Course Description
EEE 4550 Radar is the undergraduate introduction to radar systems — how a transmitted electromagnetic pulse returns information about the range, velocity and character of a distant object, and what limits how well it can do so. It draws together three strands that an electrical engineering student has previously met separately: electromagnetics supplies propagation and antennas, signals and systems supplies the processing, and probability supplies the detection theory.
The Statewide Course Numbering System description is a full syllabus in itself: “basic concepts of radar systems including: radar range equation, radar cross section calculations, random processes and noise, array antennas, beamsteering, Doppler and range processing, FM and CW systems, pulse compression, synthetic aperture radar, clutter.”
Two Florida institutions carry it, both at 3 credits: Florida A&M University as Radar, matching the statewide title, and Florida International University as Introduction to Radar Systems. The titles differ only in explicitness; this is title drift, not subject divergence. The FAMU–FSU College of Engineering bulletin lists prerequisites of EEL 3473 and EEL 3135, with EEL 4021 as a co-requisite.
⚠ Radar is unusually well matched to the Florida employment market — more so than almost any other elective in this prefix. The reasons are set out under Career Pathways, and they are concrete rather than aspirational.
Learning Outcomes
Required Outcomes
- Derive and apply the radar range equation, and use it to determine detection range from transmitted power, antenna gain, target cross section, receiver noise and required signal-to-noise ratio.
- Explain radar cross section as a physical quantity, compute it for canonical shapes, and describe its dependence on frequency, aspect angle and polarisation.
- Model receiver noise, compute noise figure and system noise temperature, and relate them to minimum detectable signal.
- Apply detection theory: the Neyman–Pearson criterion, probability of detection and probability of false alarm, receiver operating characteristics, and threshold setting.
- Apply matched filtering and explain why it maximises output signal-to-noise ratio.
- Determine range resolution from waveform bandwidth and velocity resolution from coherent processing interval, and explain the resulting design trade-offs.
- Analyse the Doppler effect in radar, compute Doppler shift, and explain range–Doppler ambiguity and the blind speeds it produces.
- Explain pulse compression, analyse linear FM (chirp) and phase-coded waveforms, and evaluate range sidelobes and their suppression.
- Analyse antenna parameters relevant to radar — gain, beamwidth, sidelobes, aperture — and the relationship between aperture size and angular resolution.
- Analyse phased array antennas, compute the array factor, and design electronic beamsteering including the grating-lobe constraint on element spacing.
- Characterise clutter statistically and analyse moving-target indication and Doppler filtering as clutter rejection methods.
- Explain the principle of synthetic aperture radar and why it achieves cross-range resolution independent of range.
- Analyse continuous-wave and frequency-modulated continuous-wave radar and identify the applications for which each is appropriate.
Optional Outcomes
- Analyse tracking radar and implement or evaluate a tracking filter (alpha-beta, Kalman).
- Analyse space-time adaptive processing for airborne radar.
- Analyse constant false alarm rate (CFAR) detection schemes.
- Analyse weather radar and dual-polarisation products.
- Analyse automotive radar architectures and their specific constraints.
- Analyse electronic warfare, jamming and low-probability-of-intercept waveforms.
- Analyse MIMO radar and distributed apertures.
- Implement a radar signal-processing chain in simulation and evaluate its detection performance.
Major Topics
Required Topics
- Radar fundamentals — pulsed and continuous-wave operation, pulse repetition frequency, duty cycle, unambiguous range, range gating, and the basic monostatic and bistatic geometries.
- The radar range equation — derivation from transmitted power through antenna gain, spreading loss, target scattering and receiver aperture; system losses; the maximum-range form; and the bistatic variant.
- Radar cross section — definition, canonical target RCS (sphere, flat plate, corner reflector, cylinder), Rayleigh, resonance and optical scattering regimes, aspect and frequency dependence, target fluctuation and the Swerling models.
- Random processes and noise — thermal noise, noise figure and system noise temperature, the statistics of noise in a receiver, envelope detection and the Rayleigh and Rician distributions.
- Detection theory — binary hypothesis testing, Neyman–Pearson detection, probability of detection and false alarm, ROC curves, integration of multiple pulses (coherent and non-coherent), and the integration improvement factor.
