Passive direction finding (DF) is an important part of today’s electronic warfare systems. Early threat detection and the need to limit friendly fire casualties have driven the development of many DF systems. With each specific application, a new and unique system configuration is necessary.
It is the intention of the Fox Hunters senior project group, working in conjunction with Syndetix Corporation, to design a passive DF system for the purpose of identifying friend or foe. The system shall operate in the 962 MHz to 1213 MHz range. The system will determine the azimuth direction of any intercepted signal in the specified range. In addition, the system will determine the signal characteristics in order to verify a friend or foe signal. The signals that will be classified as “friendly” are two military regulated forms of field communication. The Tactical Air Navigation (TACAN) channels are communication channels used by the allies that are part of the North Atlantic Treaty Organization (NATO). The TACAN system operates on 252 channels ranging from 962 MHz to 1213 MHz in 1 MHz increments. The other signals of interest are the Identification of Friend or Foe (IFF) interrogator/transponder system channels. The IFF system consists of two transmission frequencies, 1030 MHz (commonly referred to as the interrogation frequency) and 1090 MHz (commonly referred to as the transponder frequency). It is of importance to note that it will be the responsibility of the user to determine whether to classify a signal as friend or foe using the data that the system provides. The system will not be designed to make decisions, only to provide the information necessary to make decisions.
This project is divided into four modules: the antenna array, the receiver, and two phases of signal processing. The first phase in the determination of the azimuth direction of the incident signal, and the second phase is the determination of the characteristics of the signal for use in the IFF applications. The concept of amplitude comparison will be used for determining bearing of the incident signal.
The input of the system is the antenna array. This consists of four open-ended rectangular waveguide antennas arranged to provide 360-degree azimuth coverage. Christina Robinson designed the antennas. An open-ended rectangular waveguide antenna was chosen as the antenna element because it will provide suitable gain (on the order of 5 dBi) and beam-width for accurate amplitude comparison measurements. The waveguides are relatively simple and cost efficient to make. The radiating element of each waveguide is an E-field probe. The diameter of the probe was chosen to have a large diameter to increase the bandwidth of the antenna. The design parameters are shown below in Table 1.
Table 1: Design parameters for each open-ended waveguide antenna of the array.
The parameters of open-ended waveguide are depicted in Figure 3.An N type connector, UG58A/U is connected to the BNC UG290A/U RF connector. The BNC is then connected to the first stage of the receiver. Each waveguide antenna contains three tuning stubs to provide a VSWR of approximately 1.4:1. The half-power beam width of the antenna is 52 degrees at 1088 MHz.
The receiver unit was particularly complicated to design. Although changes have been made over the course of the design process, the initial block diagram of the system still holds. Figure 4 is the block diagram of one channel within a four parallel channel system. Each channel will be identical.
To ease in the understanding of the process, the parts will be explained in the order that the incoming signal encounters them.
Upon reception by the antenna, the signal will be fed into the front end micro-strip filter. This filter will have a pass band of slightly larger than the required 962-1213MHz in order to insure passage of all signals of interest. The filter will be an inter-digital micro strip filter, custom designed by John Delk, using coupled strip line theory. A prototype has already been fabricated and tested. The design will require some adjustments but the prototype was successful enough to provide confidence in the design methods.
After the front end wide band filter, the signal will be fed into the RF Low Noise Amplifier (LNA). The LNA for this part of the receiver will be the RF2361 supplied by RF Micro Devices. This part was chosen because of the wide dynamic range and stellar noise figure at the frequencies of interest, 1.4dB. The specifications for this integrated chip are included in Appendix D.
The signal will then be down converted to the intermediate frequency (IF) of 300MHz, subject to change, using frequency multiplication, commonly referred to as mixing. This process will be accomplished using a mixer and a common local oscillator (LO), common due to it being the same LO signal fed into all the parallel channels. The mixer that will be used for this process is the Analog Devices model AD8343. This mixer was chosen because of its broadband operation, up to 2.5 GHz. The specifications for this chip are also included in Appendix D.
The LO is a fairly complicated
circuit. It consists of a phase locked loop (PLL) circuit and a voltage
controlled oscillator (VCO). The VCO, without the PLL, would not be stable
enough for this application.To implement the PLL, an Analog Devices model
ADF4111 frequency synthesizer will be used. The ADF4111
is an integer N frequency synthesizer. This means that it compares the output
of the VCO, Fo, to a multiple of the reference frequency, Fref. If Fo ≠ N*Fref then
it applies a voltage to the VCO to correct the difference.
