MULTI-LEVEL QAM & PSK
Contents
1. could identify the signals at TRUNKS using only an oscilloscope But you could identify them using a lengthy trial and error process by setting up the demodulator to be described and modelled in what follows However to save time their descriptions and locations at TRUNKS will be defined by your Laboratory Manager 118 D2 2 in detail in BER measurement in the noisy channel in this Volume Multi level QAM amp PSK the noisy channel To save rack space it might be thought that the channel filtering could be included at the transmitter end of the system But it is necessary to add noise at the input to the channel so that the noise will suffer the same bandlimiting as the signal For an alternative method see Tutorial Question Q3 The complete receiver is shown modelled in Figure 4 below MULTIPLIER TUMEABLE oc Pac x m QAM IN bit clk Figure 4 model of the m QAM demodulator and decoder The error counting instrumentation is shown as a BER INSTRUMENT module This is described in the experiment entitled BER instrumentation macro model this Volume T1 patch up the system of Figure 4 Be sure to connect the cos and sin outputs from the MASTER SIGNALS module as shown since this matches their usage at the transmitter and so preserves relative carrier phasing T2 select the bandpass channel from the 100 kHz CHANNEL FILTERS module T3 select the 4 QAM signal from TRUNKS Connect it to the inp2. MULTI LEVEL QAM amp PSK PREPARATION en aE Sagan tan eaten att ct scat S 116 the m QAM cenerators caccsesv2o essa eine eelae eas tcaaeasd 116 the M QAM demodulator ccccceccsscecssscecssscecsseeeesseeeesseeees 117 carrier amp bit COCKS ws erein ee E N EE i EA 117 Carrier PHASING geinn e E R RE aE E EE ER 117 decoder and error counting eeseseseesiereersssstststtersesessesesetereesssesesee 118 POMPE SITY een ee i e A N R E 118 the noisy channel iposi ct ca saleat teu scearetotouatadaneeecoeventanenaucasetne 119 EXPERIMEN T a e a E E teenth S 119 noise and BER Mmeasurenient gacssthacancuncmnenieana ees 121 further experiments ssessessesseseesseessessreseossessesressesseeseesressee 122 bandwidths inna A AE A E A ieee tds 122 POM venae i eE a aE EE ued BS E AE EEE EE eb hee 122 TUTORIAL QUESTIONS sssssesssessesessssesessresesessssesesrssssesesresesesssses 122 Multi level QAM amp PSK Vol D2 ch 13 rev 1 0 115 MULTI LEVEL QAM amp PSK ACHIEVEMENTS demodulation of various m QAM signals including 16 PSK and 16 QAM Detection of errors PREREQUISITES completion of the experiment entitled Signal constellations in volume D1 Completion of the experiment entitled QAM and 4 PSK this Volume would be an advantage ADVANCED MODULES M LEVEL DECODER ERROR COUNTING UTILITIES NOISE GENERATOR WIDEBAND TRUE RMS METER EXTRA MODULES a total of two MULTIPLIER two TUNEABLE LPF and two ADDER mod
3. al for details and or seek the advice of your Laboratory Manager These signals are the actual inputs to the decoder circuitry You are now ready for BER measurements Detailed instructions are outside the scope of this introductory experiment Multi level QAM amp PSK p2 121 further experiments bandwidth So far no mention has been made of the bandwidth of these signals Further experiments at the very least should be devised to compare the inter relationships between channel bandwidth data rates and error rates PCM Instead of using a SEQUENCE GENERATOR as a message source a PCM ENCODER could be used This would allow the transmission of an analog message over the system TUTORIAL QUESTIONS Q1 there are two waveforms i and q and so two decision points but the DECISION POINT control moves both points together What does this say about the properties of the lowpass filters in the MULTIPLIER outputs Q2 the method of carrier phase adjustment was made using a single PHASE CHANGER in the input signal path to the demodulator Was this reasonable This method was implemented due to the shortage of free slots in a single TIMS 301 rack The method illustrated in Figure 2 would require two slots whereas the suggested method only one Q3 to save rack slots at the receiver the channel filter could be placed at the transmitter Noise could still be added at the receiver input This would save a single rack slot If the ch
4. annel filter bandwidth was wider than the lowpass filters at the demodulator output would not this be an acceptable method of bandlimiting the noise Compare the two methods What problems might there be in each 122 p2 Multi level QAM amp PSK
5. ate Figure 3 the two phasing schemes decoder and error counting The decoding is performed by the M LEVEL DECODER module acting on the i and q outputs from the demodulator Correct alignment of the system which requires carrier phase adjustment is checked by counting errors The technique of error counting using a reference sequence at the receiver has been examined in detail in previous experiments 2 In particular and of direct relevance to this experiment setting up for error free transmission using multi level signalling and error detection instrumentation was described in the experiment entitled Signal constellations Volume D1 complexity To build a complete system encoder modulator noisy channel demodulator decoder and bit error rate measuring facilities requires slots for more than 12 modules The TIMS 301 rack provides 12 slots So in this experiment only the receiver will be modelled and examined Several signals will be available at TRUNKS These will be defined by your Laboratory Manager and will be used for testing your receiver All are based on 100 kHz carriers and 8 333 kHz bit clocks Their messages will be derived from SEQUENCE GENERATOR modules similar to your own and of stated sequence lengths So a reference message can be created at the receiver and a sliding window correlator used for lining it up with the decoded sequence as has been done in previous experiments It is unlikely that you
