of the transmitted signal. The BER can also be used to record long-term system trends. It is best used to identify periodic, short-term signal defects.

BER measurement results are usually expressed in engineering terms and are often displayed as an instantaneous ratio and an average ratio. The typical target value is 1E-09, the quasi-error-free BER is 2E-04, the critical BER is 1E-03, and a BER greater than 1E-03 will result in loss of service.

How to improve BER --- using MER

The TR 101 290 standard describes measurement criteria for digital television systems. The modulation error ratio (MER) measurement is designed to provide a single quality factor for the received signal. MER provides an early indication of the receiver's ability to correctly decode the transmitted signal. In practice, MER compares the actual position of a received symbol (representing a digital value in the modulation pattern) with its ideal position. When the signal quality degrades, the received symbol is further away from the ideal position and the MER measurement decreases.

As the signal quality degrades, symbols are eventually decoded incorrectly and the bit error rate increases, reaching the threshold or peak point. The plot shown in Figure 1 was obtained by connecting a MER receiver to a test modulator. Once connected, noise was introduced gradually and the MER and pre-Viterbi BER values ​​were recorded. The initial MER without additive noise was 35dB, at which point the BER was close to zero. It is noteworthy that as the noise increased, the MER gradually decreased while the BER remained constant. When the MER reached 26dB, the BER began to increase, indicating that it was approaching the peak point. The MER shows that the signal quality of the system was degrading long before the peak point was reached.

Importance of MER

Since Tektronix equipment can measure very high extreme MER values ​​(typically 39dB in QAM systems), monitoring equipment located at the front-end modulator can provide early indications of signal quality degradation when the MER degradation factor (safety margin) of the downstream signal flow is known or can be measured at or near the user. When the MER drops to 24dB (for 64-QAM) or 30dB (for 256-QAM), ordinary set-top boxes may not demodulate correctly or may not work. Other ordinary measurement equipment with lower extreme MER measurement capabilities will not give early warnings of signal quality degradation. The typical extreme MER of the cable (QAM) front end is 35-37dB. The typical MER in analog cable systems is 45dB. The data difference between analog and digital systems is 10dB, so the digital MER in the transmission system is about 35dB.

Error Vector Magnitude EVM

The EVM measurement is similar to MER, but expressed differently. EVM is expressed as a percentage of the RMS error vector magnitude to the maximum symbol magnitude. As signal imperfections increase, EVM will increase, while MER will decrease. MER and EVM can be converted to each other. EVM is the distance between the detected carrier and its theoretically accurate position on the IQ (in-phase and quadrature) constellation diagram (see Figure 3), and is the ratio of the "error signal vector" to the "maximum signal magnitude", expressed as a percentage of RMS. EVM is defined in an annex to TR 101 290. The Tektronix MTM400 provides both MER and EVM measurement capabilities.

Variations of modulation schemes and systems

Signals in satellite, cable, and terrestrial digital television transmission systems use a quadrature modulation scheme that represents data symbols by modulating phase and amplitude. The most commonly used modulation schemes in digital television transmission are variations of quadrature amplitude modulation (QAM). For example, among the commonly used terrestrial digital modulation schemes, COFDM uses 16-QAM or 64-QAM, and 8VSB uses an 8-column system. The digital modulation scheme used in satellite digital television systems is QPSK (quadrature phase shift keying), which is equivalent to 4-QAM. QPSK is a very robust modulation scheme and has been used for many years. QPSK also makes more efficient use of the available bandwidth, but requires a higher carrier-to-noise ratio.

Cable digital television systems build on this foundation with a wider range of modulation schemes that are still evolving. Other modulation orders (16-QAM, 64-QAM, 256-QAM, and 1024-QAM) improve spectral efficiency, thus providing more channels in a given bandwidth.

In the US digital TV system, the transmission rate of 64-QAM can reach 27Mb/s, which is equivalent to transmitting 6 to 10 SD channels or 1 HD channel in a 6MHz bandwidth. New compression technology can provide up to 3 HD channels on 256-QAM. In the European system, 8MHz bandwidth can achieve a transmission rate of 56Mb/s on QAM-256.

ITU.J83 specifies three regional QAM cable standards:

* Annex A - Europe

* Annex B - North America

* Nearby C - Asia

In addition to being able to measure QPSK for satellite applications, the MTM400 has the ability to perform RF interface and measurements with all of the above-mentioned QAM standards.

