1.

Basic parameters of a communication system (power, bandwidth, noises), capacity theorems (with examples) Basic parameters of a communication system Transmitted power: the transmission power of the message signal [Watts] Frequency bandwidth: the physical spectrum available for a certain transmission [Hz, bps] Noises: undesired signals distorting the useful signal (channel noise, receiver noise, interferences) Shannon used these terms in its famous capacity theorem Examples Bandwidth: 300 3400 Hz for PSTN networks (adapted to the human hears spectrum): restricted by regulations; 1.1 MHz for twisted pairs cables (such as in PSTN): restricted by physical features; 10 Mbps: total bandwidth available for an Ethernet transmission (100 Mbps for Fast Ethernet ): restricted by regulations and physical features; Power and noises SNR (signal-to-noise ratio):

Examples: 0dB means unitary SNR, 10 dB means that the signal is ten times stronger, 20 dB means 100 times more signal power than noise power etc.

2. Channels: noises Noises Definition: unwanted signals that are inserted somewhere between transmission and reception Four categories: Thermal noise Intermodulation noise Crosstalk Impulse noise

Thermal noise Thermal noise Due to thermal agitation of electrons Uniformly distributed in frequency Generally modeled as white noise The amount of thermal noise in 1Hz N 0 = kT k = 1.3803 10 23 [J / oK ] N0 is the power spectral density [Watts/Hz] The amount of thermal noise in W Hz N = kTW Other types of noises Intermodulation Produces components having frequencies f1+f2 and f1-f2 Caused by non-linearity of the transfer function Crosstalk A signal from one line is picked up by another Electrical coupling between nearby twisted pairs Impulse Irregular pulses or spikes e.g. External electromagnetic interference Short duration and high amplitude Important source of errors for the digital signals Crosstalk noise Far end and near end crosstalk Near end crosstalk predominates Near end crosstalk falls off with frequency

Other noises Three categories: fluctuation, oscillation and pulse noise Fluctuation noise: caused by the power supply networks, radio stations etc Uniformly distributed in the useful bandwidth Oscillation noise: parasite harmonics of 50Hz Pulse noise: issued from crosstalk (pulses transmitted in the neighbor lines) or because of the switches from the telephone exchange

3. Channels: attenuation, delay distortions, crosstalk, other impairments Attenuation Signal strength decreases with distance Depends on the transmission medium Received signal strength: must be enough to be detected must be sufficiently higher than noise to be received without error (Solution: repeaters, amplifiers) higher the transmission frequency, higher the attenuation is (mainly concerns the analog signals) Delay distortion Specific to guided media (wires) Signal propagation speed depends on the frequency Frequency selectivity arises: various frequency components of the signal will arrive at receiver with different delays A kind of Inter Symbol Interference (ISI) occurs Other distortions Frequency deviation of the oscillator from the receiver, compared to the transmitter Echoes: at the transitions between 2 wires and 4 wires Counteracted by echo suppressors (echo attenuations >19dB) Short duration cuts of the signal, caused by power supply back off activation, redundancy mechanisms in case of failure They are defined as a decrease of at least 6dB of the signal level, for a duration ranging from 3 to 300 ms 4. Channels capacity (Nyquist and Shannon theorems with computation examples) Channel capacity Definition: the rate at which data can be transmitted over a given communication path, under given conditions Four important concepts in defining capacity Data rate In bits per second Rate at which data can be communicated Bandwidth In Hertz Constrained by transmitter (regulations) and medium Noise Noise Bit Error Rate (BER)

Nyquist formulation For noise-free channels: W is the bandwidth, M is the number of signaling levels Question: What is the capacity of a telephone line modem that uses 8 signaling levels? Answer:

