Hata Propagation Model

A closed-form empirical model that estimates median land-mobile path loss from frequency, distance, antenna heights, and environment-specific corrections inside a declared validity envelope.

Version

v1 · 2026-08-30 · History

Domain-specific #

1984

Origin domain

radio propagation

Subdomain

land mobile path loss

Aliases

Okumura–Hata model

Core Idea

The Hata propagation model is a closed-form empirical estimator of median transmission path loss for VHF/UHF land-mobile links. Masaharu Hata fitted formulas to the measurement-based Okumura method so that planners could compute urban loss from carrier frequency, transmitter–receiver distance, effective base-station antenna height, mobile antenna height, and an environment correction without reading families of curves.^[1][2] Suburban and open-area forms subtract calibrated corrections from the urban baseline.

The identity includes its empirical envelope. Hata's original paper states 100–1500 MHz, 1–20 km, base height 30–200 m, mobile height 1–10 m, quasi-smooth terrain, and isotropic-antenna propagation loss; many engineering summaries use a conservative lower frequency of 150 MHz.^[1] Evaluation outside those ranges is extrapolation, not automatically an application of the validated model.

Structural Signature

Recognition roles:

• land-mobile outdoor link — a base or fixed transmitter and a lower mobile receiver;
• median path-loss target — large-scale loss in decibels, not instantaneous fading;
• carrier frequency — entered in megahertz;
• link distance — entered in kilometres;
• effective antenna heights — base and mobile heights entered in metres;
• urban reference equation — the calibrated logarithmic baseline;
• mobile-height correction — a frequency- and city-size-dependent term;
• environment correction — urban, suburban, or open-area selection; and
• validity envelope — declared limits on frequency, range, heights, and terrain.

Recognition requires the whole parameterized rule, correct units, an environment class, and an in-envelope use. A generic statement that radio power decays with distance is Propagation, not Hata. A ray tracer or site-specific terrain solver is another model family.

What It Is Not

The model is not a first-principles solution of Maxwell's equations and does not trace individual reflected or diffracted rays. It is not free-space path loss, because its coefficients summarize cluttered land-mobile measurements. It is not the original Okumura method's graphical workflow, although it was derived from those data. It is not COST-231–Hata, which is a later higher-frequency extension with changed constants and an added metropolitan correction.^[3]

It predicts median large-scale path loss, not fast fading, delay spread, instantaneous received power, outage probability, or channel capacity. It also does not encode a digital cellular standard. Calling any empirical log-distance fit “Hata” erases the named coefficients, correction terms, units, and validity ranges that define this abstraction.

Scope of Application

Hata was designed for land-mobile system planning where a base antenna stands substantially above a vehicular or portable terminal. Engineers use the model for early coverage contours, link-budget comparison, candidate-site screening, and sensitivity studies across distance, height, frequency, and broad clutter class. Its computational simplicity made the Okumura measurement program usable in large planning calculations.^[1]

The urban equation is the reference. A small/medium-city mobile-height correction differs from the large-city piecewise correction. Suburban and open forms modify urban loss. Uses should state which branch was chosen and why the terrain and morphology resemble its calibration class. Modern standards and engineering reports still document Okumura–Hata as a macrocell planning model, but contemporary dense urban, indoor, irregular-terrain, or high-frequency deployments may require other validated models.^[4]

Clarity

For a small or medium city, the urban median-loss equation is

\[ L_U=69.55+26.16\log_{10}f-13.82\log_{10}h_b-a(h_m) +(44.9-6.55\log_{10}h_b)\log_{10}d, \]

with

\[ a(h_m)=(1.1\log_{10}f-0.7)h_m-(1.56\log_{10}f-0.8). \]

Here \(L_U\) is in dB, \(f\) in MHz, \(d\) in km, and \(h_b,h_m\) in metres.^[1] Because logarithms act on numerical values expressed in those prescribed units, silently supplying hertz or metres for distance produces nonsense while looking algebraically legal.

