ECE 495 — Electromagnetic Warfare and Survivability

Block 2: RF Electronic Warfare
Lesson 12: Electromagnetic Support

Juan Jurado, Col, USAF, Ph.D.
Permanent Professor and Head
Electrical and Computer Engineering
Fall 2026

←→ navigate · C draw · B board · S speaker view · F fullscreen · ? all shortcuts

Block 2 · Lesson 12 · Where We Left Off

Where We Left Off

L11 made the case: the listener detects at $R^2$ while the radar pays $R^4$. The B-21 hears every radar on the battlefield long before any of them sees it.

Hearing is not understanding. Thousands of pulses per second arrive interleaved, from every direction, on every band.

Today: the box that turns that chaos into a threat display — the RWR.

Block 2 · Lesson 12 · Objectives

Lesson Objectives

By the end of this lesson, you will be able to:

Objectives

Distinguish ES from ELINT by purpose and data depth
Name the four functions of every threat warning system
Walk the RWR block diagram from antenna to cockpit display
Explain how pulse parametrics turn raw intercepts into a threat ID

Block 2 · Lesson 12 · ES vs ELINT

ES vs ELINT: Same Spectrum, Different Job

Collects just enough to answer: what radar, what mode, right now
Bandwidth-bounded by known threats

ELINT

Collects everything: full parametric depth, future threats, library building
Bandwidth-bounded by any possible threat

ELINT’s product becomes the threat library the RWR classifies against.

ES fights the fight — ELINT studies the fight.

Block 2 · Lesson 12 · Four Functions

Four Functions of Every Threat Warning System

Function	What it does	OODA phase
Observe	Monitor the radar bands continuously	Observe
Detect	Recognize a threat signal is present	Orient
Classify	Determine radar type and operating mode	Decide
Alert	Warn aircrew and cue defensive systems	Act

Same four functions in every threat warning system, RF or IR
The hard engineering lives in Classify — the rest is plumbing and ergonomics

Block 2 · Lesson 12 · RWR Architecture

The RWR, End to End

Antennas

→

Receiver

→

Processor

→

Display

↳

Other EW systemsfrom the processor

Antennas — typically four, one per quadrant; small, broadband, low gain
Receiver — measures each pulse’s parameters
Processor — sorts, classifies, decides what matters
Display + cueing — aircrew scope, plus chaff / jammer / decoy cues

Block 2 · Lesson 12 · Bandwidth

The RWR’s Bandwidth Dilemma

The RWR wants to hear everything: all bands, all directions, all the time. Physics sends the bill:

Minimum detectable signal

\[ S_{\min} = kTB \cdot \text{NF} \cdot \text{SNR}_{\text{req}} \]

Same $S_{\min}$ contract as L8 — but now $B$ spans tens of GHz, not MHz
$\text{SNR}_{\text{req}}$ is a single-pulse requirement — L8’s integration gain is absent: the RWR cannot integrate the threat’s pulses the way a radar integrates its own echoes
Every decade of bandwidth costs 10 dB of noise floor
Saved by the one-way advantage: radar main lobes arrive loud

For warning, wide-open and deaf-ish beats narrow and sharp.

Block 2 · Lesson 12 · Receivers

Five Ways to Build the Receiver

Type	Sensitivity	Character
Crystal video	Low	Wide coverage, fast, one signal at a time
IFM	Low	Frequency only, up to an octave
Superheterodyne	Med–high	Selects one signal, recovers any modulation
Channelized	Med–high	Multiple simultaneous signals
Digital	Med–high	Wide, fast, flexible — the modern default

Classic RWRs leaned on crystal video + IFM: cheap, fast, wide
Modern RWRs digitize wholesale and let the processor do the selecting

Block 2 · Lesson 12 · Pulse Parametrics

Pulse Parametrics: The PDW

Each intercepted pulse becomes a pulse descriptor word (PDW). What you can extract grows with how many pulses you wait for.

One pulse — everything measurable as it arrives

frequency · pulse width · TOA · AoA · power · polarization · modulation

Two pulses — difference the times of arrival

Pulse repetition interval (PRI) — the single most diagnostic number

Many pulses — wait for the beam to come around again

staggered PRI patterns · carrier modulation · antenna scan rate

Easy and quick measurements run first; if the ID resolves, stop. Scan rate takes longest.

