Presented by: Group 2

 Two-level PSK (BPSK)  Uses two phases to represent binary digits Where we can consider the above two functions to be multiplied by +1 and -1 for a binary 1 and binary 0 respectively which equals

 Differential PSK (DPSK)  Phase shift with reference to previous bit ▪ Binary 0 – signal burst of same phase as previous signal burst ▪ Binary 1 – signal burst of opposite phase to previous signal burst  The term differential is used because the phase shift is with reference to the previous bit  Doesn’t require an accurate receiver oscillator matched with the transmitter for the phase information but obviously depends to the preceding phase (information bit) being received correctly.

 Four-level PSK (QPSK - quadrature PSK)  Each element represents more than one bit

I stream (in-phase) Q stream (quadrature data stream)

 OQPSK has phase transitions between every half- bit time that never exceeds 90 degrees (π/2 radians)  Results in much less amplitude variation of the bandwidth-limited carrier  BER is the same as QPSK  When amplified, QPSK results in significant bandwidth expansion, whereas OQPSK has much less bandwidth expansion especially if the channel has non-linear components

 Multilevel PSK  Using multiple phase angles with each angle having more than one amplitude, multiple signals elements can be achieved ▪ D = modulation rate, baud ▪ R = data rate, bps (note the difference in baud and bps) ▪ M = number of different signal elements = 2 L ▪ L = number of bits per signal element  If L = 4 bits in each signal element using M = 16 combinations of amplitude and phase, then if the data rate is 9600 bps, the line signaling speed/modulation rate is 2400 baud

 QAM is a combination of ASK and PSK  Two different signals sent simultaneously on the same carrier frequency

 Program 3.2 (bpsk_fading)

Amplitude DistortionAmplitude + Phase Distortion

Input  sr= ; % Symbol rate  ipoint=2^03; % Number of oversamples  ncc=1;  %******************* Filter initialization ********************   irfn=21; % Number of filter taps

 Number of symbols (nd) = 10  data=rand(1,nd)>0.5  data1=data.*2-1

 [data2] = oversamp( data1, nd, IPOINT)

 data3 = conv(data2,xh)

 demodata=data6 > 0

 Graphical eye pattern showing an example of two power levels in an OOK modulation scheme. Constant binary 1 and 0 levels are shown, as well as transistions from 0 to 1, 1 to 0, 0 to 1 to 0, and 1 to 0 to 1OOK Source: Wikipedia

 MSK is a continuous phase FSK (CPFSK) where the frequency changes occur at the carrier zero crossings.  MSK is unique due to the relationship between the frequency of a logic 0 and 1.  The difference between the frequencies is always ½ the data rate.  This is the minimum frequency spacing that allows 2 FSK signals to be coherently orthogonal.

 The baseband modulation starts with a bitstream of 0’s and 1’s and a bit-clock.  The baseband signal is generated by first transforming the 0/1 encoded bits into -1/1 using an NRZ filter.  This signal is then frequency modulated to produce the complete MSK signal.  The amount of overlap that occurs between bits will contribute to the inter-symbol interference (ISI).

 1200 bits/sec baseband MSK data signal  Frequency spacing = 600Hz a) NRZ data b) MSK signal

 Since the MSK signals are orthogonal and minimal distance, the spectrum can be more compact.  The detection scheme can take advantage of the orthogonal characteristics.  Low ISI (compared to GMSK)

 The fundamental problem with MSK is that the spectrum has side-lobes extending well above the data rate (See figure on next slide).  For wireless systems which require more efficient use of RF channel BW, it is necessary to reduce the energy of the upper side-lobes.  Solution – use a pre-modulation filter:  Straight-forward Approach: Low-Pass Filter  More Efficient/Realistic Approach: Gaussian Filter

 Gaussian Filter  Impulse response defined by a Gaussian Distribution – no overshoot or ringing (see lower figure)  BT – refers to the filter’s -3dB BW and data rate by:  Notice that a bit is spread over more than 1 bit period. This gives rise to ISI.  For BT=0.3, adjacent symbols will interfere with each other more than for BT=0.5  GMSK with BT=∞ is equivalent to MSK.  Trade-off between ISI and side-lobe suppression (top and bottom figures)  The higher the ISI, the more difficult the detection will become.

 An important application of GMSK is GSM, which is a time-division multiple-access system.  For this application, the BT is standardized at 0.3, which provides the best compromise between increased bandwidth occupancy and resistance to ISI.  Ninety-nine percent of the RF power of GMSK signals is specified to confine to 250kHz (+/- 25kHz margin from the signal), which means that the sidelobes need to be virtually zero outside this frequency band and the ISI should be negligible.

 The program bpsk.m prints the BER in each simulation loop, and this causes the program to run slowly, therefore, I stopped printing those results. Instead, I plotted the BER vs. EbN0 with a counter that displays the current value of EbN0.  I tried to plot the eye diagram for QPSK, but I didn’t succeed in that.

 Wikepedia.com  Haykin, S. 2001: “Communication Systems”. 4 th ed. New York, NY. John Wiley & Sons.  Introduction to GMSK  GMSK: Practical GMSK Data Transmission F/2003AUG29_NTEK_AN.PDF  Minimum Shift Keying: A Spectrally Effiecient Modulation iIntegrati/msk_pasupathy_1979.pdf