By changing the value of N, the receiver is incremented through the
band of interest in step sizes equal to the Fref. Figure 5 is a generic block diagram of a
For this application, a reference frequency of 1MHz will be used. This will allow the receiver to tune over the frequency band in increments of 1MHz because the PLL will be comparing to multiples, N, of 1MHz. This is exactly the channel spacing of the TACAN frequencies of interest.
The VCO that will be used to generate the mixer frequency will be the ELCO-807/128-02 manufactured by Emhiser Micro-Tech. Our sponsor recommended this manufacturer. The data sheet for this device is included in Appendix D.
Upon completion of the down conversion to the IF the signal will then be applied to a narrowband IF filter. This filter will be custom designed if time allows. If time does not allow, it will be purchased from a vendor yet to be determined.
Following the IF filter, the signal will pass through a cascade of IF amplifiers before proceeding to the next part of the system, the DF computer and the signal characterization unit. Syndetix Corporation, the project sponsor, will supply the cascaded amplifiers for this portion of the receiver. All relative schematics, data sheets, simulations and code used in the receiver design are included in the Appendices.
A received signal is completely described by three main characteristics: amplitude, frequency and phase. In direction finding applications, characterization is of primary importance because it is the signal characteristics that distinguish the friendly signal sources from others. Aircraft communicate with ground and with other aircraft via streams of pulses that contain embedded information. For this particular direction finding system, two main things must be known about the pulsed signals: the pulse width (PW) and the pulse repetition interval (PRI). Thus, the duration that a pulse is active high and the time interval between each consecutive pulse rise time must be measured, stored, and displayed. This module receives the signals from the receiver module that is capable of detecting overall signal frequencies within the range of 962 MHz to 1213 MHz in 1 MHz steps.
A particularly efficient and economical way to process these signals is by using the programmable logic devices (PLDs) such as Field Programmable Gate Arrays (FPGAs). For this project, devices from the FLEX10K and MAX7000 families are being implemented to perform such tasks as measurement and routing of the PW and PRI information, frequency division, multiple clocking schemes, and PLL/Attenuator information routing.
The accomplishments to date for this portion of the project include the completion, compilation, and simulation of the code for the frequency division, PW and PRI measurements, and information routing to the FIFOs. The programs were coded in AHDL (Altera Hardware Description Language) using Altera’s MAX+PLUSII Baseline software. Proper operation of the programs were verified in hardware as of Feb. 17, 2003.
As always, we will keep the bits of golden advice from our sponsor, Evan, fresh in our minds at all times:
“Test early! Test often!”
"If something goes wrong check the small and simple stuff first! (Is you power supply plugged in?...)"
One of the deterministic properties that this project needs to determine is the direction or bearing that a signal has relative to the position of the antenna. This is to be accomplished by using the amplitude of the incoming signal. The antennas are to be designed in a certain configuration. These antennas are supposed to be identical therefore having the same properties. The only properties that would change are the relative positions among each other.
To accomplish this DF technique the signal is taken from the receiver output. The output signal should have both a phase and amplitude associated with it. The amplitude of a signal should vary from antenna to antenna, depending on the relative location of a signal. This difference in amplitudes is used to triangulate a relative position of the signal.
The direction of the signal will be determined using an FPGA board. First the signals amplitude will be converted to digital signal using an analog to digital converter, which is one the board that has been provided by the people at Syndetix. This digital signal will then be compared among the other signals being received through the system. After some comparisons the output will be displayed on LED's. There will be four outputs. The value of these outputs will then be looked up on a table. Each degree of bearing will correlate to four different output values of the system.
Another issue is cost. To obtain a high quality signal at microwave frequencies is very expensive. This is particularly true in the receiver. The signal-to-noise ratio must be very high in order to be able to obtain a good quality signal for the signal characterization and DF module. The parts either need to be high quality (expensive) or the system needs to be perfectly designed. These decisions will be give and take.
The project at this point in time is about 75% complete. The antennas, the receiver board and the signal characterization module have all been designed, built and tested. The direction finding module will be completed at a later date. The microprocessor and the user interface will be built by Syndetix.
The Fox Hunter Team would like to give a great deal of thank to the following people:
All the staff at Syndetix
All the Professors in the ECE department