Shannon theorem Formulation This is the original Shannon formulation

Interpretation: What Shannon says? Reminder: Capacity [bps], W[Hz] Shannons formula expresses the theoretical maximum rate that can be achieved Shannon decoded: Give me enough bandwidth (power) and I will shake the world 5. Digital encoding: key terms (definition, uni/bi-polar encoding, differential vs absolute encoding, baseband transmission, data rate ) Digital encoding What is digital encoding anyway? Information is an abstract concept, the bit concept too When transmitting, we must face the physical reality Signals are used to figure and to transmit the bits Digital encoding = a set of rules used in order to map a sequence of bits to a signal Depending on the transmission environment, the signal can be an electrical signal, a radio wave (modulation) or a sequence of light pulses

Mostly, digital encoding refers to transmission through guided media (cables) Key terms Unipolar: all signal elements have same polarity (+ or -) Polar (bipolar): one logic state represented by positive voltage the other by negative voltage Mark stands for 1 and Space stands for 0 (from telegraphy) Differential encoding: the signal element(s) which encode a bit are a function of the previous bit representation E.g.: a bit of 1 determines a transition from the previous signal state Absolute encoding: the signal element(s) encoding a bit are only a function of the current bit to be encoded Baseband transmission: the signal occupies a spectrum which is located nearby the 0 frequency (and NOT centered to a high frequency carrier) Data rate

The modulation rate is different of the data rate (can be lower or higher) Defined as: Rate at which the signal level changes (the reversed of the duration for which the signal level stays constant) Measured in baud = signal elements per second 6. The objectives of the digital encoding Objectives of digital encoding [1] Spectral efficiency Lack of high frequencies reduces required bandwidth Concentrate power in the middle of the bandwidth No DC component: allows AC coupling via transformer, providing isolation Synchronization The receiver must have exactly the same time reference as the transmitter (beginning of the bit interval, bit duration etc) Sampling must be very precise, especially at high data rates External clock versus sync mechanism based on the received signal Objectives of digital encoding [2] Error detection Usually the task of the upper layers, but some mechanisms can be incorporated in the digital encoding strategy too (phy layer) Signal interference and noise immunity Some codes are better than others (in terms of BER versus SNR) Cost and complexity Higher signaling rate lead to higher costs Some codes require modulation rate greater than data rate The relation between the modulation and the data rate is critical

7. Digital encoding techniques (NRZ, Binary AMI, bi-phase): principles, versions, advantages and drawbacks NRZ codes Rule: bits are represented with constant levels, for the whole bit duration No voltage for 0, constant positive voltage for 1 (unipolar) In practice, negative voltage is assigned to 1 More often, bipolar version used (+ and pulses, zero voltage avoided) Used in slow speed communication interfaces (e.g. RS232) and storage media (e.g. magnetic tapes) NRZI is the differential version of NRZ (1 encoded by transition from the voltage level which represents the previous bit) NRZ pros and cons Pros Easy to engineer Makes good use of bandwidth: can achieve 2 bps per Hz of spectral efficiency Cons DC component (for the unipolar version) Lack of synchronization capability: long runs of identical bits mean no transition in the signal Multi-level binary codes Alternate Mark Inversion (AMI) and pseudoternary Three signal levels (0, +V and -V) AMI rule: 0 means no signal, 1 is represented by voltage pulses with alternating sign Pseudo-ternary is the opposite of AMI Used in first generation PCM equipments Improved versions of AMI are currently in use Pros and cons of multi-level binary codes Pros No loss of sync if a long string of ones (in AMI) Zero DC component by the nature of these codes Bandwidth efficiency (main amount of power concentrated in less bandwidth than in NRZ case) Simple mechanism for error detection (two consecutive pulses of the same sign can only be caused by an error) Cons: synchronization issues for long runs of 0 (AMI) or of 1 (pseudo-ternary) They introduce some kind of redundancy: three levels to encode two bits 3dB SNR loss versus the BER performance of NRZ Biphase codes