Block 2 · Lesson 12 · Deinterleave

Deinterleave, Then Classify

The receiver hears one merged stream of pulses from every emitter in view. The processor must take it apart — and it does it with the same three tiers.

Deinterleave — cluster PDWs into per-emitter buckets

by frequency, AoA and PW — the one-pulse parameters

Estimate PRI — difference the TOAs inside each bucket

the two-pulse measurement, now one per emitter

Classify — match the bucket against the threat library

frequency · PRI · PW · scan

Determine mode — search, track, or launch

PRF jumps betray the handoff

The RWR’s first job is sorting, not hearing.

Block 2 · Lesson 12 · Blind Spots

When the RWR Is Blind

The RWR only warns about things that radiate:

Nothing on the scope

LPI radars — spread, low-power waveforms below the RWR’s detection floor
EMCON — the threat simply doesn’t transmit until late
Passive seekers — IR and EO weapons give no RF launch indication
Silent terminal phases — GPS / inertial / home-on-jam guidance leaves the scope quiet until impact

This is why survivability never rides on one sensor — B3 covers the IR half.

Today’s Interactive Demo: RWR Scope

Companion script: L12_RWRPulseSort.m
In L13 we add the question the PDW can’t fully answer yet: where is it?

Live demo

RWR Scope

usafa-ece.github.io/ece-495-ew

Demo hand-off; drive it from here. Start in the merged view: three emitters’ pulse trains arrive on one receiver channel as a single time-of-arrival stream. Ask them to read a PRI off it by eye before touching anything. Hit Deinterleave: the stream clusters on MEASURED frequency into per-emitter lanes, and each lane’s PRI is estimated as the median gap between its times of arrival. Point out that the frequencies and PRIs in the table are measurements, not lookups — the receiver is carrying fifty megahertz of frequency error and two percent of TOA jitter, so the numbers land near the library values rather than on them. A cluster is named only if it falls within ten percent of a library entry in both frequency and PRI; otherwise it reads UNKNOWN. Now raise the clutter slider: spurious pulses that will not cluster into anything land in the rejected lane, and the count under the table climbs. Push the slider to the top and the early-warning radar — the sparsest emitter in the scene, four pulses in the window — is the first ID to fail. Ask them why that one and not the tracker. Close with the mystery emitter: turn it on and have the class call it from its parametrics before anyone clicks. Companion script runs the same pipeline in MATLAB; they run it on their own.

Block 2 · Lesson 12 · Quick Exercise

Quick Exercise / Type-along: Deinterleave the Stream

Type this with me — the B1 cast, merged into one receiver channel.

pri = [5.0 2.0 0.1];  f = [0.5 3.0 10.0];   % ms and GHz
toa = [];  fq = [];                    % EW, ACQ, TTR
for k = 1:3
    t   = pri(k)/3 + (0:pri(k):20);    % 20 ms, staggered
    toa = [toa t];  fq = [fq f(k)+0*t];
end
[toa,i] = sort(toa);  fq = fq(i);  nv = median(diff(toa));
fprintf('%d PDWs; naive PRI %.3f ms\n', numel(toa), nv);
fprintf('  f(GHz)     N   PRI(ms)   PRF(Hz)\n');
for k = 1:3
    d = diff(toa(fq == f(k)));         % deinterleave, then diff
    fprintf('%7.1f %5d %9.2f %9.0f\n', f(k), numel(d)+1, ...
            median(d), 1e3/median(d));
end

The naive PRI names only one emitter. Which one, and what erased the other two?

Then: which row goes on the aircrew’s scope first, and why?

Block 2 · Lesson 12 · Wrap-Up

Wrap-Up

Key ideas

ES answers what and now; ELINT builds the library it answers from
Observe, Detect, Classify, Alert — Classify is where the engineering lives
PDWs + PRI turn an interleaved pulse stream into named threats with modes
A silent scope is not a safe sky

Next lesson

L13: Angle of arrival — amplitude comparison and interferometry
Pre-flight quiz for L13 due before class

Closing question: the 10 GHz cluster on your scope goes silent — power gone, nothing else changes. Are you safe?