The architecture of the aiLPI Radar option addresses almost all known techniques for minimizing the probability of intercept by a RWR (Radar Warning Receiver). The aiLPI Radar option when combined with the aiRadar Radar Programming Language (aiRPL), and the Radar Processing Unit (aiRPU) provide access to these techniques on our standard Research Radar Interferometer series of research instruments.
The aiRPU is, conceptually, as shown below
Technology Block Diagram
From an LPI Point of view, the key elements of the aiRPU are the aiRPU Tx Sequencer and the aiRPU BF Sequencer, and their associated memories (RAMs).
LPI radar systems aim to reduce the probability that an RWR can detect a threat signal in two ways. First, by carefully managing when and where power is transmitted through its radiation and scan patterns. Second, the radar waveform aims to reduce the average power density by using high duty cycles or continuous wave signals with advanced modulation techniques that cover wide bandwidths.
In particular, reducing the radar’s peak radiated power using some form of pulse compression techniques is a commonly used approach to realize LPI radar signals. Pulse compression based on intrapulse modulation is very effective in practice, leading to a large pulse energy, with the time-power product spread out over a large bandwidth. Other techniques have been exploited and developed recently to design radar waveforms that can increase the performance of LPI radar systems. LPI waveforms include Frequency Modulation Continuous Wave (FMCW), several polyphase-coded CW waveforms such as Frank, P1, P2, P3, and P4, frequency hopping, and combined frequency hopping-phase coding.
The power of aiRPL for LPI.
To fully appreciate the power of aiRPL for an LPI radar requires a basic understanding of aiRPL itself.
A simple example shows how a complex series of transmit pulses and PRIs may be written using aiRPL, and executed on the aiLPI configured RRI Research Radar. Of course, this example, by it’s simplicity, is by no means an exhaustive description of the almost infinite capability of aiRPL.
What this example illustrates is that one technique of LPI, hopping the PRI timing and the pause between PRIs, is relatively simple within aiRPL.
The example also illustrates how the transmit pulse itself may be hopped from pulse to pulse, or from pulse group to pulse group; in this case the example uses two predefined FMCW pulses, named “txfmcwup”, “txfmcwdown “.
Of course, a simple example of FMCW up and FMCW down also does not represent the almost infinite capability of aiRPL. This will be discussed a little further on.
The most important method in aiRPL is PRI (Pulse Repetition Interval), which generates a specified pulse repetition interval. It is defined like this:
PRI ( duration , TRC , BF , TX , RX)
The `duration’ parameter specifies the total duration of the PRI. The TRC, BF, TX and RX arguments are name strings which identify pre-defined hardware / firmware specific data structures which precisely specify the operation to be performed on the target hardware /firmware. For example, a TX structure designed for FMCW (one of many possible transmitted pulse types) would contain parameters, such as start frequency and slope, for the transmitter DDS (Direct Digital Synthesizer).
An example PRI could be:
PRI (5e-3, ” tr0rx12 “, ” bfwide7 “, ” txfmcw4 “, ” rxfmcw1 “)
In this example, a 5.0 millisecond PRI is programmed, accessing four structures, “tr0rx12”, “bfwide7”, “txfmcw4”, “rxfmcw1”. (Time values in aiRPL are specified in floating-point seconds, optionally using scientific notation, e.g. 1E-3 or 1e-3 to indicate a millisecond.)
The named data structures are assumed in this example to have been previously defined using the aiRPL Ecosystem tools, Acquisition Designer and maintained in the aiRPL database.
These data structures use a common header, defined in aiRPL, so that aiRPL can access and handle these structures during compilation, during uploads and during radar operation.
The following is an illustration of aiRPL’s looping construct, generating a sequence of 10 pairs of FMCW pulses, one a 5ms upchirp, and the next a 6ms downchirp, each with a changing pause of 1 millisecond, then 2ms between them.
REPEAT (10)
{
PRI (5.0e-3, ” tr0rx12 “, ” bfwide7 “, ” txfmcwup “, ” rxfmcw1 “)
PAUSE (1e-3)
PRI (6.0e-3, ” tr0rx12 “, ” bfwide7 “, ” txfmcwdown “, ” rxfmcw1 “)
PAUSE (2e-3)
}
aiRPL supports programmer-defined functions (implemented as macros) in addition to its built-in source methods.
A function definition could look like:
fmcwPulse (priInt1, priInt2, pause1, pause2, trcf , txwst1 , txwst2 , rxwst , bfwst )
{
PRI(priInt1 , trcf , bfwst , txwst1 , rxwst )
PAUSE (pause1)
PRI(priInt2 , trcf , bfwst , txwst2 , rxwst )
PAUSE (pause2)
}
The above REPEAT loop could then be written:
REPEAT (10)
{
fmcwPulse (5e-3,6e-3, ” tr0rx12 “, ” txfmcwup “, ” txfmcwdown “, ” rxfmcw1 “, ” bfwide7″)
}
The following is an example aiRPL program, which uses the above fmcwPulse function definition. An aiRPL program is defined within a main() call. After loading a set of beamforming tables, this program generates 8 sets of 32 pulses. Each set of pulses is identical. There is a one millisecond pause between each set of pulses. All text following a double slash `//’ to the end of the line is a comment.
main ()
{
LOAD LUT (10e-3, ” fmlut 15 “) // load beamforming tables
REPEAT (8) // outer loop repeats 8 times
{
REPEAT (32) // inner loop repeats 32 times
{
fmcwPulse (5.0e-3, 6.0e-3, 1e-3, 2e-3, ” tr0rx12 “, ” txfmcwup “, ” txfmcwdown “, ” rxfmcw1 g”)
PAUSE (1e-3)
}
}