“Median” is equally important. A prediction of 147 dB does not assert that every location or instant has exactly that loss. Shadowing, buildings, terrain, foliage, and small-scale multipath produce dispersion around the large-scale estimate.

Manages Complexity

The model compresses a difficult environment-dependent propagation field into a few interpretable coordinates. Frequency, distance, two antenna heights, and a morphology branch replace exhaustive drive-test curves at the preliminary-planning stage. The distance coefficient itself changes with base height, so the model retains an interaction that a single universal path-loss exponent would discard.

This compression is useful because it supports rapid comparison: raising the base antenna changes both the intercept and distance slope; changing frequency alters baseline loss and mobile-height correction; changing environment selects another calibrated offset. What is discarded—street orientation, individual building geometry, terrain profiles, foliage, weather, and stochastic fading—must reappear as uncertainty or later-stage measurement rather than being treated as zero.

Abstract Reasoning

Within the envelope, increasing distance increases predicted loss because the distance coefficient remains positive for allowed base heights. Raising the base antenna generally lowers the predicted loss, though the full change includes its effect on distance slope. Increasing frequency generally raises the urban baseline in this range. Applying the suburban or open correction lowers loss relative to the urban reference for the same numeric inputs.

The model also licenses diagnostics. An unexpectedly large residual across many sites may signal a mismatched morphology class, incorrect effective height, out-of-range use, or a systematic local clutter effect. A consistent offset can motivate local calibration, but changing coefficients creates a calibrated derivative model rather than retroactively changing Hata's identity. Predictions from two models should not be compared until units, environments, and validity domains match.

Knowledge Transfer

The exact model transfers among qualifying land-mobile planning cases because the same variables, coefficients, branches, and median-loss output recur. It transfers from urban to suburban or open settings through defined correction formulas, not by metaphor. COST-231 and other derivatives demonstrate instrument lineage, but their altered ranges and coefficients keep them distinct models.

The portable skeleton—signal source, medium, attenuation, distance, and reach—belongs to Propagation. Empirical calibration and prediction travel more broadly as modeling practices, but a traffic forecast or epidemiological spread curve does not instantiate Hata. The named model remains bound to radio units, antenna geometry, land-mobile morphology, and measurement-derived constants.

Examples

Worked urban case. Take \(f=900\) MHz, \(h_b=50\) m, \(h_m=1.5\) m, and \(d=5\) km in a small/medium city. The mobile correction is

\[ a(h_m)\approx0.0159\ \mathrm{dB}. \]

Substitution into the urban equation gives \(L_U\approx146.94\) dB. Every input lies within the original envelope. Recomputing with base-10 logarithms confirms the value; using natural logarithms would be a formula error.

Environment comparison. The suburban correction

\[ L_S=L_U-2[\log_{10}(f/28)]^2-5.4 \]

gives approximately \(137.00\) dB for the same numeric inputs. The roughly 9.94 dB change is a model-class correction, not evidence that moving a real receiver across an administrative boundary instantly changes its channel.

Boundary non-example. Applying the base formula at 3.5 GHz and 100 m separation violates both frequency and distance ranges. A numerical result can still be produced, but it is unsupported extrapolation. A suitable higher-frequency or short-range model must be selected and validated.

Structural Tensions

• Computational simplicity versus environmental detail. Few parameters enable fast planning but suppress site geometry. Diagnostic: compare residuals against drive tests before using the model for final design.
• Standard coefficients versus local calibration. Fixed coefficients make studies comparable, while local tuning can improve fit. Diagnostic: label tuned equations as derivatives and retain an untuned baseline.
• Broad morphology classes versus continuous clutter. Urban, suburban, and open branches are usable but coarse. Diagnostic: inspect whether the survey environment actually matches the chosen calibration class.
• Interpolation versus extrapolation. In-range evaluation inherits empirical support; out-of-range arithmetic does not. Diagnostic: report every input beside the original envelope before accepting an estimate.
• Autonomy versus reduction. Hata instantiates Propagation yet adds a stable calibrated estimator. Diagnostic: if the named coefficients, corrections, units, and envelope can be removed without changing the claimed method, the case is generic propagation rather than Hata.