- Matched filtering and waveforms — the matched filter and its SNR optimality, the ambiguity function as a unified description of range and Doppler resolution, and waveform design trade-offs.
- Pulse compression — the energy-versus-resolution problem, linear FM chirp, compression ratio and time–bandwidth product, range sidelobes and windowing, and Barker and polyphase codes.
- Doppler and range processing — Doppler shift, coherent processing intervals, range–Doppler maps, ambiguity in range and velocity, blind speeds, and low/medium/high PRF trade-offs.
- Antennas and arrays — gain, beamwidth and aperture, sidelobe levels, the array factor, uniform and tapered illumination, electronic beamsteering by phase shift, grating lobes and element spacing, and beam broadening at scan.
- Clutter — surface and volume clutter, clutter-to-signal ratio, statistical clutter models (Rayleigh, Weibull, K-distribution), moving-target indication, delay-line cancellers and their frequency response, and Doppler filter banks.
- Synthetic aperture radar — the synthetic aperture principle, cross-range resolution and its independence from range, stripmap and spotlight modes, azimuth compression, and the basics of SAR image interpretation.
- FM and CW systems — unmodulated CW Doppler radar, FMCW range measurement by beat frequency, the range–Doppler coupling in FMCW, and applications in altimetry, automotive and short-range sensing.
Optional Topics
- Tracking: monopulse angle estimation, track-while-scan, alpha-beta and Kalman filtering, data association.
- CFAR detection: cell-averaging, ordered-statistic and adaptive variants.
- Space-time adaptive processing for airborne and space-based radar.
- Weather radar, reflectivity and dual-polarisation moments.
- Automotive radar at 24 and 77 GHz; MIMO automotive architectures.
- Electronic warfare: jamming, chaff, ECM and ECCM, low-probability-of-intercept design.
- MIMO and distributed radar; passive and multistatic radar.
- Radar hardware: transmitters, T/R modules, receivers and digital beamforming.
Resources & Tools
- Introduction to Radar Systems (Merrill Skolnik) is the classic and most widely cited text and shares FIU’s course title. Principles of Modern Radar: Basic Principles (Richards, Scheer and Holm) is the modern standard and is the better match for a signal-processing-oriented treatment. Radar Systems Analysis and Design Using MATLAB (Mahafza) is widely used because it pairs the theory with runnable code, which suits this course particularly well.
- Fundamentals of Radar Signal Processing (Mark Richards) is the reference for the processing half, and Detection of Signals in Noise (McDonough and Whalen) for the detection theory.
- MATLAB is the dominant environment in radar — more decisively than in most areas of electrical engineering, because the defence industry standardised on it. The Radar Toolbox and Phased Array System Toolbox are the specific tools, and familiarity with them is directly marketable.
- Python with NumPy and SciPy is a workable free alternative for coursework.
- Electromagnetic simulation — Ansys HFSS, CST or FEKO for antenna and RCS work where the course goes that far.
- Software-defined radio — where a laboratory component exists, an SDR platform (USRP, ADALM-PLUTO, or an FMCW radar development kit from Texas Instruments) allows a genuine working radar to be built at low cost. Building an FMCW radar that detects a walking person is a memorable and achievable project.
- Reference material — the IEEE Aerospace and Electronic Systems Society, IEEE Transactions on Aerospace and Electronic Systems, the IEEE Radar Conference, and the Radar Handbook (Skolnik, ed.) as the field’s reference work.
Career Pathways
- Radar systems engineer and radar signal processing engineer (SOC 17-2071, Electrical Engineers) — the direct destination.
- RF and microwave engineer; antenna engineer — the array and aperture material leads directly into both.
- Electronic warfare engineer — a substantial and growing specialisation, and one where clearance-eligible graduates are in short supply.
- Remote sensing engineer — synthetic aperture radar for earth observation, at NASA, NOAA and commercial satellite operators.
- Automotive radar engineer — FMCW radar at 77 GHz is standard equipment on new vehicles, and this is one of the fastest-growing radar employment areas outside defence.
- Weather radar engineer — particularly relevant in Florida, where hurricane observation is a national priority.
- Air traffic and navigation systems engineer with the FAA and its contractors.