Manchester and differential Manchester Principle: supplementary transitions introduced compared to NRZ, AMI Manchester rule: there is always a transition in the middle of the bit period, and sometimes at the beginning E.g.: 1 is represented by a low-high transition at the middle of the bit interval, 0 is the reversed Inherently, a transition will be performed at the beginning of the bit, if the current bit its identical with the previous one Differential Manchester: always transition at the middle of the bit (clocking only, it doesnt meter the sense of the transition) Information is carried by the transition at the beginning of the bit: 0 means transition, 1 means no transition Manchester is used in IEEE 802.3 Ethernet Biphase codes pros and cons Pros Predictable transition at the middle of each bit, which can be used for synchronization (selfclocking codes) Error detection possible due to the same feature No DC component Cons Bandwidth efficiency issues (half the bandwidth efficiency of NRZ) Modulation rate can be two times the data rate

8. Improved versions of AMI (HDB-3, B8ZS): motivation, principles, advantages and drawbacks, examples. B8ZS Bipolar With 8 Zeros Substitution Can be considered as an improvement of the bipolar-AMI Rule: 8 consecutive zeros are NOT encoded with no signal for eight bit periods, a signal which has 4 transitions being used instead When an all-zero octet occur: encode as 000+-0-+ if last non-zero voltage pulse was positive encode as 000-+0+- if the last non zero pulse was a - AMI code rules are broken twice by polarity violation: once inside of the eightzeros group, once between the first non-zero pulse preceding the group and its correspondent within the group Unlikely to occur as a result of noise Receiver detects and interprets as octet of all zeros

HDB3 High-density Bipolar, order 3 Another improvement of the bipolar-AMI Rule: 4 consecutive zeros are NOT encoded with no signal for eight bit periods, but with a signal which has at least one transition When a group of 4 zeros occurs: encode as 000V if the number of bits of 1 after the last polarity violation is odd B00V if the number of bits of 1 after the last polarity violation is even V means the same polarity as the previous one, B respects AMI pattern This rule aims to remove the DC component (polarity violation always changes its sign) Scrambling pros and cons Pros: Classical drawbacks of AMI are eliminated No redundancy, therefore the same rate is maintained Error detection capabilities Good spectral efficiency Cons: Extra-processing needed at transmitter and, especially at receiver side B8ZS employed in 1.544 Mbps-T1, HDB-3 in 2.048 Mbps-E1

9. Baseband transmission chain (with explanations about the role of each block) Model for baseband data transmission

Legend: an: sequence of bits to be transmitted PAM: pulse amplitude modulator GE(), GR(): emission/reception filters (their transfer function) s(t): signal which carries data, transmitted in the channel C(): channels (physical environment) transfer function n(t): additive white noise r(t): received signal (based on it, a decision is made) Threshold comparison: e.g. positive value leads to 1, negative to 0 ean: estimation of the received bits (ideally, identical to an) 10. Nyquist theorem for ISI-less data transmission in the baseband (deduction)

Nyquist theorem

In a channel which is equivalent with an ideal low-pass filter having the cutoff frequency F, it is possible to transmit symbols with a modulation rate equal or less to 2F symbols/sec, without ISI The characteristics of such a channel are shown below

Graphical view of the Nyquist theorem

11. Nyquist filters (infer the mathematical criterion for the raised-cosine Nyquist filters) The raised-cosine family The previously formulated criteria is met if:

The impulse response will be:

12. Transmission with controlled ISI (motivation, the cosine filter: frequency and impulse response, precoding, decision) Transmission with controlled ISI Motivation: In practice, the raised cosine filters will not reach the Nyquist rate Closer they are to this objective, higher is the transmission sensitivity to synchronization errors Solution: Some degree of ISI can be tolerated, if the ISI is controlled Higher rates can be obtained The cosine filter allows to reach the Nyquist rate Principle: every waveform g(t) will carry two-bits of information This is the reason for calling this transmission duobinary

The cosine filter

Pre-coding Disadvantage of transmission with controlled ISI: every decision depends on two successive bits (error propagation) Solution: pre-coding performed Instead of ak, another sequence bk is transmitted, encoded as:

