1. Environment Simulation – The S-Parameter
Doc.: IEEE /0719r0
July 2005
July 2005
Environment Simulation – The S-Parameter
Date:
Authors:
Notice: This document has been prepared to assist IEEE It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein.
Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE
Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures < ieee802.org/guides/bylaws/sb-bylaws.pdf>, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE Working Group. If you have questions, contact the IEEE Patent Committee Administrator at
Dr. Michael D. Foegelle, ETS-Lindgren
Dr. Michael D. Foegelle, ETS-Lindgren

2. Doc.: IEEE /0719r0
July 2005
July 2005
Abstract
This presentation introduces the concept of S-Parameters, which are a well known standard way of describing the RF properties of a multi-port black box, as a way of characterizing the basic properties of any RF channel, including antennas and environment, from transmitter port(s) to receiver port(s).
It will be shown that this concept can be used to achieve the same results as a real-world environment, including the effects of interferers, when evaluating the performance of an radio and higher layer metrics.
Dr. Michael D. Foegelle, ETS-Lindgren
Dr. Michael D. Foegelle, ETS-Lindgren

3. Overview Introduction The S-Parameter Multipath Environments
July 2005
Overview
Introduction
The S-Parameter
Multipath Environments
Environments
Metrics
Measurement Framework
Environment
Dr. Michael D. Foegelle, ETS-Lindgren

4. July 2005
Introduction
Many of the metrics that TGT is interested in are influenced by the physical channel used.
There is a desire to evaluate these metrics as a function of variations in the physical channel.
It would be nice to have a simple way to test the effects of the physical channel on these metrics without having to result to creating real-world physical channels (real-world environments) for each variation we wish to test.
Understanding the concept of S-Parameters will show how this can be accomplished.
Dr. Michael D. Foegelle, ETS-Lindgren

5. July 2005
The S-Parameter
S-Parameters (short for “Scattering Parameters”) represent the transmission and reflection coefficients between incident and transmitted or reflected waves.
They completely characterize the behavior of a device under linear conditions.
RF cables, passive antennas, and most radiating environments are linear devices.
Dr. Michael D. Foegelle, ETS-Lindgren

6. The S-Parameter The simplest & most common case is the two-port model.
July 2005
The S-Parameter
The simplest & most common case is the two-port model.
An RF cable is a two-port device.
So is a pair of antennas in a real-world environment.
However, multi-port networks can be represented too.
Dr. Michael D. Foegelle, ETS-Lindgren

7. July 2005
The S-Parameter
If ai is the complex voltage of the wave going into port i, and bj is the complex voltage of the wave coming out of port j,
Then the complex S-Parameter for any port combination i and j is defined as:
when all other ports are all terminated into matched loads.
Dr. Michael D. Foegelle, ETS-Lindgren

8. July 2005
The S-Parameter
Representing this graphically then, a two port network has four S-parameters, S11, S21, S12, and S22.
The Sii terms (i = j) represent reflectivity of Port i.
The Sij terms (i  j) represent the transmission coefficient from Port j to Port i.
Dr. Michael D. Foegelle, ETS-Lindgren

9. The S-Parameter For a multi-port network with N ports:
July 2005
The S-Parameter
For a multi-port network with N ports:
There N reflectivity S-Parameters (Sii), one for each port.
There are N(N-1) transmission parameters (Sij, i  j), two for each possible port pair.
For a total of N2 S-Parameters.
Dr. Michael D. Foegelle, ETS-Lindgren

10. July 2005
The S-Parameter
The relationship between all of the input signals to the various ports, ai, and the resulting output signals of the various ports, bi, can be represented in terms of the S-Parameters, Sij , in matrix form:
Dr. Michael D. Foegelle, ETS-Lindgren

11. July 2005
The S-Parameter
The neat thing about looking at S-Parameters this way is that matrices can be multiplied.
That is, if and then if
we get
where is some new S-Parameter matrix.
Dr. Michael D. Foegelle, ETS-Lindgren

12. July 2005
The S-Parameter
This is just a fancy way of saying that networks can be combined:
Dr. Michael D. Foegelle, ETS-Lindgren

13. July 2005
The S-Parameter
Or, looking at it another way, we can measure all of the components of a system (antennas, cables, etc.) at once:
Dr. Michael D. Foegelle, ETS-Lindgren

14. July 2005
The S-Parameter
or we could measure each component and combine the results:
Dr. Michael D. Foegelle, ETS-Lindgren

