Time to first fix (TTFF): what it is and how to make GNSS fixes faster

A device powers on, the global navigation satellite system (GNSS) receiver starts listening, and the user waits for a valid position. The length of that wait is the time to first fix (TTFF).

It is one of the most visible measures of how well a positioning device works, shaped by everything from the freshness of the receiver’s orbital data to the efficiency of the antenna that hears the satellites.

Time to first fix is the time a GNSS receiver takes, from power-on or reset, to compute its first valid position, velocity and time (PVT) solution. A fix is that solution, a position the receiver trusts enough to report, worked out from the satellite signals it has acquired and decoded. Until that happens, the device is powered on but not yet usable.

Three conditions have to hold at once: the receiver needs signals from at least four satellites, enough to solve for position in three dimensions plus its own clock error; it needs current orbital data for each of them to know where they are; and those signals have to be clean enough to decode, which is where the environment and the antenna come in.

GNSS covers GPS, Galileo, GLONASS and BeiDou, and a receiver that can hear more of them has more satellites to work with.

How quickly those conditions are met depends heavily on what the receiver still remembers from its last session.

The single biggest driver of TTFF is how much valid data the receiver still holds when it wakes, which sorts start-ups into three named cases.

The difference between these modes comes down to two kinds of orbital data. The cold, warm and hot framework is standard GNSS test terminology, defined by receiver and test authorities such as Safran Navigation and Timing. The almanac is a coarse, long-life description of the whole constellation, good for weeks, that tells the receiver roughly where to look. The ephemeris is the precise, short-life orbital data for one satellite, valid for a few hours, and the receiver needs a current ephemeris for every satellite it uses to compute an accurate position.

On the legacy GPS signal each satellite broadcasts its navigation frame over 30 seconds and the full almanac takes 12.5 minutes to receive, so a receiver forced to decode fresh orbital data from the sky pays a penalty measured in tens of seconds or more (ESA Navipedia, GPS navigation message).

A cold start carries it in full, while a hot start avoids it because the data is already sitting in memory.

What counts as a good TTFF depends on the start mode and the environment, so it comes as a range rather than one fixed number. Under open sky, a modern multi-constellation receiver can hot-start in a few seconds and cold-start in well under a minute, and assisted GNSS can cut that cold start to a few seconds.

The same device in an urban canyon, under foliage, or indoors can take several times longer, or fail to fix at all. TTFF is judged against the conditions the device actually ships into, not a bench in open sky.

TTFF is never down to one component. It emerges from the satellite data on hand, the receiver’s state and firmware, the RF signal quality and the product’s physical design, and any one of them can be the limiting factor.

A stable PVT fix needs enough satellites in view, and spread across the sky rather than bunched in one patch. Open sky offers both, with more satellites and cleaner signal paths, while urban canyons, foliage, metal enclosures, a vehicle interior, even the user’s own body, block or attenuate the signal and thin out what the receiver has to work with.

Signal quality compounds the problem. The satellites sit more than 20,000 km up, so what reaches the antenna is already faint, and obstruction, multipath and atmospheric attenuation pull the signal-to-noise ratio (SNR) lower still. Reflections in an urban canyon are the hardest case, because the same signal arrives by several paths at once; indoors or under dense foliage the receiver may see satellites yet struggle to pull clean enough data for a stable first fix.

The satellite data a receiver keeps in memory goes stale over time. Current almanac and ephemeris let it acquire quickly, while missing or aged data forces it to rebuild that information from the satellites before it can compute a fix.

A device that wakes often therefore tends to fix faster than one that powers all the way down for long stretches, because keeping some time and satellite data alive cuts the work left for the next start.

Receiver sensitivity governs how well acquisition holds up in weak signal. Firmware does just as much of the work, through its search patterns, its handling of assisted data, how it selects constellations, its power-saving modes and the way it treats cached ephemeris.

The same receiver tuned for low power will not behave like one tuned for fastest acquisition, so power budget and TTFF usually end up as a deliberate trade-off.

Antenna efficiency and placement decide how much of that faint signal actually reaches the receiver, the largest lever a design team controls.

GNSS is a receive-only, weak-signal task. The signal arriving from a satellite is extraordinarily faint, well below the noise floor by the time it reaches the ground, and every decibel the antenna fails to capture is a decibel the receiver has to claw back before it can acquire and hold a lock.

Antenna efficiency, the share of available signal power the antenna actually delivers to the receiver, therefore sits directly on the TTFF path, and its effect grows exactly where fixes are hardest, in cities and under cover.

Because satellites sit at arbitrary elevations and a device can be held or mounted in any orientation, a GNSS antenna should radiate omnidirectionally rather than lean on directional gain aimed at one part of the sky. A common design target is at least 70% radiation efficiency, omnidirectional, across the GNSS band.

