•Lecture 3 •Paul Flynn Modulation

Frequency Spectrum

Long Waves, Medium Waves and Short Waves

Modulation of Analog Signals Amplitude Modulation (AM) Frequency Modulation (FM) Phase Modulation (PM)

Figure 5.24 Analog-to-analog modulation

Types of analog-to-analog modulation

Amplitude modulation (AM)

Frequency modulation (FM)

Phase Modulation (PM)

Digital modulation •Three basic methods •– Amplitude shift keying (ASK) •– Frequency shift keying (FSK) •– Phase shift keying (PSK

Amplitude shift keying (ASK) •Use different amplitude to represent 0 and 1. •– Simple, low bandwidth – Sensitive to interference. • Multi-path propagation, noise or path loss heavily influence the amplitude. •– A constant amplitude in wireless environment can not be guaranteed. • Used in wired optical communication. •– A light pulse =1, no light =0.

Frequency shift keying (FSK) •Binary FSK (BFSK) – One frequency for 0 and one •frequency for 1. – needs larger bandwidth • Avoid discontinuity – Discontinuity creates high •frequencies as side effects. – Continuous phase modulation (CPM) can be used. • Demodulation: – Use two bandpass filters for 2 frequencies.

Phase shift keying (PSK) •Use shift in phase to represent data. • Binary PSK (BPSK) – Shift the phase by 180. • Synchronization is important • More resistant to interference • More complex transmitters and receivers.

Digitizing Analog Signals 1.Sample 2.Quantize 3.Encode

Sampling Analog Signals Original Signal Sample Times Sample Values Reformed Signal

Digitizing Analog Signals Original Signal and Samples Actual Sample Values Quantized Sample Values Reconstructed Signal Original Signal

Linear Quantizing Actual Sample Values Quantized Sample Values Reconstructed Signal Original Signal

Linear Quantizing Linear: quantizing steps for all segments are the same height. SQRSQR 2-42

Quantizing Noise TransmitReceive

Voice Digitization Process

Pulse Code Modulation Waveform-Time Domain-Nondifferential Sample Values Quantized value of each sample is coded

Coding How many bits are needed to code this many levels? Answer: 3 bits for 8 levels L=2 n

Companding or Non-linear Encoding zCompanding = compressing + expanding zWhy companding? zQuantization levels not evenly spaced zReduces overall signal distortion zCan also be done by companding

Binary Signaling Encoding Schemes 0 = 1 0 to 1 1 to 0 1=sq sq 1 & 0 c x on 1 chg on 0 alt mark inversion (B8ZS) sq wv 1 inv 2 nd 1 nnnnssssnnnnnnssssnn EIA 232 B8ZsB8Zs

2B1Q (ISDN) EACH LEVEL REPRESENTS TWO BITS 00, 01, 10, AND 11

The digital Loop Signal

AMI (Alternate Mark Conversion) Violation START OF CONTROL INFO NO MORE THAN 15 0s IN A ROW