15. July 2005
Examples
Now let’s see how this concept applies to our RF Physical Channel by examining some wave propagation cases.
Since we know that we can combine S-Parameters for different components, we can ignore the effects of cables, antennas, etc. and just look at the behavior of point sources in an environment.
The complex electric field at a point some distance d away from a point source is given by:
where E0 is a magnitude constant and  is the wavelength of interest.
Dr. Michael D. Foegelle, ETS-Lindgren

16. July 2005
Examples
The value E0 is related to the voltage output from the transmitter circuitry by some complex S-Parameter representing the loss of the transmit cables, the gain of the transmit antenna, etc.
Similarly, the voltage seen by the receive circuitry is related to the electric field applied to the receive antenna by some other complex S-Parameter that represents the receive cables and antenna.
Thus, the transmission coefficient S-Parameter, S21, of everything between the transmitter and receiver is also proportional to e-i2d//d.
Dr. Michael D. Foegelle, ETS-Lindgren

17. July 2005
Examples: Single Path
Applying this to a single path at a 3m separation distance
Dr. Michael D. Foegelle, ETS-Lindgren

18. July 2005
Examples: Single Path
Assuming K is constant, the magnitude and phase vs. frequency becomes:
Dr. Michael D. Foegelle, ETS-Lindgren

19. July 2005
Examples: Single Path
From this information, we can determine a time domain transform:
Dr. Michael D. Foegelle, ETS-Lindgren

20. Examples: Single Path Dividing by c, the speed of light, we get:
July 2005
Examples: Single Path
Dividing by c, the speed of light, we get:
Dr. Michael D. Foegelle, ETS-Lindgren

21. July 2005
Examples: Multi-Path
Expanding this to represent a Multi-Path environment with direct path of 3 m and reflected paths of 5 and 10 m.
Dr. Michael D. Foegelle, ETS-Lindgren

22. July 2005
Examples: Multi-Path
Expanding this to represent a Multi-Path environment with direct path of 3 m and reflected paths of 5 and 10 m.
Formulation becomes:
K values include directionality of antennas, losses in reflections, etc.
Again, for simplicity we’ll let the Ks = 1
Assumes perfect reflections and omnidirectional antennas.
Dr. Michael D. Foegelle, ETS-Lindgren

23. July 2005
Examples: Multi-Path
Direct and reflected paths “beat” against each other vs. frequency.
Dr. Michael D. Foegelle, ETS-Lindgren

24. July 2005
Examples: Multi-Path
Transforming to time domain shows the three signals arriving at different times corresponding to signal path length.
Dr. Michael D. Foegelle, ETS-Lindgren

25. July 2005
Examples: Multi-Path
Remember, S-Parameter is a qualification of a black box.
Dr. Michael D. Foegelle, ETS-Lindgren

26. July 2005
Examples: Multi-Path
All we know about our black box is what we measure externally in the way of its S-Parameters.
Dr. Michael D. Foegelle, ETS-Lindgren

27. July 2005
Examples: Multi-Path
So, if we measure this data, what can we tell about what is inside our black box?
Dr. Michael D. Foegelle, ETS-Lindgren

28. July 2005
Examples: Multi-Path
S21 only tells us about distances (times) to reflections.
Dr. Michael D. Foegelle, ETS-Lindgren

29. July 2005
Examples: Multi-Path
S21 only tells us about distances (times) to reflections.
It has no information about location (direction)!
Dr. Michael D. Foegelle, ETS-Lindgren

30. July 2005
Examples: Multi-Path
There are many black boxes that can produce the same S21.
Dr. Michael D. Foegelle, ETS-Lindgren

31. July 2005
Examples: Multi-Path
The same result can come without even having a direct path!
Dr. Michael D. Foegelle, ETS-Lindgren

32. July 2005
Examples: Multi-Path
Or even from multiple antennas connected to the same source!
Dr. Michael D. Foegelle, ETS-Lindgren

33. Examples: Multi-Path Given that:
July 2005
Examples: Multi-Path
Given that:
many different configurations can generate the same S-Parameters
the transmitter and receiver interact across the RF physical channel solely through these S-Parameters.
Dr. Michael D. Foegelle, ETS-Lindgren

34. July 2005
Examples: Multi-Path
Then any black box that has the desired S-Parameters can be used as the test communication channel!
Dr. Michael D. Foegelle, ETS-Lindgren

