EITF25 Internet Techniques and Applications L2: Physical layer Stefan Höst
Data vs signal Data: Static representation of information For storage Signal: Dynamic representation of information For transmission 2
Analog vs Digital data Analog data: Continuous representation Examples: Audio, Images (on paper) Digital data: Discrete representation (states). Examples: Text, Images (in computer) 3
Analog vs Digital signal Analog signal: Continuous amplitude Eg summation of sinusoids Digital signal: Discrete amplitude (finite levels). Eg number of customers in a queue Conversions: Digital Analog!! Analog: DAC, Reconstruction, Digital: ADC, Digitalization 4
Analog v. Digital signals 5
Digital vs analog Digital data digital signal Ethernet, PON Digital data analog signal ADSL, 3G, LTE, WLAN Analog data digital signal Telephone transport network, IPTV Analog data analog signal FM radio 6
Digitalization The process is performed in three steps: 1. Sampling Discretization in time 2. Quantization Discretization in amplitude 3. Encoding Binary representation of amplitude levels 7
Bandwidth F! The bandwidth of a signal is the frequency band where its main power lies. 8
Sampling Sampling: The process of discretizing the time of a continuous function. s(t) s(t) 4 2 0 2 4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 4 t 2 0 2 4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 4 t 2 s[n] 0 2 4 0 5 10 15 20 25 30 35 40 45 n 9
Aliasing y(t)=cos(14πt) 1 0.5 0 π Ω T π 0.5 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Reconstruction to lowest possible frequency y(t)=cos(6πt) 1 0.5 0 0.5 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 10
Nyquist Sampling Theorem If s(t) is a band limited signal with highest frequency component F max, then s(t) is uniquely determined by the samples s n = s( nt ) if and only if F s = 1 T 2F max The signal can be reconstructed with F max s(t) = n s n sinc t nt T is the Nyquist frequency and F s the Nyquist rate 11
Reconstruction Example y(t)=cos(14πt), F s =1Hz 1 0.5 0 0.5 1 2 0 2 4 6 8 10 12 1 0.5 0 0.5 1 2 0 2 4 6 8 10 12 12
Quantization Linear Quantization 2 N equidistant levels Represent sample with N bits Telephony: 2 1.5 1 0.5 0 0.5 1 x(t) x[n] Quant level N=8: 256 levels CD: 1.5 2 0 1 2 N=16: 65 536 levels 13
Quantization Delta modulation Represent change in amplitude with 1 bit s(n) 1: +1 0: 1 5 10 15 n 1 0 1 0 0 1 0 1 1 1 1 1 1 1 1 1 0 1 14
Digital communictions 15
Data rate vs signal rate The data rate is the number of bits transmitted per second. It has the unit b/s (or bps) and is often denoted R b. The signal rate is the number of signal alternatives transmitted per second. It has the unit Hz (sometimes baud) and is often denoted T s. If there are k bits/signal, R b =kt s. 16
Example: Bit rate for telephony links Analog signal in frequencies 0-4kHz. Nyquist theorem gives that the sampling frequency is 2x4 khz = 8000 samples per second. 8-bit encoding. The bit rate is 64 kbps. 17
Performance metrics Throughput: The real transmission capacity between a source and a destination (bits/sec). Latency: The time required to send a message from a source to a destination. Is a sum of propagation time, transmission time, queuing time and processing delay. Bandwidth-delay product: Bandwidth (Throughput) x Latency 18
Transmission Medium Anything that can carry information from a source to a destination. Guided media (wired communication): Twisted-pair cable, Coaxial cable, Fiber-optic cable Unguided media (wireless communication): Radio, Microwaves, Infrared 19
Digital communictions 110101 Network adapter link Network adapter Digital transmission: The bits are represented by digital signals. Analog transmission: The bits are represented by analog signals. 20
Digital signals Digital signals are represented by changes in the voltage amplitude 21
Non-return to zero (NRZ-L) 0 = high voltage amplitude 1 = low voltage amplitude 0 0 1 0 1 1 0 1 0 0 22
Base-band transmission All data transfer occur on a channel with finite bandwidth. In base-band transmission the used frequency band is located close to zero s(t) Rs(f) t f 23
Pass-band transmission All data transfer occur on a channel with finite bandwidth. In pass-band transmission the used frequency band is located far from zero s(t)*cos(wt) R(f) t f 24
