Electromagnetic Interference Reduction Study using a Self-Structuring Antenna

Transcription

1 Electromagnetic Interference Reduction Study using a Self-Structuring Antenna A. M. Patel (1), E. J. Rothwell* (1), L.C. Kempel (1), and J. E. Ross (2) (1) Department of Electrical and Computer Engineering Michigan State University, East Lansing, MI (2) John Ross & Associates, Salt Lake City, UT URSI 07 Thursday, 1

2 Overview Self-Structuring Antenna (SSA) Overview Goals Motivation Simulation Setup Conclusions Future Work 2

3 SSA Overview The Self-Structuring Antenna (SSA) is an adaptive antenna that changes its electrical shape in response to the environment by controlling electrical connections between the components of a skeletal template. The SSA automatically configures itself to accommodate changes in signal strength, orientation, and atmospheric conditions through the control of simple on/off switches. Changes in switch states cause the electrical shape of the antenna to be altered, allowing it to adjust to changes in its electromagnetic environment. The effect of different antenna configurations is unknown to the designer, only a statistical approach is utilized in testing. 3

4 SSA Overview The template is comprised of a large number of wire segments or patches interconnected by controllable switches. For a template with n switches, there are 2 n possible configurations. The template can be highly structured or random and can be placed on a planar or conformal surface. For each configuration, the states of the switches determine the electrical characteristics of the antenna. An asymmetric topology provides more diversity and less repeated states than a symmetric topology. 4

5 SSA Overview Array of wires or patches interconnected with electronically controlled switches SELF- STRUCTURING ANTENNA TEMPLATE. m. control lines antenna feed line feedback control Receiver/Transmitter with feedback signal for VSWR, S-Meter, BER, etc SENSOR MICROPROCESSOR Uses smart (evolutionary) algorithms to select switch positions which optimize feedback signal 5

6 Research Goals Overall goal is to investigate the ability of an SSA to reduce interference between two antennas. As a simple starting point: the ability of an SSA to receive a plane wave signal from one direction, while rejecting a signal from a different direction, is investigated. This is a receiving problem. The goal is to not only achieve an acceptable level of rejection for the unwanted signal, but also to ensure that the antenna is reasonably well-matched to the desired signal. How close can we bring the two incident waves and still be able to reject one signal and accept the other? 6

7 Motivation Electromagnetic Interference (EMI): Electromagnetic radiation generated by an external source that induces an unwanted signal into the circuitry of a device disrupting the performance of the device Directly Through the antennas Unintentional Interference By-product of operation of another device Intentional Interference Cell phones, broadcast transmitters, wireless internet Jamming signals Reducing the vulnerability of systems to EMI has become a major focus. 7

8 Simulation Setup Simulation approach will be used to search for SSA configurations that: Maximize the receive power of the wanted signal. Minimize the receive power of the interference signal. Create the largest difference between the reception of the two signals but ensure that the receive power for the desired signal is comparable in magnitude to what the optimum receive power for that template is. Fitness function is constructed to meet these objectives. GA-Suite (John Ross and Associates) General purpose Genetic Algorithm optimizer and GUI that interfaces to: Berkeley Spice LLNL NEC-2 and NEC-4 Ohio State University BSC 4.x Developed by John Ross 8

9 Simulation Setup C++ code written to generate templates. NEC- 4 used as EM solver. Optimization performed by GA-Suite based on a fitness function. Fitness A single numerical quantity describing how well an individual meets predefined design objectives and constraints. Often based on the outputs of multiple analyses using a weighted sum. Fitness function used is based on how close the reception for the wanted signal is to an upper constraint in addition to how close the reception for the unwanted signal is to a lower constraint. Priority is placed on reaching constraint for wanted signal (to ensure good reception of the desired signal) by using weights. Generally use following setup: Simple GA with elitist strategy Single point crossover Population size: 50 Mutation probability = 0.1 Crossover rate = 0.06 Maximum generations = 12 to Ratio of weight on wanted signal constraint to weight on unwanted signal constraint = 3:2 Upper and lower constraints are set in a manner that will encourage at least db rejection Simulation terminates as soon as constraints are met. 9

10 Simulation Setup SSA OVERVIEW GOALS MOTIVATION SIMULATION SETUP Number of panels = 5 Frequency = 0 MHz Size: = 4 x 3 wavelengths = 3 x 2.25 meters Number of switches = 27 Number of possible configs. = 134,217,728 Polarization (η) = 90 degrees Receive Impedance = 50 Ohms RESULTS CONCLUSIONS FUTURE WORK Number of panels = Frequency = 0 MHz Size: = 4 x 3 wavelengths = 3 x 2.25 meters Number of switches = 57 Number of possible configs. = 1.44 E+ 17 Polarization (η) = 90 degrees Receive Impedance = 50 Ohms