This leads to a one-sample based decision, as follows:

13. Modulation: key terms (carrier, message) with one graphical example for binary AM, FM, and PM. Modulations key terms Modulator signal: the signal to be transmitted The signal can be analog or digital Especially in the digital signal context, the modulator signal can be referred to as message Carrier signal: used to transport the message signal Carrier signal is a sine wave (continuous wave modulation) or a periodic rectangular wave

Amplitude modulation AM is referred to as linear modulation The amplitude of the carrier is changed by the signal to be transmitted When the message signal is digital, we obtain Amplitude Shift Keying (ASK) The simplest form of ASK is called On/Off Keying (OOK)

Frequency Modulation Definition: FM is a method used to transmit analog or digital signals, in which the information is carried by the instantaneous frequency of a high-frequency carrier Examples: FM radio (modulator=audio signal), FM modems (modulator=digital data) When the modulator is digital, the frequency modulation is referred to as frequency shift keying

Phase Modulation Definition: PM is a method used to transmit analog or digital signals, in which the information is carried by the initial phase of a high-frequency carrier PM is not so popular as FM, especially for analog signal modulators (because of its complexity) When the modulator is digital, the frequency modulation is referred to as phase shift keying (PSK)

14. AM types (mathematical model of the AM, AM with supressed carrier, SSB-AM, VSB-AM). AM Frequency domain view [2] The spectrum of the modulated signal is:

Suppressingand un-suppressing the carrier If the carrier has a DC component:

The modulated signal is:

In this case, in the spectrum of the AM signal, we will retrieve the Dirac pulses corresponding to the sine carrier

SSB-AM SSB signal can be expressed as:

is the Hilbert transform of g, which can be obtain by passing g(t) through the filter:

SSB-AM is spectrally efficient Difficult to implement in practice: the filters which separate the side-bands must be very selective AM with Vestigial Side-Band (VSB) Only part of the side-bands is suppressed Lower frequencies transmitted with both Side-Bands, upper frequencies with one side-band This allows easier filtering to separate the bands (frequencies near the carrier must not be filtered) 25% more bandwidth required than in SSB, but easier to implement Example: NTSC TV system: all upper sideband of bandwidth W2 = 4 MHz, but only W1 = 1.25 MHz of the lower sideband are transmitted

15. Coherent demodulation for AM signals (mathematical model, principles, the importance of the phase synchronization) Coherent demodulation [1]

Principle: locally generated carrier must be used in receiver This carrier must be synchronized (frequency, phase) with the one used for transmission The modulated signal can be expressed as:

The LPF cuts-off the 2 0 component, and r(t) is obtained 16. The Costas loop for receivers carrier synchronization

Comments on the previous scheme Costas loop: used by the coherent AM demodulators to regenerate the carrier It uses the received data signal If we assume that there is an initial phase shift , this shift is used to tune the VCO, via an error signal (e) The error signal is obtained from u(t) (eq. 10) and it is a kind of average value of x2(t), weighted by This average value is computed by the LPF (only the frequency components around 0 are maintained after the filter)

17. Frequency modulation: definition, principles, mathematical approach (formulas for frequency modulated signals, instantaneous frequency definition, continuous and discontinuous frequency modulation), advantages and drawbacks Introduction Definition: FM is a method used to transmit analog or digital signals, in which the information is carried by the instantaneous frequency of a high-frequency carrier Examples: FM radio (modulator=audio signal), FM modems (modulator=digital data) When the modulator is digital, the frequency modulation is referred to as frequency shift keying Mathematical approach Considering a rectangular form of the modulator signal:

g(t): rectangular pulse-shaper an: data sequence to be transmitted The FM modulated signal is:

Instantaneous frequency The instantaneous frequency represents the derivate of the phase:

