# Reading — Electromagnetic Support By the end of this lesson you should be able to: 1. Distinguish **ES** from **ELINT** by purpose and by the depth of data each collects. 2. Name the **four functions** common to every threat-warning system and map them to OODA. 3. Walk the **RWR block diagram** from antenna through receiver and processor to the cockpit display. 4. Explain how **pulse parametrics** turn raw intercepts into a named threat with a known mode. ## Hearing is not understanding L11 settled the asymmetry: the listener detects at one-way $R^2$ while the radar pays round-trip $R^4$, so the B-21 hears every radar long before any of them sees it. But what *hearing* delivers is thousands of pulses per second, interleaved, from every direction, on every band. A bare antenna voltage is not a threat picture. This lesson is about the box that turns that chaos into a labeled scope: the radar warning receiver (RWR). ## ES versus ELINT: same spectrum, different job Two communities listen to the same emissions for opposite reasons. **Electromagnetic support (ES)** collects just enough to answer one tactical question — *what radar, what mode, right now* — fast enough to matter to the aircrew flying through the threat. **Electronic intelligence (ELINT)** collects everything: full parametric depth, on every emitter it can find, including threats never yet fired in anger, all to build a library. The split sets the bandwidth each must cover. ES is bounded by *known* threats — it only has to recognize emitters already in the catalog. ELINT is bounded by *any possible* threat, because its job is to find the ones not yet catalogued. And the two are linked: ELINT's product *is* the threat library the RWR classifies against. :::{admonition} Key Concept :class: key-concept ES fights the fight; ELINT studies the fight. ES answers *what and now* against known threats; ELINT collects full parametric depth against any possible threat and produces the library ES matches against. ::: ## The four functions, mapped to OODA Every threat-warning system — RF or IR, on a fighter or a frigate — performs the same four functions, and they line up cleanly with the OODA loop: - **Observe** — monitor the radar bands continuously (*Observe*). Always listening across the whole coverage. - **Detect** — recognize that a threat signal is present in the noise (*Orient*). Something coherent arrived; flag it. - **Classify** — determine the radar's type and operating mode (*Decide*). This is the hard part. - **Alert** — warn the aircrew and cue defensive systems (*Act*). Put a symbol on the scope, hand the cue to chaff, jammer, or decoy. Note where the difficulty concentrates. Observe, Detect, and Alert are mostly plumbing and ergonomics — wide antennas, a threshold, a display. The hard engineering lives in **Classify**: resolving a merged pulse stream into named emitters with current modes. Everything below is really about how the RWR classifies. ## The RWR, end to end The RWR is a short pipeline, antenna → receiver → processor → display, with one branch out to the rest of the defensive suite. - **Antennas** — typically four, one per quadrant, giving all-aspect coverage and a coarse bearing from the start. They are small, broadband, and low-gain: a warning receiver trades antenna gain for the ability to hear across the whole spectrum at once. - **Receiver** — measures each pulse's parameters as it arrives. Its output is not audio or video but a stream of measured numbers, one set per pulse. - **Processor** — the brain. It sorts the merged pulse stream, classifies each emitter against the library, and decides what is worth the aircrew's attention. - **Display + cueing** — the cockpit scope, plus cueing lines that drive chaff dispensers, onboard jammers, and decoys. The same classification that draws the symbol also tells the countermeasures what they are fighting. ## The bandwidth dilemma The RWR wants to hear *everything*: all bands, all directions, all the time. Physics sends the bill through the same minimum-detectable-signal contract from L8, $$ S_{\min} = kTB \cdot \text{NF} \cdot \text{SNR}_{\text{req}}, $$ with $k$ Boltzmann's constant, $T$ the system temperature, $B$ the bandwidth, NF the noise figure, and $\text{SNR}_{\text{req}}$ the SNR needed for a reliable detection. The terms match a radar receiver's — but $B$ does not. A tracking radar lives in a few MHz around one carrier; an RWR must stay open across **tens of GHz** to cover every threat band at once. Since the noise floor rises linearly with $B$, each decade of extra bandwidth costs about 10 dB of sensitivity, so a wide-open RWR is far less sensitive than the radars it listens to. What rescues it is the L11 advantage: the RWR detects *transmitters*, not faint echoes, and a radar's main lobe arrives one-way and *loud*. For warning, wide-open and deaf-ish beats narrow and sharp — you cannot warn about a band you are not listening to. :::{admonition} Key Concept :class: key-concept An RWR pays for its tens-of-GHz coverage with a noise floor roughly 10 dB worse per decade of bandwidth than a narrowband radar receiver. It gets away with it because radar main lobes arrive one-way and loud — coverage matters more than raw sensitivity for warning. ::: ## Five ways to build the receiver There is no single right receiver; the design trades sensitivity, simultaneity, and the ability to recover modulation: - **Crystal video** — low sensitivity, but wide and very fast; one signal at a time. The classic cheap, instantaneous detector. - **Instantaneous frequency measurement (IFM)** — low sensitivity, frequency only, over about an octave. Fast, good for tagging a carrier. - **Superheterodyne** — medium-to-high sensitivity and very selective. Tunes