11 Simulation Setup Specification of an incident plane wave: For each desired signal, pick 6 undesired signals for each angle of separation (β) Find 6 solutions to equation: cos(β) = sin(θ a )sin(θ r ) [cos(φ a )cos(φ r ) + sin(φ a ) sin(φ r )] + cos(θ a )cos(θ r ) 11

12 Accept Signal: θ = 60 Φ = 45 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

13 Accept Signal: θ = 60 Φ = 45 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

14 Accept Signal: θ = 60 Φ = 45 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

15 5 PANEL Accept Signal: θ = 60 Φ = 45 Reject Signal: θ = 21.7 Φ = 250 Separation Angle = 80 15

16 PANEL Accept Signal: θ = 60 Φ = 45 Reject Signal: θ = 21.7 Φ = 250 Separation Angle = 80 16

17 Accept Signal: θ = 60 Φ = 45 Dependence of Rejection on Angle of Seperation for Accept Signal: Theta = 60deg Phi = 45deg 5 Panel Template Rejection(dB) Degrees of Separation between Desired and Undesired Signals (β) Dependence of Rejection on Angle of Seperation for Accept Signal: Theta = 60deg Phi = 45deg Panel Template Rejection (db) 5 60 Degrees of Separation between Desired and Undesired Signals (β)

18 Accept Signal: θ = 60 Φ = 45 Six Signal Average Rejection vs. Seperation Angle for Theta = 60deg, Phi = 45deg Template Comparison Six Signal Average Rejection (db) 15 5 panel panel Degrees of Separation between Desired and Undesired Signals (β) 80 panel 5 panel Template 18

19 Accept Signal: θ = 30 Φ = 1 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

20 Accept Signal: θ = 30 Φ = 1 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

21 Accept Signal: θ = 30 Φ = 1 Angle of Separation (β) θ (reject) Φ(reject) Reject in db (5 panels) Reject in db ( panels)

22 Accept Signal: θ = 30 Φ = 1 Dependence of Rejection on Angle of Seperation for Accept Signal: Theta = 30deg Phi = 1deg 5 Panel Template Rejection(dB) Degrees of Separation between Desired and Undesired Signals (β) Dependence of Rejection on Angle of Seperation for Accept Signal: Theta = 30deg Phi = 1deg Panel Template Rejection(dB) 5 60 Degrees of Separation between Desired and Undesired Signals (β)

23 Accept Signal: θ = 30 Φ = 1 Six Signal Average Rejection vs. Seperation Angle for Theta = 30deg, Phi = 1deg Template Comparison Six Signal Average Rejection (db) panel panel Degrees of Separation between Desired and Undesired Signals (β) 80 panel 5 panel Template 23

24 Average Rejection vs. Beta for 5 panel Template Theta = 60, Phi = 45 Theta = 30, Phi =1 Average Rejection vs. Beta for panel Template A v e r a g e R e j e c t i o n ( d B ) A v e r a g e R e j e c t i o n ( d B ) Degrees of Separation between Desired and Undesired Signals (β) Degrees of Separation between Desired and Undesired Signals (β) 24

25 Average Rejection by Angle of Seperation (Beta) Average Rejection (db) Degrees of Separation between Desired and Undesired Signals (β) 25

26 Conclusions And Observations In general The ability to reject any specific undesired signal depends heavily on the number of panels. In many cases, good rejection is achieved with the 5-panel template but extremely poor rejection is achieved with the -panel template (or vice versa) for the same desired and undesired signal pair. For low β (under deg), undesired signal rejection increases monotonically as angle of β increases. Rejection appears to stagnate or even drop slightly for higher values of β. Rejection is higher on average for -panel template than for 5-panel template. More db+ rejections are feasible using -panel template than 5-panel template. -panel template gave less dispersed (lower standard deviation) rejection numbers for various undesired signals at a fixed β, especially at higher values of β. -panel template is capable of generating patterns with more complexity than 5-panel template. Ability to have large rejection is dependant on the desired signal as well as the undesired signal Accept Signal: θ = 30 Φ = 1 had an average rejection higher than Accept Signal: θ = 60 Φ = 45 regardless of template or separation angle β. 26

27 Future Work Explore dependence on: Template geometry Size Shape (layout of wires) Number of switches Number of panels Desired signal Incidence angle Polarization Undesired Signal Incidence angle Polarization Receiving impedance Frequency bandwidth Matching input impedance 27