Structural–Framed Character

The model combines structural and empirical framing. Its logarithmic equation and variable roles are structural. Its constants, height definitions, morphology labels, and validity ranges are framed by the Okumura measurement program and Hata's fit. Those empirical choices are not optional decoration: they are precisely why the formula predicts one class of radio environments rather than every propagation medium.

Structural Core vs. Domain Accent

The structural core is calibrated attenuation prediction from distance, scale, source/receiver configuration, and environment. The domain accent is decisive: MHz, km, antenna heights, dB loss, urban morphology, and the Okumura-derived coefficients. Removing that accent yields generic propagation or empirical prediction, not the Hata model.

The candidate therefore passes the domain-specific bar but not the prime bar. Its roles recur across telecom planning practices, not across unrelated substrates with unchanged operative vocabulary.

Instantiates / Related Primes

The minimal parent is Propagation: the model estimates how a radio signal attenuates while spreading from a source through a cluttered environment. Measurement is historically important because the constants derive from field data, but the model is a predictor rather than an instrument–target measurement operation. Approximation describes its empirical simplification but does not capture the transmitted signal and medium. Channel Capacity is declined because path loss may feed a link budget, while capacity additionally depends on bandwidth, noise, interference, and coding assumptions.

Relationships to Other Abstractions

Current abstraction Hata Propagation Model Domain-specific

Parents (1) — more general patterns this builds on

• Hata Propagation Model presupposes Propagation Prime

  The minimal parent is Propagation: the model estimates how a radio signal attenuates while spreading from a source through a cluttered environment.

Hierarchy path (1) — routes to 1 parentless root

• Hata Propagation Model → Propagation

Neighborhood in Abstraction Space

Hata Propagation Model sits in a sparse region of the domain-specific corpus (90^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

• Rugosity — 0.80
• Wireless triangulation — 0.79
• Ziggurat Algorithm — 0.79
• Variogram — 0.79
• Floor Effect — 0.78

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With

• Okumura model: measurement-curve method from which Hata derived closed-form fits.
• COST-231–Hata: later extension for higher bands with different constants and correction.
• Free-space path loss: ideal unobstructed spreading without Hata's urban calibration.
• Log-distance model: a broader family with fitted reference loss and exponent.
• Ray tracing: site-specific deterministic geometry rather than broad empirical morphology.
• Fast-fading model: predicts short-scale variation, not median large-scale loss.
• Link budget: combines path loss with powers, gains, feeder losses, margins, and receiver sensitivity.
• Channel capacity: information-rate limit, not a propagation-loss estimator.

The discriminating test is whether the calculation uses Hata's prescribed coefficients, corrections, units, and validated parameter envelope to estimate median land-mobile loss.

References

[1] Masaharu Hata, “Empirical Formula for Propagation Loss in Land Mobile Radio Services,” IEEE Transactions on Vehicular Technology 29, no. 3 (1980): 317–325, https://doi.org/10.1109/T-VT.1980.23859. registry ↩a ↩b ↩c ↩d

[2] Yoshihisa Okumura, Eiji Ohmori, Tomihiko Kawano, and Kaneharu Fukuda, “Field Strength and Its Variability in VHF and UHF Land-Mobile Radio Service,” Review of the Electrical Communication Laboratory 16, nos. 9–10 (1968): 825–873. registry ↩

[3] COST Action 231, Digital Mobile Radio Towards Future Generation Systems: Final Report, EUR 18957 EN (European Commission, 1999), ISBN 92-828-5416-7, https://op.europa.eu/en/publication-detail/-/publication/f2f42003-4028-4496-af95-beaa38fd475f. registry ↩

[4] NIST, User's Manual for the NIST Wireless Infrastructure Deployment Tool, NISTIR 7761 Revision 1 (2014), section 5.2.1.3.1, https://doi.org/10.6028/NIST.IR.7761r1. registry ↩