FM: pros and cons Pros Higher resilience to additive noise and interference, compared to AM Possibility to use non-linear amplifiers to amplify the FM signals Non-coherent detection is an available option Cons: Much larger bandwidth than AM (e.g. 240 KHz vs 30 KHz for a 15 KHz audio channel)

18. Frequency demodulation using a narrowband limiter Demodulation Two types of demodulation: coherent and noncoherent A non-coherent demodulator is presented below:

BPF: eliminates out-of-band signals (noises, interference) Narrowband limiter: limits the peaks of the signal and than applies a band-pass filtering around the carrier 19. Frequency demodulation using a wideband limiter A different approach

The limiter is wide-band It transforms the modulated sine into rectangles (trapezoidal) waveforms The differentiator highlights the transitions in these waveforms The signal is rectified to obtain only positive impulses They command a constant-duration pulse generator, which generates pulses having the same duration, but different periods given by the bit value LPF averages the signal over one bit duration, thus transforming the frequency information into an amplitude one 20. Differential frequency demodulation Differential demodulator

The delayed version of the signal u(t) is:

After, the multiplier, we get:

21. Phase modulation: principles, mathematical approach, graphical examples, advantages and drawbacks. Introduction Definition: PM is a method used to transmit analog or digital signals, in which the information is carried by the initial phase of a high-frequency carrier PM is not so popular as FM, especially for analog signal modulators (because of its complexity) When the modulator is digital, the frequency modulation is referred to as phase shift keying (PSK) Mathematical approach Considering a rectangular form of the modulator signal, x(t):

g(t): rectangular pulse-shaper (pulse with amplitude 1 and duration T) an: data sequence to be transmitted (bits) If we associate a phase, n with each symbol an, we get :

x remains constant during the symbol interval (T) n is a phase value, associated with the symbol to be transmitted In the digital PM (PSK) we have a finite number of possible symbols, M M=2, corresponds to BPSK (e.g.: 90o phase shift corresponds to 1, -90o phase shift corresponds to 0) PM: pros and cons Pros Narrow bandwidth compared to FM Non-coherent detection, with small performance degradation Cons High complexity (especially for analog PM)

22. QPSK modulator (QPSK principles, modulators scheme, constellation, phase table, mathematical expression for the generated signals) QPSK Modulator For the BPSK case, the modulation is implemented by simply inversing the sign (0o/180o phase shift) In QPSK (Quadrature PSK), the modulation symbol is a group of two bits

Comments on the previous scheme Every symbol is a group of two bits (AB) On the upper branch, the signal corresponding to bit A (e.g.:+V voltage for 1, -V voltage for 0) modulates a cosine On the lower branch, the signal representing the bit B, modulates a cosine of the same frequency, but shifted by 90o Thats the reason behind the Q in QPSK The signal from the two branches is summed an transmitted in the channel 23. 8-PSK modulator ((QPSK principles, modulators scheme, constellation, phase table, mathematical expression for the generated signals) 8-PSK modulator

A group of three bits composes a symbol a and b will be 4-levels AM signals The second modulator is a phase modulator

A B C a (sign) (sign) (level) 0 0 0 Negative, small amplitude 0 0 1 Negative, High amplitude 0 1 0 Negative, Small amplitude 0 1 1 Negative, High amplitude 1 0 0 Positive, Small amplitude 1 0 1 Positive, High amplitude 1 1 0 Positive, Small amplitude 1 1 1 Positive, High amplitude

b Negative, high amplitude Negative, Small amplitude Positive, high amplitude Positive, Small amplitude Negative, High amplitude Negative, small amplitude Positive, high amplitude Positive, Small amplitude