Manchester Combines NRZ with a clock pulse. 0: 1: t t 0 0 1 0 1 1 1 1 0 1 0 Doubled bandwidth compared to NRZ Built in synchronization. 25
Differential Manchester 0 = Inversion in the beginning of the bit 1 = No inversion in the beginning of the bit 0 0 1 0 1 1 1 1 0 1 0 26
Analog transmission Analog transmission uses modulation. The digital data is represented by sinus waves. 27
Types of modulation schemes 28
Amplitude Shift Keying (ASK) In Amplitude Shift Keying (ASK), the amplitude of the carrier signal is varied. Often the term Pulse Amplitude Modulation (PAM) is also used. 29
Frequency Shift Keying (FSK) In Frequency Shift Keying (FSK), the frequency of the carrier signal is varied. 30
Phase Shift Keying (PSK) In Phase Shift Keying (PSK), the phase of the carrier signal is varied. 31
Transmission impairment When a signal travels on a link, it will deteriorate due to transmission impairment. Attenuation (swe: dämpning): Loss of energy Distortion: Change of signal shape Noise: The signal is corrupted due to, e.g., thermal noise or crosstalk (swe: överhörning). Signal-to-noise ratio (SNR) = Average signal power Average noise power 32
Attenuation 33
Distortion 34
Noise 35
Additive White Gaussian Noise (AWGN) n(t) s(t) y(t) If the noise is white (i.e. R(f)=N 0 /2) the corresponding discretized noise samples are i.i.d. Gaussian η n N(0, N 0 / 2) 36
Shannon capacity If s(t) is a band limited signal with band width W transmitted over an AWGN channel, the maximum achievable data rate [b/s] is given by C = W log 2 1+ P N 0 W = W log 2 ( 1+ SNR) 37
Data communication Two stations that transmit data on a physical link cannot do this simultaneously on the same frequencies with the same coding scheme. 38
Data flow concepts 39
Multiplexing of links Also, physical links need to be shared. This is called multiplexing, where one physical link is divided into several channels. 40
Multiplexing techniques 5 basic types of multiplexing techniques: Space-Division multiplexing (SDM) Frequency-Division Multiplexing (FDM) Wavelength-Division Multiplexing (WDM) Time-Division Multiplexing (TDM) Code-Division Multiple Access (CDMA) 41
Space-Division Multiplexing (SDM) SDM is used in fibre-optic cables. Each channel uses one optical fibre. 42
Frequency-Division Multiplexing (FDM) FDM is an analog multiplexing technique where each link has its own frequency band Ch 1 Ch 2... Ch N f Each channel uses a unique carrier frequency. 43
FDM process 44
Wavelength-Division Multiplexing (WDM) WDM is an analog multiplexing technique to combine optical signals. 45
Time-Division Multiplexing (TDM) In TDM, each channel occupies a portion of the time in the link. 46
Synchronous TDM If a channel has nothing to send, its time slots will be empty! 47
Example: Empty slots 48
Frame synchronizing If the multiplexor and the demultiplexor are not synchronized, bits may be received by the wrong channel. Therefore, synchronization bits (framing bits) are added in the beginning of each frame. 49
Statistical Time-Division Multiplexing In Statistical TDM, the channels have no reserved time slots. Instead, slots are dynamically allocated. Data about the destination is added to each slot. Statistical TDM usually has better performance than Synchronous TDM when not all channels transmit data all the time. 50
TDM comparison 51
Code-Division Multiple Access (Spread Spectrum) CDMA, or Spread Spectrum (SS), is a multipple access technique for wireless links. The original signal is changed in a spreading process. 52
Spread Spectrum techniques Frequency Hopping Spread Spectrum (FHSS) A source uses many carrier frequencies. One frequency is used at a time, but the frequencies are changed very often (e.g. 1000 times a second). Direct Sequence Spread Spectrum (DSSS) Each data bit is replaced with n bits (called chips) using a unique spreading code. The code is chosen so that the combination of all other sources can be treated as white noise. 53
FHSS A source use a pseudorandom code generator to find the next carrier frequency. 54
FHSS cycles 55
Comparison with FDM 56
DSSS The original signal is combined with the spreading code. 57
DSSS example 58