- ⚠⚠ Florida is one of the strongest radar employment markets in the United States, and this is not a general claim. L3Harris in Melbourne and Palm Bay is one of the largest radar and electronic warfare employers in the country and recruits heavily from Florida universities. Lockheed Martin (Orlando), Northrop Grumman (Melbourne and St. Augustine), Raytheon/RTX (Largo) and Leidos all run radar programmes in the state. NOAA’s Aircraft Operations Center at Lakeland flies the hurricane-hunter aircraft and their radar systems; the National Hurricane Center is in Miami. NSWC Panama City works on maritime sensing and recruits directly from the FAMU–FSU College of Engineering. The Space Coast adds range safety and tracking radar at Cape Canaveral and Patrick Space Force Base.
- ⚠ Security clearance is worth planning for. Most defence radar work requires a clearance, which requires US citizenship and takes months to process. Employers sponsor it, but eligibility should be considered when planning a career in this direction — and an internship with a cleared employer is the usual route to starting the process early.
Special Information
⚠⚠ Three prerequisites, and most students are weak in one of them
Radar sits at the intersection of three subjects, and this is the single most useful thing to understand before registering. The FAMU–FSU requirements make the structure visible:
| Requirement | Supplies |
|---|
| EEL 3473 (electromagnetics) | Propagation, antennas, apertures, polarisation |
| EEL 3135 (signals and systems) | Matched filtering, Fourier analysis, sampling, correlation |
| EEL 4021 (corequisite — random signals) | Noise statistics, detection theory, clutter models |
The statewide record lists EEL 3512 and EEL 3473; institutional requirements differ, so check your own catalog.
⚠ Almost every student arrives strong in two of the three and weak in the third, and the weak one determines where the course becomes difficult:
- Weak in electromagnetics — the antenna and array material (gain, aperture, array factor, grating lobes) will feel like a separate course. This is the most common gap.
- Weak in probability — detection theory will be the obstacle, and it arrives early. Probability of detection and false alarm, ROC curves and clutter statistics are unavoidable and are where the course is genuinely quantitative.
- Weak in signals — matched filtering, the ambiguity function and pulse compression will be hard, and these are the heart of modern radar.
Identify your weak side before the term starts and address it deliberately. The random-signals course being a co-requisite rather than a prerequisite at FAMU–FSU is a warning in itself: it means detection theory may be taught in this course before you have met the probability underlying it.
⚠ Course-code variation across Florida
Radar is offered narrowly at undergraduate level in Florida:
- EEE 4550 — FAMU (Radar) and FIU (Introduction to Radar Systems).
- EEE 5557 Introduction to Radar Systems — the graduate counterpart, carried at FAMU, FAU, FIU and UCF, which is a notably wider spread than the undergraduate number and reflects where the subject is usually taught.
- Related undergraduate material appears in RF and microwave courses and in EEE 4510 Digital Signal Processing, but no other institution carries a dedicated undergraduate radar course under a different EEE number.
⚠ Note the graduate/undergraduate asymmetry. Radar is more commonly a graduate subject in Florida, which means an undergraduate who takes it has covered material most of their peers meet later — a real advantage in recruitment, and worth saying explicitly in an interview. SCNS equivalency does not cross course numbers, so expect this to articulate as a technical elective elsewhere.
Position in the curriculum
EEE 4550 is a senior-level elective following electromagnetics and signals and systems. It pairs naturally with EEE 4510 Digital Signal Processing — students who have taken DSP first find the matched filtering, Doppler processing and pulse compression material substantially easier, because it is filtering theory in a specific application. It leads into graduate radar (EEE 5557) and into RF, antenna and remote sensing coursework.
Difficulty and time commitment
This is a demanding senior elective, and its characteristic difficulty is the number of decibels. That sounds glib but is accurate: radar analysis is conducted almost entirely in logarithmic units, link budgets chain a dozen terms together, and a student who is not completely fluent in decibel arithmetic will make errors that are hard to spot because the numbers remain plausible. Become fluent in dB, dBm, dBW and dBsm early — it is the highest-return preparation for this course.
The second difficulty is that the range equation is deceptively simple and its application is not: a real link budget involves system losses, atmospheric attenuation, integration gain, fluctuation loss and scan loss, each of which must be justified.
