Title: Telecommunications
1Telecommunications
2Communication System Designers Goal
- Maximize information transfer
- Minimize errors/interference
- Minimize required power
- Minimize required system bandwidth
- Maximize system utilization
- Minimize cost
3Useful Relationships
- Decibels
- A logarithmic unit originally devised to express
power ratios but used today to express a variety
of other ratios as well - where P1 and P2 are the two power levels being
compared
4Examples
- Loss
- 1,000 watts (P1) 10 watts (P2)
- Gain
- 10 watts (P1) 1,000 watts (P2)
-
P1
P2
Power in
Power out
The unit decibel was named after Alexander Graham
Bell. The unit originated as a measure of power
loss in one mile of telephone cable. Also,
hearing is based on decibel levels.
20 dB means 100 times more
5Derived Decibel Units
- The dBm
- Example 20W is what in dBm?
- The dBW
- Example conversions
6Voltages (examples)
7Gains and Losses
- Power is gained via amplification and lost via
absorption or resistance - Gains and losses are expressed in dB (usually the
W or m are dropped)
8Communications Example
Attenuation x dB
Gain b dB
Attenuation y dB
Pin
Gain c dB
Gain a dB
9Special Values
10Other Examples
- Sound levels
- If Pref is the sound power resulting in a barely
audible sound,
11Radio Frequency Radiation
- RF signals travel at the speed of light in air
(atmosphere) and space (vacuum) - c speed of light in vacuum
- 2.998x108 m/sec (186,200 miles/sec)
- Wavelength, ?c/f
- f frequency
- Beam width ? (rad) ? ?/D
- D aperture width or diameter
- Defines how spread out the beam is
12Half Power
- A 3 dB drop in power represents the half-power
point
13Isotropic Radiation
- Aperture area of a receiving or transmitting
antenna through which all signal is assumed to
pass. - If transmitting antenna radiates equally in all
directions, it is called isotropic - The fraction of power received from an isotropic
radiator at a distance, d, is - where Ar is the aperture area of the receiving
antenna
14Isotropic Radiation (contd)
- Receiver is not 100 efficient, so including
efficiency factor, z, - Z ? 0.55
- Transmitting antenna designed to focus radiation
(i.e. not isotropic) - Can also be expressed in dB
15Typical Antenna Patterns
16Parabolic Reflector Antenna
parallel beams
focal point
- D diameter
- - wavelength
- z - efficiency
17Lobes
18Cassegrain Reflector Antenna
19Modulation
- Definition
- Altering a signal to make it convey information
(either analog or digital) - AM (Amplitude Modulation)
- Changes amplitude (frequency constant)
- FM (Frequency Modulation)
- Changes frequency (amplitude constant)
Frequency modulation
20Modulation (contd)
- Changing the phase of the signal
- For digital data, these methods are also called
- ASK amplitude shift keying
- FSK frequency shift keying
- PSK phase shift keying
21Link Budget
- Allocation of various losses and gains in the
communication link between Earth and the
spacecraft - Similar to signal-to-noise ratio, but Eb/No
pertains to digital data
22Link Requirements
- For data
- (Eb/No)estimated (Eb/No)required ? 3 dB
- For commands
- (Eb/No)estimated (Eb/No)required ? 20 dB
- This difference is known as the link margin
23Terms
- P transmitter power
- Ll line loss (between transmitter and antenna)
- Gt transmitter antenna gain
- Ls space loss (inverse square in distance)
- k Boltzmanns constant
- La transmission path loss (atmosphere and rain
absorption) - Gr receive antenna gain
- Ts system noise temperature
- R data rate
- Li implementation loss (?-2 dB)
24More Details
25More Details
26More Details
27More Details
- Calculate Link Margin (Eb/No)est (Eb/No)req
Fig. 13-9, SMAD
Acceptable BER
28Example
- If acceptable BER (bit error rate) is one bit
error in every 100,000 bits, then BER10-5 - Using BPSK modulation with Reed/Solomon coding,
this requires an Eb/No2.5dB - If BPSK is used without coding, Eb/No9.5dB
- Increase transmitter power by 7 dB
- Multiplicative factor of 100.75
- Increasing the transmitter and receiver antenna
gains by 7dB (combined) - Antennas then more sensitive to pointing errors
29Data Rates
- For each sensor, data rate
- Sample size is determined based upon required
level of accuracy - Example temperature sensor needed to monitor
propellant tank temperature in range -10?C to
80?C - Amplitude range80?C-(-10?C) 90?C
30Data Rates (contd)
- Sensor generates voltage proportional to
temperature - A/D converter generates a digitized
representation of this temperature an n-bit
word - Number of quantized levels that are represented
2n - Quantization step here
Quantized steps
31Data Rates (contd)
So, if n8, then quantized step 0.35oC and Eq
0.175oC Typically, one needs to find the required
value of n. Using same example, if required Eq ?