The signal environment sets what that target has to overcome. These illustrative ranges from Ignion’s GNSS design work show how far it can move TTFF:

Typical, illustrative ranges, not a measured result for any one build.

The efficiency figures behind that target are measurable. On real reference boards, Ignion’s RUN mXTEND™ multiband antenna booster (part NN02-224) was characterized for L1-band GNSS in Ignion’s Satimo STARGATE 32 anechoic chamber.

126.5 × 60 mm board

86 × 54 mm board (same component, retuned only through its matching network)

RUN mXTEND™ booster (part NN02-224), measured by Ignion in its Satimo STARGATE 32 anechoic chamber. Figures belong to these specific boards; a different layout, clearance and enclosure will shift them.

That is precisely the kind of shift integration teams validate on their own board before committing to a prototype.

The booster is one part of Virtual Antenna® technology, a non-resonant component that uses the PCB ground plane to radiate. Oxion™ estimates GNSS efficiency and checks the matching network on the design’s own board ahead of the prototype build, working from measured behavior rather than a simulation.

Design resource: the GNSS antenna integration design guide. Placement, matching and isolation for GNSS.
→ Download the guide

One more antenna-level factor decides whether that efficiency survives contact with the rest of the device. GNSS almost always shares a board with a cellular or low-power wide-area radio, and that transmitter is many orders of magnitude stronger than the faint satellite signal the receiver is trying to hear.

Isolating the GNSS antenna from the cellular one, with at least 5 mm of separation and a dedicated matching network, protects the weak GNSS downlink from being desensitized by its noisy neighbor. Getting that isolation and the matching right is a large part of solving GNSS integration challenges in a compact multi-radio device, and it is where the choice between a chip and a patch antenna for GNSS is really made.

By the time a device is in the field, most of its TTFF is already locked in by decisions made on the bench. Five of them move the needle most:

• Turn on assisted GNSS. Assisted GNSS (A-GNSS, also called A-GPS) delivers predicted ephemeris plus a rough time and location to the receiver over the network, so it skips the slow on-air ephemeris download. This is the single largest lever, and it can turn a cold start into something close to a hot one.
• Receive multiple constellations and bands. Tracking GPS, Galileo, GLONASS and BeiDou together puts far more satellites in view, which improves geometry and shortens acquisition. Adding a second band, L1 plus L5, further improves accuracy and resistance to multipath, and it is the direction receivers are heading.
• Give the antenna a clear view of the sky. Place it away from the battery, the display, large metal and the user’s hand, and away from the cellular antenna. Sensible placement recovers signal that no amount of receiver tuning can.
• Specify a high-efficiency, omnidirectional GNSS antenna with clean matching. Efficiency lost here is acquisition speed you cannot recover downstream, so a high-efficiency GNSS antenna booster such as RUN mXTEND™ with a properly tuned matching network pays back directly in TTFF.
• Keep the receiver warm or hot where the power budget allows. Holding recent ephemeris, time and position in memory, or backfilling them with A-GNSS on wake, avoids paying the full cold-start cost every time the device powers up. This is how a real GNSS tracker in the field keeps fixes quick without draining its battery.

There is no single setting that delivers a faster TTFF. It comes out of the whole system, when fresh satellite data, enough visible satellites, clean RF design, sound receiver configuration and an efficient antenna all pull in the same direction.

For compact trackers, wearables and multi-radio IoT devices, the antenna decision should be made before the board layout is locked, while placement, matching and isolation can still change cheaply.

See what a Virtual Antenna® booster does for GNSS efficiency on your own board. Sign up to Oxion™ 

TTFF means time to first fix. It is the time a GNSS receiver needs after power-on, reset or signal loss to calculate its first valid position, velocity and time solution. It includes satellite acquisition, navigation-data handling and the first usable position output.

A good TTFF depends on start mode and environment. Hot starts should usually be measured in seconds. Warm starts are commonly under a minute. Cold starts can take minutes because the receiver may need to rebuild satellite data from the broadcast navigation message.

Slow GPS TTFF usually comes from stale ephemeris, poor sky view, weak SNR, multipath, low antenna efficiency, receiver power-saving behavior or poor GNSS antenna placement. In compact multi-radio devices, cellular noise and weak GNSS isolation can also lengthen acquisition time.

A cold start begins with little or no reliable time, position, almanac or ephemeris data, so the receiver must search broadly and collect more data. A hot start begins with valid time, position, almanac and ephemeris, so the receiver can usually calculate a fix in seconds.

The antenna affects TTFF because it controls how much usable GNSS signal reaches the receiver. Low efficiency, poor placement, detuning or weak isolation reduces acquisition margin. In obstructed environments, that can turn a seconds-level fix into a much longer wait.

Assisted GNSS provides satellite orbit, time or coarse location data through a network connection instead of relying only on the satellite broadcast. That can reduce TTFF because the receiver starts with data it would otherwise need to collect over a slow GNSS link.