Phase -112.5o

-157.5o

+112.5o

+157.5o

-67.5o

-22.5o

+67.5o

+22.5o

24. Coherent demodulation of QPSK signals. QPSK demodulation

Phase +45o -45o +135o -135o

SA + + -

SB + + -

sd 11 10 01 00

Local reference oscillator After the LPF, the 20 component is removed After sampling, a positive value indicates a 1, a negative sample indicates a 0 (on both branches) Equivalently, a single phase detector and a multi-level comparator can be used 25. DBPSK: principles, demodulation. DBPSK Principles DBPSK=Differential Binary Phase Shift Keying At modulator, a pre-coder provides a differential phase shift Instead of transmitting an (the initial bit sequence), the data sequence transmitted in the channel is precoded:

Example:

DBPSK Demodulator

For the n-th bit, the received signal (after) BPF is:

After the LPF (which eliminates the second harmonic), we get: Remarks on the DBPSK demodulator The signal delay, T, introduced in the demodulator, must equal the symbol period The delayed signal will then correspond to the previously transmitted bit The signal entering the LPF will be:

By transforming this product into a sum we will obtain an oscillating component (20 and a DC component, indicated by eq. 8)

26. Classification of the data transmission systems for the synchronicity point of view. Synchronous systems: every bit has the same duration, T The same time interval is preserved between any two consecutive bits (multiple of T) Bit-level synchronized systems Referred to as start/stop asynchronous systems Synchronization realized at bit level No synchronization between two successive characters (words) Step-by-step synchronized systems Fully asynchronous systems No synchronization neither at bit, nor at character level

27. One-step digital synchronization Digital approach: one-step Synchronizer

Input: data signal Output: clock signal synchronized with data One-step synchronizer: comments on the scheme Transition time instant compared to the clock transition If there is a time shift between them, the receivers time basis is tuned The scheme uses a higher resolution clock (clock frequency: 64/T) Frequency divisors are used to tune the clock frequency 28. Two steps digital synchronization (fine and coarse tuning) Two-step synchronization

Input: data signal Output: clock signal synchronized with data Two-step synchronization Principle: the high and the low synchronization errors in the receivers clock treated differently If the synchronization error is high, a coarse tuning takes place An impulse is added/cleared, having a weight of 1/24 (4%) The weight is given by the multipliers (4x6=24) If the synchronization error is small, a fine tuning takes place Impulse added/cleared, having the weight 1/192 The weight results from 4x2x6x4=192

29. USB (definition, channels, transmission speed, signal encoding at PHY layer, power supply). USB Interface USB = Universal Serial Bus Definition: serial bus standard to interface devices to a host computer Used for plug-and-play peripherals connection: mouse, keyboard, PDA, printer, scanner, digital camera Provides power for low-consumption devices (no need for external power supply) USB Channels 4 data channels for the communication between the device and the USB host controller Control data transfer- uses pipe 0: commands towards the device or status reporting Isochronous transfer- for the transmission at a guaranteed speed of real time data (e.g. voice, video) Interruption data transfer- from some peripherals to the host (e.g. keyboard, mouse, joystick), requires small delay guarantees Bulk transfers: they may use the whole bandwidth, but they are not time-critical (e.g transfers to/from a storage device) Transmission speeds Low speed (<1.5 Mbps) for human interface devices (mouse, keyboard, etc) Full speed (<12 Mbps): legacy from USB 1.0, bandwidth shared using the principle first-came-first-served High speed (<480 Mbps), characteristic to USB 2.0 Not all USB 2.0 devices support high speed Theoretical maximum speed of a USB 2.0 device: 60 Mbps Most of the devices barely support one half of this speed USB-IF gives certificates of compliancy PHY layer details

PHY layer details [2] Signal encoding: NRZI with bit-stuffing Only 0 introduces a change in the transmitted signal After six bits of 1, a 0 is introduces and it is ignored by the receiver Synchronization byte at the beginning of the frame (01111110) Power supply of 5V (between 5.25 V and 4.75 V allowed between the line +D and -D)

30. V.24 (RS232) serial interface: number of pins, data pins, electrical and functional specification (circuit types) V.24 interface ITU-T v.24 Only specifies functional and procedural References other standards for electrical and mechanical EIA-232-F (USA) RS-232 (more popular name) Mechanical ISO 2110 Electrical V.28 Functional V.24 Procedural V.24