Constellation display

The constellation display is the digital equivalent of a vectorscope display and shows the in-phase (I) and quadrature (Q) components of a QAM signal. A symbol is the smallest information component transmitted in a particular modulation system. For QAM-64, a symbol represents 6 bits and is plotted as a dot on the graph. These symbol bits go through a complex transcoding process from the original MPEG-2 transport stream. This process includes Reed-Solomon encoding, interleaving, randomization, trellising for QAM Annex B systems, and convolutional (Viterbi) encoding for QPSK systems. The purpose is to protect and correct bit errors, provide immunity to burst noise, and evenly distribute energy in the spectrum. After the process is reversed in the decoder, a quasi-error-free bit stream must be reconstructed. Because of this error correction process, simply examining the transport stream will not give any indication that the channel or the modulator and processing amplifiers are introducing errors, bringing the system closer to the "digital spike point." By the time the Transmission Error Flags (TEFs) begin to be reported in the MPEG stream, it is usually too late to take any error correction measures.

Constellation map

The constellation diagram can be viewed as a "two-dimensional eye diagram" array of digital signals, with reasonable limits or decision boundaries for the location of the symbols in the diagram. The closer the points representing the received symbols are in the diagram, the higher the signal quality. Since the graphics on the screen correspond to amplitude and phase, the shape of the array can be used to analyze and determine many defects and distortions in the system or channel, and help find their causes.

Constellation diagrams are useful for identifying the following modulation problems:

* Amplitude imbalance

* Orthogonality error

* Correlated interference* Phase noise, amplitude noise

* Phase error

* Modulation Error Ratio

Remote constellation diagram

The MTM400 uses web-based technology, so it is unique in that the constellation diagram can be viewed from a different location or even a different country, through the Internet or a dedicated network to an unattended test probe location. The user interface also has adjustable persistence, which can fade spots in the early received carriers, just like on traditional instruments. Note: The following MTM400 screen graphics are from instruments with test settings that make the MER and EVM displays similar, only the constellation diagram display is different.

Orthogonality Error

Quadrature error causes the symbols to be positioned closer to the boundary limits in the diagram, thus reducing the noise margin. Quadrature error occurs when the I and Q phases are not exactly 90 degrees apart. The result is that the constellation diagram is no longer square, but looks like a parallelogram or a diamond.

Noise Error

Noise is the most common and unavoidable impairment in any signal, including QAM. Additive White Gaussian Noise (AWGN) is a common type of noise impairment. Because it is white (a flat power density function in frequency) and Gaussian (mathematically a "normal" amplitude density), the received symbols are distributed around the ideal position.

Gain Compression

The MTM400's graphic signal display allows the operator to observe the gain compression phenomenon that causes the corner edges to be smoothed on the I and Q axes, but this only occurs when the modulator or fiber transmission system approaches its limit. At this time, the signal amplitude is high and nonlinear. When gain compression occurs, the graphic display looks like a "hemisphere" or "fish eye".

Related interference

When correlated interference occurs, the channel interference or harmonic components are phase-locked to the IQ signal. In this case, the graphical display is a set of rings, or "doughnut" shapes.

Phase noise (I, Q jitter)

Any carrier source or local oscillator in the signal chain generates phase noise or phase jitter that is superimposed on the received signal. Phase noise appears as concentric arcs around the carrier symbol.

Acceptable signal

In modern all-digital modulators, IQ gain and phase errors are usually negligible. These errors do not occur due to misalignment but rather due to device failure. On the other hand, compression may also occur in the modulator, upconverter, and transmission network.

Conclusion

It is better to anticipate system problems before digital TV service fails than to wait until problems occur and then try to solve them. MER measures small changes in transmitter and system performance and is one of the best quality factors in any cable and satellite transmission system. EVM and the more traditional BER are useful for standard cross-equipment checks and to help identify short-term signal degradation. Constellation diagrams can indicate defects, distortions or equipment deviations, helping to provide a reliable "sanity check" for RF transmission systems. By combining these key RF measurements with comprehensive MPEG transport stream monitoring and alarm capabilities in a single probe, digital TV operators can detect system problems at an early stage before they affect viewers' viewing. With the MTM400, Tektronix now offers all the key RF measurements and interfaces, and integrates MPEG measurements in a single cost-effective monitoring probe.

References

International Telecommunications Union, ITU-T J.83, Series J: Digital multi-program system for television, sound and data services for cable distribution (04/97).

Measurement Guidelines for DVB systems, ETSI Technical Report, TR101 290 V1.2.1 (2001-05) Digital Video Broadcasting (DVB); European Telecommunications Standard Institute.

Digital Video Broadcasting (DVB); Framing structure, channel coding and modulation for cable systems EN 300 429 V1.2.1

(1998-04) European Standard (Telecommunications series).