Week 4 - Lecture 8 - Segmentation and Clustering Flashcards
Name the three modulation techniques
Amplitude modulation (AM)
Frequency modulation (FM)
Phase modulation
Signals are modulated to allow higher bandwidth, longer transmission distance and managing noise, interference and losses.
What is phase shift keying (PSK)?
The phase of a constant amplitude carrier is modified to represent different binary symbols
For binary PSK two phase values encode the 1 and 0 logic values, typically the phase difference is 180 degrees
Other types of PSK
Differential (DPSK)
Quadrature (QPSK)
pi/4 QPSK
What is frequency shift keying (FSK)?
In FSK the frequency of a constant amplitude carrier signal is varied to represent different binary symbols
For binary FSK, the frequency changes between two values (high (FC1) and low (FC2) tones) corresponding to binary 1 and 0
With 8 frequencies you can represent 3 bits in one change
What is spread spectrum modulation (SSM)?
SS methods require bandwidth which is orders of magnitude greater than other modulation techniques
They spread the signal onto many frequencies
Ideal for multiple users
Spread spectrum signals are pseudorandom and exhibit noise like properties
SS techniques utilise pseudo-noise sequences produced by appropriate circuits
- The pseudo-noise codes are also available to the receiver to demodulate the spread spectrum signal
Signal recovery is performed by cross-correlation with a locally generated version of the pseudorandom carrier with which the signal was modulated
Cross-correlation with the local pseudorandom sequence of an unintended user yields a very weak wide-band noise at the receiver
They have uniform energy over a very large bandwidths at any given time only a small portion of the spectrum will suffer from fading.
How do pseudo-noise (PN) sequences work?
Binary sequences that have an autocorrelation that over a period of time resembles the autocorrelation of random binary sequence
Generated by Feedback shift registers, these consist of a series of flip flops equal to number of bits of the sequence, say m-bits, usually implementing a higher order polynomial using XOR-gates (linear feedback shift register (LFSR))
The period of an m-bit shift register is 2^m - 1 symbols. This means a sequence may be repeated after 2^m-1 symbols.
Useful in military, only talk to one frequency then it can be intercepted, spreading over more frequencies mean it’s harder to read the frequency
How does Direct Sequence Spread Spectrum (DS-SS) work?
Directly multiplies the baseband data pulses with a PN sequence generated by a PN sequence generator
At the receiver the narrowband signal is recovered by correlating the wide-band signal with the appropriate PN
How does frequency hopped spread spectrum (FHSS) work?
A periodic change of transmission frequency
May be seen as a sequence of modulated bursts of data with time-varying, pseudorandom carrier frequencies.
The hopping sequence of frequencies is only known to the desired receiver
- Time between hops is called the hop duration or hopping period
- Total bandwidth of the hopset and the instantaneous BW of each frequency are notated as Bss and B
- The processing gain for FH systems is Bss/B
What is channel coding?
Introducing redundant bits in the transmitted data to reduce errors
Can either detect or correct errors
Higher tolerance to noise comes at a cost to bandwidth efficiency (yet is necessary!)
Block codes
forward error codes that can detect a specific number of errors, depending on complexity, and correct those errors.
Parity bits are added to the bits of the message forming codewords or codeblocks. In general, a block coder, k data bits are encoded into n code bits. Thus, the redundant bits are n - k, code is usually referred to as an (n,k) code. Examples include Hamming, Hadamard and cyclic codes
Convolutional codes map a continuous sequence of information bits into a continuous sequence of encoder output bits. A convolutional code can demonstrate a higher coding gain as compared to block codes with the same level of complexity. To produce a convolutional code, the information sequence is shifted in through a shift register. Typically the shift register contains N k-bit stages and m linear algebraic function generators.
Turbo codes have been invented more recently and are integrated with 3G wireless standards. Combines convolution with channel estimation theory and can be described as nested or parallel convolutional codes. Their coding gains are much higher than the other types of codes allowing the capacity of a wireless channel to reach the Shannon capacity bound.
What are multiple access schemes?
Multiple access methods allow many (mobile) users to share simultaneously a finite amount of radio spectrum.
What is Frequency Division Duplexing (FDD)
Separate frequency channels for transmitting and receiving
What is time division duplexing (TDD)
Assign separate time slots for transmitting and receiving
What are the four primary ways to partition the common medium (“air”) based on
Frequency
Time
Space
Coding
What are the multiple Access Methods?
Frequency Division Multiple Access (FDMA)
Time Division Multiple Access (TDMA)
Code Division Multiple Access (CDMA)
How does FDMA work?
Assigns multiple individual channels (frequency bands) to individual users
- Only the assigned user can use the channel even if it’s not used for some time (but the call is still active)
The number of channels supported by FDMA is given by an equation of the total allocated spectrum, the allocated guard band and the channel bandwidth
What is TDMA?
Splits the spectrum into time slots and each user uses one time slot to either transmit or receive.
The specific time slots are periodically repeated.
The time slots form a repeated TDMA frame
