Physical Layer

Transmission

• Transmission a sequence of bits over an analogue channel
• addresses issues like noise, clock skew, hardware limitations.

Interface to Layer 2(Physical Layer)

• provides a sequence of bits
  • Hides the complexity of encoding/decoding from higher layers

Wired Media

Unshielded Twisted Pair(Cable)	optical fibre(Fibre)
electrical cable using two wires twisted together	glass core and cladding, contained in plastic jacket for protection
-each pair is unidirectional: signal & ground	-fragile
-twist reduces interference and noise	-unidirectional: transmission laser at one end; photodetector at the other
-susceptible to noise increased with length	-very low noise,electromagnetic interference does not affect light
	-Very high capacity: 10s of Gbps over 100s of miles
	-cheap to manufacture

Wired Data Transmission

• signal usually directly encoded onto the channel
• Encoding preformed by varying the voltage in an electrical cable, or intensifying light
• Many encoding schemes exist:
  • NRZ (Non-Return to Zero)
  • NRZI (Non-Return to Zero Inverted)
  • Manchester
  • 4B/5B
  • etc…

Baseband Data encoding

encoding the signal, 1 if high 0 if low

Carrier Modulation

The process of applying a carrier wave to a channel at a specific frequency(C) Modulation: the act of varying one or more properties (amplitude, frequency or phase) of the carrier wave in relation to the information signal Purpose:

• Shift baseband signal: modulates the signal onto the carrier to shift it from the baseband to a higher carrier frequency
• Multiple signals on a single channel: allows multiple signals to share a single channel by using different carrier frequencies Applications: used in wireless links can also be used in wired links(how ADS^[Asymmetric Digital Subscriber Line, A type of DSL technology used for transmitting digital data over traditional copper telephone lines, offering higher download speeds than upload speeds] and voice telephone share the same line)

Bandwidth & Signal Capacity

Determines the frequency range it can transport Based on physical properties of the channel, design of the end points

Digital Sampling

the (Nyquist’s) sampling theorem : needs 2H samples per second to accurately digitise an analogue signal

• $R_{max}=2H{\log }_{2}V$
  • $R_{max}$: Maximum transmission rate of channel (bits per second)
  • $H$: Bandwidth (Hz)
  • $V$: number of discrete values per symbol
  • Assumption: perfect, noise-free channel

Noise & Signal-to-Noise Ratio

Real world noise includes:

• Electrical interference, cosmic radiation, thermal noise, etc.

Computing the Signal-to-noise Ratio($\frac{s}{n}$ or SNR^[SNR (Signal-to-Noise Ratio) ▪ The ratio of the Signal Power (S) to the Noise Floor (N)])

• Typically quoted in decibels(dB): $10{\log }_{10}\frac{s}{n}$

Shannon’s Theorem

The maximum transmission rate of a channel grows logarithmically to the SNR: $R_{max}=H{\log }_2\left(1+\frac{s}{n}\right)$