Plan on nine to eleven hours a week, more where a MATLAB simulation project is set.
Articulation and transfer
SCNS records EEE 4550 as guaranteed to transfer to an institution offering the same course. Two Florida institutions carry it, both at 3 credits. The course is upper-division and carries no general-education or Gordon Rule designation.
⚠ Export control and restricted material
Radar sits close to controlled technology. Students should be aware that some radar performance data, waveform designs and system parameters are subject to ITAR (International Traffic in Arms Regulations) or EAR (Export Administration Regulations). Coursework uses unclassified, published material and is unaffected, but three practical consequences follow: internships and projects with defence employers may require citizenship or clearance; some conference and journal material is restricted; and a student should not assume that specifications encountered during an internship can be discussed in a class presentation. When in doubt, ask the employer.
FE exam relevance
The NCEES Fundamentals of Engineering (Electrical and Computer) exam covers electromagnetics, signal processing and communications, all of which this course applies, but does not test radar specifically. The link-budget and noise-figure work is directly transferable to the communications portion of the exam.
AI Integration
Radar is a field where machine learning has been adopted rapidly and where the classical theory remains indispensable, which makes the boundary between them unusually clear.
Where AI is genuinely used in the discipline. Automatic target recognition from SAR imagery and range profiles is now dominated by learned classifiers. Radar-based human activity recognition from micro-Doppler signatures — distinguishing a walking person from a vehicle, or detecting a fall in an assisted-living setting — works well and is deployed. Interference mitigation in automotive radar, where many vehicles share a band, is increasingly learned rather than designed. And cognitive radar, in which the transmitted waveform adapts to the environment, is an active research programme. A graduate entering radar will meet these.
⚠ Where the classical theory is not replaceable. The radar range equation is physics: no amount of learning recovers a return that is below the noise floor. Detection theory gives a provable optimum under stated assumptions — the Neyman–Pearson test is optimal, and a learned detector cannot beat it when the assumptions hold, only when they do not. And in safety-critical and regulated applications, a detector whose false-alarm rate cannot be characterised analytically is difficult to certify. The classical material tells you what is achievable; the learned methods sometimes get closer to it in messy conditions.
Where a general-purpose assistant helps in coursework. Explaining the ambiguity function or why grating lobes appear at half-wavelength spacing; checking a link-budget calculation; generating MATLAB code for range–Doppler processing, pulse compression and array factor plots; and explaining the field’s dense acronym vocabulary.
⚠ Where it fails, and why the failure is this course’s own subject. The characteristic error of an AI tool asked a radar question is to produce a range-equation calculation with unstated or inconsistent units and no loss terms — mixing dB and linear quantities, omitting system losses, atmospheric attenuation and integration gain, and returning a detection range that is optimistic by an order of magnitude. Getting a link budget right is the core professional skill this course teaches, and the errors are invisible: the answer is a number in the right general vicinity, and nothing in it signals that four loss terms are missing.
A second failure matters as much: models routinely quote a radar cross section as though it were a property of an object — “the RCS of a fighter aircraft is 1 m²” — when RCS depends strongly on frequency, aspect angle and polarisation and varies by orders of magnitude as a target manoeuvres. The Swerling fluctuation models exist precisely because RCS is not a constant, and a single number with no aspect or frequency attached is the misconception the course exists to remove.
Third, and specific to processing: models will describe range and Doppler resolution without acknowledging the ambiguity function that couples them, presenting bandwidth and coherent processing interval as independent knobs when the waveform ties them together.
The engineer’s responsibility. A radar performance prediction is a claim about whether a system will detect something, and in defence and safety applications people rely on it. The discipline is the itemised link budget: every gain and every loss stated, with its source and its assumed conditions. The habit to form is to ask of any detection-range figure: at what RCS, at what aspect, at what probability of detection and false alarm, with what losses assumed? A number without those is not a prediction.
Academic integrity. FAMU and FIU each maintain academic integrity policies covering AI-generated work. Analysis and link-budget derivations are normally expected to be your own even where coding assistance is permitted, and generated simulation results are data fabrication. Ask before you rely on a tool, and disclose its use where the syllabus requires it.