QPSK Acoustic Software Radio

1
QPSK Acoustic Software Radio

• Orr Srour Naftali Zon
• Under the supervision of
• Ami Wiesel

2
The problem

• RF Communication systems, and especially MIMO
  communication systems are
• expensive
• have long development time
• require wide technological knowledge

3
The need

• Besides their public use, in the academic world
  communication systems are needed for vary of
  reasons, for example new algorithms testing.
• In many cases, making a real communication system
  is simply unreasonable.

4
The goal

• Real time communication system.
• Rapid development
• Easy to construct, manipulate and upgrade
• Modular

5
The solution

• Using ACOUSTIC waves instead of electromagnetic.
• Fully software implemented - Simple ordinary
  computer is enough
• We will use the computer as our processing unit
  (we use Matlab-Simulink), and ordinary speakers
  and microphones as our antennas.

6
Systems overview - TOC

• Simple SISO receiver/transmitter system
• Virtually linked SIMO antenna selection
• Fully SIMO system running the Alamouti space-time
  algorithm

7
System Parameters

• We use QPSK modulation
• Data frequency 50 symbols/sec 100
  bit/sec
• Carrier frequency 800Hz

8
Communication Protocol

• We send data in packets.
• Each packet is constructed as follows

13 symbols
50 symbols
9
Training Packets

• The training sequence are constant predefined
  series of symbols.
• They allow us to distinguish actual data packet
  from random noise.
• They allow us to estimate the propagation channel
  and reconstruct the data.

10
Alamouti Transmit Diversity Technique

• The Alamouti diversity technique allows us to
  send data from two antennas to one with the
  highest theoretical SNR possible, without the
  need of a delay system or pre-knowledge of the
  channel.
• To do that, the transmission is done using 2
  antennas as follows

11
Alamouti Transmit Diversity Technique

• Where S0 and S1 are two data symbols after QPSK
  modulation.

12
Alamouti Transmit Diversity Technique

• In order to reconstruct the data, the following
  mathematical function is preformed

• Here h0 and h1 are the two channels propagation
  factors estimated by the detectors (see below)

13
Down to Top overview our basic building
blocks

• We will now introduce our basic building blocks,
  which will later be shown inside the different
  type of systems.

14
Packets Creator

• This unit simply receives bits and returns them
  in packets according to the mentioned protocol.

15
M-PSK Modulator/Demodulator

• These units transform between complex phase
  symbols and integer symbols

16
Raised Cosine Filter

• This unit both upsamples and filters the input
  signal, using raised consine filter.

17
Amplification Vector Creator

• When the bits of data are being transmitted, each
  bit has a unique amplification factor that
  determines its amplitude.
• These bits are sorted in a vector created in the
  "Amplification Vector Creator unit".

18
Up Mixer / Down Mixer

• The up-mixer block is in charge of shifting the
  incoming (complex) data into a real carrier
  signal.
• The down-mixer has the opposite functionality.

Amp
19
Squaring Timing Recovery

• The Squaring Timing Recovery block is in charge
  of sampling the incoming signal at the right
  time. It uses the knowledge of the number of
  constant phase samples in the incoming signal.

20
The Estimator

• The estimator block uses the predefined reference
  training signal in order to estimate the free-air
  channel propagation factor.
• We assume here that the channel can be modeled by
  a complex number, representing the attenuation,
  delay and noise the signal has suffered, and that
  this complex number will not change within the
  transmission of one data packet.

21
The Estimator cont

• All this is done by correlating the incoming
  signal with the training sequence.
• This block
• rises a trigger flag representing the reference
  signal has been discovered and is now over.
• samples and holds the conjugate phase of the
  correlation result
• samples and holds the amplitude of the
  correlation result.

22
The Detector

• The detector block is in charge of canceling the
  free-air channel effect of the incoming data.

23
The Signal Output block

• This block is in charge of sampling the actual
  data and regrouping the different packets into
  one long vector.

24
Antenna Selection Amplification For the antenna
selection system only

• This block is in charge of the creation of the
  amplification vector for the two antennas, in
  accordance to the "antenna selection" parameter
  received from the receiver.

25
EstimatorFor the antenna selection system only

• This block uses two ordinary estimators blocks
  (mentioned in the SISO system), each for
  different reference sequence.
• In addition to the ordinary estimator outputs, it
  also outputs the ID of the antenna which had the
  strongest signal in the receiver.

26
To AlamoutiFor the Alamouti system only

• This block is in charge of transforming the input
  data into 2 antennas input data, transformed by
  the Alamouti Diversity Technique mentioned before.

27
Solve by AlamoutiFor the Alamouti system only

• This block has the opposite functionality to the
  previous mentioned "To Alamouti" block. Its
  purpose is to demodulate the input signal using
  the Alamouti Diversity Technique.

28
Putting it all togetherThe SIMO Transmitter
29
Putting it all togetherThe SIMO Receiver
30
Putting it all togetherThe SIMO Receiver
Demonstration
31
Putting it all togetherThe Antenna Selection
32
Putting it all together The Antenna Selection
Demonstration
33
Putting it all togetherThe Alamouti Transmitter
34
Putting it all togetherThe Alamouti Receiver
35
Putting it all together The Alamouti Receiver
Demonstration
36
Future Applications

• short-range communication system on a single chip
  (using ultrasound waves)
• submarine communication systems
• implementation of MIMO sonar system

37
Questions
?