Mechanical specification

25 wires for serial communication Far fewer ports used in practice (oftentimes) Electrical specification Digital signals transmitted Values interpreted as data or control, depending on circuit less than -3V is binary 1, more than +3V is binary 0 (NRZ-L) Signal rate < 20kbps Distance <15m For control, more than-3V is off, +3V is on Functional specification V.24 interface Data, control, timing and ground circuits Full-duplex and half-duplex operation possible Full-duplex using 103, 104 Half-duplex using 118, 119 Control circuits For asynchronous functioning (105, 106, 107, 108.2, 125, 109) Signal quality detector (110) control of the secondary channel (120, 121, 122) Loopback testing (140, 141, 142) Timing circuits Provide timing for synchronous transmission When DCE is sending synchronous data over 104, it also transmits transitions on the circuit 114 (for bit timing) Ground /Common return: return circuit for all data leads

31. Session establishment and communication DTE-DCE using the RS232 interface Example 1: Asynchronous private line modem When turned on and ready, modem (DCE) applies negative voltage to DCE ready (pin 6) When DTE ready to send data, it asserts Request to Send (pin 4) Also inhibits receive mode in half duplex Modem responds when ready by activating Clear to send (pin 5) DTE sends data When data arrives, local modem asserts Receive Line Signal Detector and delivers data

Example 2: Telephone line-modem Supplementary circuits required for the operation over telephone line (DTE ready, Ring indicator) 32. Present briefly, the OFDM principles. The need for OFDM Most of the transmission channels are frequencyselective frequency components from the input signal affected differently by the channel meaning that the channels transfer function H(f) is not flat over the bandwidth this introduces Inter-Symbol interference (ISI) ISI can be seen as a time-domain manifestation of the frequency selectivity OFDM is exceptionally robust to IIS Well suited to all the dispersive channels, and especially to the wireless channel 33. The cyclic prefix in OFDM. Cyclic prefix: why? If the channel is not ideal (is time-dispersive), the successive OFDM blocks will interfere This happens because each symbol is dispersed in time by the channel (see silde 6)

Inter-Block Interference (IBI) arises Sometimes referred to as Inter-(OFDM)Symbol Interference OFDMs cyclic prefix Sometimes referred to as guard time Cancels IBI Simplifies equalization Eases synchronization (a signal is always transmitted) Cyclic Prefix is removed by the demodulator

Comments on the cyclic prefix IBI cancellation Occurs only if the CP is longer than the multipath delay Longer CP duration means higher protection against ISI, but smaller efficiency Synchronization is simpler if a signal is always transmitted E.g. a peak in the autocorrelation of the OFDM symbol could indicate the OFDM symbol start Equalization is simpler, because the linear convolution (x[n]*h[n]) is transformed into a periodic one This allows to use a very simple, one-tap, frequency domain equalizer Every received sample R[k], must be divided by H[k], the channels response at the k-th frequency line

34. Explain the OFDM's carriers orthogonality in time and frequency domain (graphical illustration too). Orthogonal carriers The OFDM carriers are orthogonal, their frequencies being f0, 2f0, 3f0 etc.

Complex exponentials of limited duration used in practice Their duration equals OFDMs symbol time (T) The orthogonality is met if: f0=1/T OFDM: the carriers and their spectra

Comments on the previous slide In the time domain, every carrier covers an integer number of cycles (periods) during the symbol time (T) This is a condition for orthogonality The carriers are time-bounded by a rectangular window, giving the symbol duration The sinc shape of the spectrum corresponds to a sine carrier multiplied by a rectangular time window At the central frequency of each carrier, all the other carriers cross zero This is the orthogonality frequency view Carriers must be separated by 1/T on the frequency axis

35. OFDM block diagram (transmitter, receiver).