35. July 2005
Examples: Multi-Path
Then any black box that has the desired S-Parameters can be used as the test communication channel!
It doesn’t even have to be a radiated signal path.
Dr. Michael D. Foegelle, ETS-Lindgren

36. Examples: Interference
July 2005
Examples: Interference
How does this help us for other environmental effects like interference?
Dr. Michael D. Foegelle, ETS-Lindgren

37. Examples: Interference
July 2005
Examples: Interference
Remember that our black box can have as many ports as we want.
Thus, we can introduce an interference signal to another port and test the effects of varied signal levels and phase relationships with receiver.
Dr. Michael D. Foegelle, ETS-Lindgren

38. Examples: Interference
July 2005
Examples: Interference
Again, this can easily be done in a cabled environment.
Dr. Michael D. Foegelle, ETS-Lindgren

39. July 2005
Equipment
The first approximation to emulating an environment is the use of a variable attenuator.
This adds magnitude variation, but not phase control.
A brute force method for emulating multiple paths involves using splitters/combiners and different cable lengths as delay lines.
Provides necessary phase change vs. frequency behavior and basic path loss.
Can be coupled with a variable attenuator to vary path loss.
Relatively inexpensive way to create some specific repeatable test cases, but rather clunky.
Dr. Michael D. Foegelle, ETS-Lindgren

40. July 2005
Equipment
Commercial Wireless Channel Emulators currently exist that are capable of generating programmable multi-path behavior capable of emulating real-time fading scenarios over many paths (up to 24 on models I’m aware of) with multiple I/Os allowing testing interaction of multiple signals.
Dr. Michael D. Foegelle, ETS-Lindgren

41. OTA OTA measurements ARE NECESSARY.
July 2005
OTA
OTA measurements ARE NECESSARY.
However, they are only necessary to determine the specific RF behavior of the antenna and DUT in normal operation.
While it is possible to measure the antenna performance separately and combine it with the conducted performance of a device, cables, packaging, etc. affect the radiated performance of the antenna.
Effect of the impedance of the antenna (S11) can have a significant effect on the radio circuitry.
Impedance can change between test case and in-use operation.
Radiate signals from antenna can cause internal interference with circuitry and unintentional radiation from circuitry can interfere with DUT performance (Receiver Desensitization).
Dr. Michael D. Foegelle, ETS-Lindgren

42. July 2005
OTA
While all metrics could be measured OTA, they all will be affected by the exact same set of RF communication channel affects.
These effects, related to antenna performance on the DUT, can be determined using only a small subset of the total set of metrics.
Best solution is to perform OTA measurements that combine an antenna pattern measurement with transmitter and receiver performance measurements.
All other tests can be performed conducted with OTA results used as modifiers.
Dr. Michael D. Foegelle, ETS-Lindgren

43. Conclusions and Summary
July 2005
Conclusions and Summary
Many of the metrics that TGT is interested in are influenced by the physical channel used.
There is a desire to evaluate these metrics as a function of variations in the physical channel.
It is not necessary to use real-world physical channels (real-world environments) in order to determine what their effect would be on these metrics.
The effect of the RF physical channel on the DUT(s) is defined by the S-Parameters of that channel.
Dr. Michael D. Foegelle, ETS-Lindgren

44. Conclusions and Summary
July 2005
Conclusions and Summary
There is only a limited amount of information about the environment that can be obtained from measurements of the physical channel.
Thus, it is possible that many different physical channels can have the same S-parameters.
Physical channels can have multiple inputs/outputs.
This tells us that it’s possible to emulate any real-world physical channel with appropriate test equipment.
Conducted tests using variable attenuators is the first step in this direction.
More accurate methods include adding delay lines or using commercial channel emulators.
Dr. Michael D. Foegelle, ETS-Lindgren

45. Conclusions and Summary
July 2005
Conclusions and Summary
OTA tests are still critical, primarily for determining true performance of the antenna in its use case.
OTA tests only need be performed for a subset of the desired metrics, primarily focused on transmit (total radiated power) and receive (total isotropic sensitivity) performance.
Results from these tests can be used to adjust results from conducted tests of other metrics by comparing conducted power and sensitivity with the OTA results.
Note: This is the model used by the CTIA for performance testing of mobile phones and similar methodology is currently being developed for converged devices.
Dr. Michael D. Foegelle, ETS-Lindgren