Multirate Signal Processing

1
Multirate Signal Processing

• Multirate Signal Processing The implementation
  of a digital signal processing application using
  variable sampling rates
• Can Improve the flexibility of a software radio
• Reduces the need for expensive anti-aliasing
  analog filters
• Enables processing of different types of signals
  with different sampling rates
• Allows partitioning of the high-speed processing
  into parallel multiple lower speed processing
  tasks ? reduced costs
• Can lead to a significant saving in computational
  power
• Wideband receivers take advantage of multirate
  signal processing for efficient channelization
• Offers flexibility for symbol synchronization and
  downconversion of software radios

2
Multirate Signal Processing

• Sample Rate Conversion Principles
• (Integer Rate Conversion)
• - Decimation
• - Interpolation zero-insertion, zero-order-hold
  (ZOH), zero-insertion, raised-cosine filtering,
  fast Fourier transform (FFT) expansion
• Multi-Stage Sampling rate conversion
• - offer less computation and more flexibility in
  filter design

3
Cascaded Integrator Comb Filter (CIC)

• Implement an interpolation or decimation filter
• CIC Decimation Filter
• CIC Interpolation Filter

4
Polyphase Filter

• A way to reduce computational cost for a
  multirate system, by replacing high-speed
  elements with low-speed processing elements
• Polyphase Decimation
• Polyphase Interpolation

5
Digital Filter Banks

• Applications -To break a wideband signal into
  multiple bands
• -Allowing lower-speed computational techniques
• -Converting a signal from frequency division
  multiplexing (FDM) to time division
  multiplexing (TDM) or vice versa.
• Implementation
• Two Basic Architectures
• 1. Filter Bank Analyzer decomposes the input
  signal into several channels
• 2. Filter Bank Synthesizer reconstructs the
  input by combining the channel
  signals

6
Time Recovery in Digital Receivers Using
Multirate Digital Filters

• Reduces complexity compared to conventional
  timing recovery techniques

Timing recovery in an Analog Receiver
Timing recovery in a First Generation Digital
Receiver
A Multirate Filtering Approach to Synchronization
7
Direct Digital Synthesis

• Direct Digital Synthesis (DDS) offers a number of
  advantages to signal generation in software
  Radio. Its digital nature in particular makes it
  very attractive.
• Precision - Possible to set frequency accurately,
  with high resolution.
• Flexibility - Easy to change output parameters.
• Switching Frequency - High switching speeds
  possible, Output is smooth and transient free
  during frequency change. Possible to have
  continuous phase during frequency switching.
• Equipment size - DDS can be implemented at a
  fraction of the size of analogue synthesizers.
• Spectral purity If accumulator size is an
  integral multiple of the step size, the will be
  no phase truncation, and signal quality will be
  very high.

8
Approaches to DDS

• Pulse Output DDS
• Simplest of all DDS models
• Consists of N-Bit Adder and register to produce
  saw tooth waveform.
• Fout ?rFclk/2N
• Limited use due to high spur and jitter levels.

• ROM Look Up Table
• Sampled Values of periodic wave stored in ROM
• Wave period defined by ?r.
• Subject to phase truncation

9
Performance Of DDS

• Spectral Purity and Sideband noise are major
  drawbacks of DDS.
• DDS performance closely related to DAC
  performance
• In practice, Clock frequency impacts performance
  of DDS.
• If output frequency is chosen to be an exact
  fraction of clock frequency, spurious outputs are
  reduced.