to one signal and recovers any modulation, at the cost of looking at one thing at a time. - **Channelized** — medium-to-high sensitivity across many parallel channels, handling multiple simultaneous signals — the dense-environment answer. - **Digital** — medium-to-high sensitivity, wide, fast, flexible. The modern default: digitize the spectrum wholesale and let software do the rest. Classic RWRs leaned on crystal video plus IFM — cheap, fast, wide, good enough to warn. Modern RWRs digitize the band and push the selecting and classifying into the processor, where it can be reprogrammed as the library grows. ## Pulse parametrics: the PDW When a pulse arrives, the receiver reduces it to a **pulse descriptor word (PDW)** — a compact record of measured parameters. What you can extract grows with how many pulses you wait for: - **From one pulse:** carrier frequency, pulse width (PW), time of arrival (TOA), angle of arrival (AoA), power, polarization, and any intra-pulse modulation. - **From two pulses:** the **pulse repetition interval (PRI)** — the gap between successive pulses, and the single most diagnostic number an RWR measures. PRI (and its inverse, PRF) goes a long way toward naming an emitter and its mode. - **From many pulses:** staggered or patterned PRI sequences, carrier modulation, and the antenna **scan rate** — how often the threat's beam sweeps back across you. The processor runs the cheap, fast measurements first and stops as soon as the ID resolves. Scan rate takes longest because it cannot be hurried — you must wait for the beam to come around again to time its rotation. ## Deinterleave, then classify The receiver hears *one* merged stream — every emitter in view, pulses interleaved in time. The processor's first job is to take that apart: 1. **Deinterleave** — cluster the PDWs into per-emitter buckets using the parameters stable pulse-to-pulse: frequency, AoA, and PW. Pulses that share those almost certainly came from the same radar. 2. **Estimate PRI** — within each bucket, difference the TOAs to recover that emitter's repetition interval. 3. **Classify** — match the bucket's signature (frequency, PRI, PW, scan rate) against the threat library and name it. 4. **Determine mode** — decide whether the emitter is searching, tracking, or guiding a launch. A PRF jump betrays the handoff: a radar going from a leisurely search PRF to a high track PRF is the IADS kill chain advancing on your scope in real time. Reading that transition *is* kill-chain awareness. :::{admonition} Key Concept :class: key-concept The RWR's first job is sorting, not hearing. Deinterleaving the merged pulse stream into per-emitter buckets comes before any classification — and the mode read off the sorted stream (search → track → launch) is the kill chain made visible. ::: ## When the RWR is blind For all its reach, the RWR only warns about things that *radiate* in RF. Several real threats give it nothing: - **LPI radars** — low-probability-of-intercept waveforms spread their energy thin at low power, sliding below the RWR's detection floor even while tracking you. - **EMCON** — under emission control, the threat simply does not transmit until the last moment, denying the RWR anything to detect. - **Passive seekers** — infrared and electro-optical weapons emit no RF, so there is no launch warning on the scope. - **Silent terminal phases** — a weapon on GPS, inertial, or home-on-jam guidance can run its endgame with the seeker quiet, leaving the scope clean until impact. The takeaway is doctrinal: survivability never rides on a single sensor. A quiet scope is an ambiguous report, not a safe sky — Block 3 picks up the infrared half the RWR cannot see. ::::{admonition} Quick Exercise :class: quick-exercise Your RWR reports four PDW clusters. Call the threat and the mode for each: 1. 0.5 GHz, PRI 5 ms, wide PW, AoA drifting slowly aft. 2. 10 GHz, PRI jumped from 1 ms to 0.1 ms, steady AoA, rising power. 3. 3 GHz, PRI 2 ms, beam sweeps past every 6 seconds. 4. The 10 GHz cluster from Q2 goes silent — power gone, nothing else changes. :::{admonition} Solution :class: dropdown 1. **EW radar, searching.** Low band and long PRI mark an early-warning set; the slow aft drift is just the B-21 flying past its sector — routine, not a lock. 2. **TTR transitioning search → track.** The PRF jump (PRI dropping ten-fold), the steady AoA, and the rising power together mean it has stopped scanning and is staying on you while you close. Cue countermeasures. 3. **ACQ in search.** Mid-band with a moderate PRI, and the 6-second beam reappearance gives you its scan rate directly — a rotating acquisition radar. 4. **Ambiguous and dangerous.** Silence could be EMCON, or it could be a handoff to a silent semi-active or passive terminal phase. Either way, silence is not safety — treat the disappearance as a possible engagement, not the end of one. ::: :::: ## Wrap-Up ES answers *what and now* against known threats; ELINT collects full parametric depth and builds the library ES matches against. Every threat-warning system runs the same four functions — Observe, Detect, Classify, Alert — and the engineering all lives in Classify. The RWR pays for tens-of-GHz coverage with a degraded noise floor and gets away with it because radar main lobes arrive one-way and loud. PDWs and PRI turn an interleaved pulse stream into named emitters with modes, and the mode read off the scope (search → track → launch) is the kill chain made visible. Finally, a silent scope is not a safe sky: LPI, EMCON, and passive seekers all leave the RWR blind. Next, **L13 — Angle of Arrival** answers the question the PDW only hints at: not *what* is out there, but *where* is it — how a passive receiver turns AoA into a bearing.