6. ese these connections duplicate those at the transmitter Since the properties of the input signal are known you already know how many levels to expect in the eye pattern For 4 QAM there should be 2 levels in each of the i and q paths If there are more than 2 levels what adjustments are available to you Only one Think about it then proceed T7 adjust the phasing of the demodulator PHASE CHANGER until there are the expected number of levels in the eye pattern two levels for 4 OAM There will be four phases which will give this result two positions of the front panel variable COARSE control and two with the 1 80 toggle switch How many will be correct How do they differ This cannot be decided yet T8 accept any one of the four phases of the previous Task T9 using the gain controls of the TUNEABLE LPF modules adjust the levels of the signals to the i and q inputs of the M LEVEL DECODER to result in 0 to 5 volt signals at the I and Q outputs T10 set up the M LEVEL DECODER front panel toggle switches to accept the coding scheme of the TRUNKS signal T11 attempt to align the DATA output from the M LEVEL DECODER with the X output from the reference SEQUENCE GENERATOR If there are continuous errors the only system error must be the demodulator carrier phasing T12 if there are errors adjust to another phase and repeat the previous Task One and only one of the four possible phases should give error free trans
7. idetracked by having to model carrier and bit clock recovery schemes These have already been examined in the experiments entitled Bit clock regeneration and Carrier acquisition both in this Volume and Carrier acquisition and the PLL and The Costas loop both in Volume A2 phasing The local quadrature carriers to the two MULTIPLIER modules will be obtained from the MASTER SIGNALS module These two will be connected with the same relative phasing as those at the transmitter This will save much time in investigating the many phase combinations possible at the receiver only one of which will satisfy the condition for error free decoding The only phase adjustment will be introduced by a single PHASE SHIFTER inserted in series with the bandpass input to the modulator The PHASE SHIFTER module is a narrow band device introducing different amounts of phase at different frequencies However the QAM signal itself is also narrow band and so this use of the PHASE SHIFTER is not entirely inappropriate See Tutorial Question Q2 Once error free decoding has been achieved the correct conditions at the demodulator are known and the approximate single PHASE CHANGER scheme could be replaced with a precise method namely a PHASE CHANGER in each carrier path These two schemes are illustrated in block diagram form in Figure 3 below Multi level QAM amp PSK 1 this differs from the method shown in F igure 2 p2 117 approxim
8. mission 120 n2 Multi level QAM amp PSK T13 examine the signal constellation of the demodulated signal So far only the 4 QAM format has been decoded T14 examine the other signals at TRUNKS In each case before error free decoding can be assured it will be necessary to a set the reference SEQUENCE GENERATOR to the same sequence length as that at the transmitter b adjust the decision point on the eye diagram c check for the correct number of levels in the eye diagram and make a fine adjustment of the phasing if necessary d repeat Tasks T7 to T13 noise and BER measurement The system is now ready for the addition of system impairments in the form of noise Follow the procedures described in detail in the experiment entitled BER measurements in the noisy channel this Volume Before making these measurements however now is the time to fine tune the system The effectiveness of these adjustments can be judged by aiming for an improvement of the bit error rate They should be made with a moderate amount of added noise Note that if only 12 slots are available in a single TIMS 301 rack and 13 are required it is possible to share one slot between the WIDEBAND TRUE RMS METER and the ERROR COUNTING UTILITIES modules These two are not required simultaneously T15 trim the DC offsets at the Q and I outputs of the M LEVEL DECODER module first read the appropriate section of the Advanced Modules User Manu
9. ules TRUNKS see your Laboratory Manager for details of signals at TRUNKS PREPARATION This experiment is about the use of the quadrature amplitude demodulator for the demodulation of m QAM and m PSK signals A modulator which produces quadrature amplitude modulated QAM signals is examined in the analog experiment entitled Phase division multiplex Volume A2 and a digital experiment entitled QAM and 4 PSK this Volume The more general m QAM and m PSK signals will be examined in this experiment the m QAM generator Figure 1 shows the block diagram of a multi level encoder and a quadrature modulator The first of these has been examined in the experiment entitled Signal constellations Volume D1 which it is assumed you have completed That experiment also examined the multi level decoder serial serial to in parallel output converter encoder Pit n ne quadrature modulator Figure 1 a multi level QAM modulator 116 D2 Multi level QAM amp PSK the m QAM demodulator Figure 2 shows the block diagram of a quadrature demodulator and a multi level decoder quadrature demodulator _ _ _ _ Pid ee decoder parallel to serial serial converter out baseband carrier carrier Figure 2 an m QAM demodulator amp bit clocks Stolen carriers and bit clocks will be used at the demodulator As usual while pursuing a major topic one does not wish to become s
10. ut of your receiver T4 initially introduce no noise to the input ADDER Adjust the gain through this ADDER to give a TIMS ANALOG REFERENCE LEVEL signal 4 volts peak to peak into the quadrature demodulator T5 connect the Z modulation output signal to the Z modulation input of the oscilloscope Multi level QAM amp PSK p2 119 The eye pattern is first used to adjust the decision instant displayed as a bright spot on the oscilloscope trace vary the intensity of the display for best spot visibility The front panel DECISION POINT control on the M LEVEL DECODER allows adjustment over a little more than one clock period If it cannot be located at the desired point step it across by pressing the HUNT button For 4 QAM the next press of the button will return to the previous condition there are two clock periods per frame but for 16 QAM for example there are four clock periods per frame and so four accessible regions each of which can be reached by repeated presses of the button Use the same techniques as are described in the experiment entitled Detection with the DECISION MAKER Volume D1 See Tutorial Question Q1 T6 look at the eye pattern at either LPF output of the demodulator Even though two carriers may be available in phase quadrature there are many combinations in which they may be connected to the two multipliers for desired results Connection in the manner shown in the model of Figure 4 will eliminate many of th
Download Pdf Manuals
Related Search