0.05oC, then quantized step 0.1oC and
32Sample Rate
- Determined based upon estimated rate of change of
quantity being measured - Examples
- Thermal sensors typically sample at low rates
(once per minute) - Attitude sensors sample at high rates, especially
during attitude maneuvers (1-5 samples/sec)
33Sampling Oscillatory Phenomena
- Must sample at 2.2 times the highest frequency
present - Human voice has frequency range of 3.5 KHz
- Sample at 7.7 KHz (7,000 samples/second)
- Commercial audio (telephony) requires 8
bits/sample - Data rate 7,700 samples/sec x 8 bits/sample
- 62,000 bits/sec (bps)
34Data Compression
- Compression/encoding allow lower data rates
- Make use of repeated patterns in the data and/or
transmit only parts of data that changes since
previous sample - Voice data can be reduced to 9.6 Kbps
- Compressed video (videophone) 28 Kbps
- Full video with color 256 Mbps (40 Mbps with
coding)
35Telemetry
- Packet telemetry format
- Each sensor forms packet of data
- When packet complete, microcomputer interrupts
main computer - Main computer formats main block
- Main block transmitted
- Advantages
- Flexible data rates for sensors
- Disadvantages
- Spacecraft processing more complex
- Ground station equipment more complex
36Error Detection and Correction
- Once our telemetry data is set to transmit, we
must concern ourselves with possible induced
errors in the transmission - With digital data, there are several ways to
check for errors - Parity check (with retransmission)
- Error correction (without retransmission)
- Ref Spacecraft Attitude Determination and
Control, J.R. Wertz (ed), Reidel Publishing Co.,
1978
37Parity Check
- Simplest method of detection
- Example
- M 1,1,0,0 original message
- Add another parity bit to M
- M now becomes 1,1,0,0,p
- Even parity scheme
- m1 m2 m3 m4 p even number ?p0
- Odd parity scheme
- m1 m2 m3 m4 p odd number ?p1
- Receiving equipment then checks each message
vector
38Parity Check
- Suppose receiving equipment receives
- M 1,1,0,0,1
- If both transmitter and receiver are employing
even parity scheme, then an error has occurred - m1 m2 m3 m4 p 3 not an even number
- Receiver requests retransmission
- What if two bits are flipped?
- Parity scheme fails (much lower probability of
two bit flips than one bit flip)
39Error Correction without Retransmission
- Example self-correcting developed by Hamming
- Extra set of bits equal in number to the original
message bits added to message vector - Before M a,b,c,d
- After M p1,p2,p3,a,p4,b,c,d
40Hamming (contd)
- Multiply MT by the Hamming matrix,
- to get SHMT (syndrome vector)
41Hamming (contd)
- Need to determine the values of p1,p2,p3, and p4
- Set these such that S 0,0,0,0T (mod 2)
- (Any even number 0 mod 2)
- Arrangement of parity bits in M so that only one
new parity bit is involved in each successive
calculation of p1,p2,p3,p4
42Hamming Example
- Intended message vector Mo0,0,1,1,1,1,0,0
- Received message vector M10,0,1,1,1,0,0,0
- Correction scheme
- If s4 0, then a, b, c, and d are correct
- If s4 1, then error occurred in message bit
s1s2s3 (101)25
43Hamming Example (contd)
- M b0 b1,b2 b3,b4 b5,b6 ,b7
- M10, 0, 1, 1, 1, 0, 0, 0
- Correct M1 is M10, 0, 1, 1, 1, 1, 0, 0
- So the original message data is 1 1 0 0
-
a b c d
Error
44Probability of Errors Simple Parity
- If probability of error in 1 bit is 1,
probability of at least one error in a 4-bit
message is 4 - Adding one parity bit increases error rate to 5
- Can detect, but not correct this error
- Need to retransmit 5 of the data
- Probability of 0.25 that error occurs in 2 or
more of the original 4 bits
45Probability of Errors Hamming Code
- Using the 8-bit Hamming code will increase
probability of error to 8 - One bit error can be corrected
- Errors in 2 bits of M will occur in 0.64 of
messages received - Two bit errors cannot be corrected
- Hamming will detect two errors, so retransmission
can be requested - Undetected errors in 3 or more bits will occur in
0.051 of the messages received.