GPS and GNSS Overview

Getting started in GNSS can be confusing. For starters, GPS is often used as a blanket term when intending to mention GNSS. While a full description of the topic is out of scope of this article, let’s take a look at some of the important concepts you’ll need to get started in GNSS. Note: This guide will only discuss GNSS as it relates to civilian use.

What is GNSS?

GNSS stands for Global Navigation Satellite System, a network of satellites that broadcast their orbital data with timestamps that allow receivers to calculate their time and position on the Earth (longitude, latitude, and height). GPS (Global Positioning System), is just one of the few GNSS that exist. The full list includes:

• GPS (Global Positioning System) – United States
• GLONASS – Russia
• BeiDou – China
• Galileo – European Union

Additionally, there are also Regional Navigation Satellite Systems (RNSS) that service specific regions of the Earth:

• QZSS (Quasi-Zenith Satellite System) – Japan
  • Intended to compliment GPS with additional satellites and augmentation service
• IRNSS (Indian Regional Navigation Satellite Sytem) or NavIC – India

There are more local augmentation systems such as at airports, but these are beyond the scope of this article. The rest of this article will focus primarily on GPS, but just know that the concepts and system parameters are often similar or the same between the different GNSS.

How does GPS work?

The numerous satellites that make up the GPS network orbit roughly 20km above the Earth. Each one of these satellites broadcasts its time and orbital data (ephemeris) continuously over radio (light) waves. GPS receivers receive these messages and calculate their position and time by using the data from four satellites.

How do the GPS satellites know their position?

Onboard each GPS satellite are atomic clocks that are synchronized to each other and reference stations on the ground. These reference ground stations on Earth receive the GPS signals from the satellites, do some processing to determine each satellite’s location and orbital data, then transmit this information back to the satellites.

How do the GPS receivers determine their position and time?

The GPS messages from the satellites include each satellite’s orbital data. Knowing the position of a satellite and the time it took for that satellite’s message to reach the receiver, the receiver can calculate the distance it’s away from that satellite by multiplying the speed of light (radio waves the message is carried over) by the time it took for the message to reach the receiver. With three satellites, the receiver can determine its position, normally output as longitude, latitude, and height above mean sea level.

However, this explanation assumes that the receiver clock is synchronized with the satellite clocks, thereby allowing it to identify the reception time. Such an implementation is not feasible. To find the reception time, the receiver uses the message from a fourth satellite to determine the time difference between the receiver clock and the GPS satellite clocks.

This is why we need a minimum of four signals from GPS satellites to determine our time and position. And while four may not seem like much, there are only about 30 satellites in orbit (minimum of 24) and not all of them will be above the horizon. Also consider that satellite signal strength and noise will affect the ability and accuracy of the receiver. Mountainous terrain, tall buildings, being indoors, all can affect the receivers accuracy.

More information on these calculations can be found here.

What frequencies do GPS satellites broadcast on?

The most applicable carrier frequencies are:

• L1 – 1575.42 MHz
• L2 – 1227.60 MHz
• L5 – 1176.45 MHz

The other GNSS broadcast on the same (or similar) frequencies but may be labelled differently. For a more comprehensive list, see ESA’s Navipedia.

What are the GPS codes?

To simplify, the GPS codes are the different encodings of information on the carrier frequencies. They are often preceded by the frequency they reside on:

• L1 C/A - Original code, stands for Course Acquisition.
• L1 C – Added to allow for easier operability between GPS & Galileo. Stands for Civilian.
• L2 C – Added so a second code can be used for local ionospheric correction. Note that L2C is not a “protected” band (it’s used for other applications), so it’s not suitable for safety critical applications. Stands for Civilian
• L5 – Name of the frequency band and the code. Essentially like L2C but can be used for safety critical applications.

What’s in a GPS message?

Each satellite broadcasts a message that includes its ephemeris (orbital position) data, timestamp, clock correction data, an almanac containing rough orbital data for all satellites in the constellation, its health status, and more.

The entire message takes about 12.5 minutes to fully transmit.

Almanac

The almanac contains the rough orbital data for all satellites in the constellation. It also includes an ionospheric model. While not precise enough to determine an accurate position, it gives the receiver knowledge as to which satellites will be within the horizon (if the prior knowledge of its position and time is known).

It’s particularly useful when attempting to obtain a fix or for mission planning (identifying satellite coverage along a route).

The almanac is available online for download, meaning that receivers don’t have to download the almanac from the GPS messages (which would normally take ~12.5 min).

Although a new almanac is uploaded to the satellites typically every day, almanacs may be valid for up to two weeks.

How does the receiver obtain a fix?

Without getting too detailed, the receiver checks every signal it receives against the known signatures of each satellite. If it determines which satellite it’s receiving data from, it begins downloading data from it and seeking other satellite signals. Once the receiver is tracking a minimum of four satellites and it determines its position and time, it’s said to have obtained a fix. The time it takes to achieve this is called the Time to First Fix (TTFF).

Cold Start – Typically 2 to 4 min TTFF

In a cold start, the receiver has no previously known position or time data of itself and no almanac data to rely on. It must begin searching for signals much like was described earlier – comparing the received signals to all known satellite signatures.

Warm Start – Typically < 45s TTFF

In a warm start, the receiver has some leftover almanac, position, and time data from previous observations (within 20s time, 100km position, and 25m/s velocity). Using the almanac, it will restrict its search for satellites to only those satellites that are currently above its horizon. Less satellite signatures to check means TTFF will be quicker than a cold start.

Hot Start – Typically < 20s TTFF

In a hot start, the receiver very recently had fixes on satellites and knows which ones to acquire. It has a current almanac, ephemeris (for each satellite), time, and position.

Improving TTFF

Especially from a cold start, obtaining the latest almanac and satellite ephemeris data can greatly improve TTFF. Services such as Assisted GNS (A-GNSS), which is used for smartphones, allow receivers to download almanac and ephemeris data over the internet, sometimes providing receiver position and time data, instead of from GPS satellite.

Errors

Satellite Geometry

Calculated as Dilution of Precision (DOP), when satellites are clustered together as viewed by the receiver, the errors compound such that the error in position is significant. See the example below for visualization. The red and blue lines indicate the true distance measurement, the gray lines the potential error, the green area the potential result.

Reference

Ionospheric Delay

As signals pass through the Earth’s atmosphere, they are slowed by the Ionosphere. This delay will affect the affect the distance calculation to a satellite, introducing error. Satellites closer to the horizon will have their signal affected much more than those overhead given they travel longer through the Ionosphere.

Since this delay is proportional to the frequency of the signal, dual frequency receivers can significantly reduce this error by comparing the difference between two signal bands. Since dual frequency receivers tend to be more expensive, GPS and SBAS satellites will broadcast Ionosphere data for correction. However, this correction data is based on the location of the reference ground stations, meaning that the less local the ground station, the less accurate the correction data will be.

Humidity

Humidity can cause delays similar to ionospheric delay. Since it’s more localized than ionospheric delay, it may be more difficult to compensate.

Multipath Effects

Signals can reflect off surrounding terrain, buildings, ground, etc… These reflections introduce delays that can affect the accuracy of the fix. As such, the more open the sky, the better.

Ephemeris

The ephemeris data for each satellite is transmitted every 30 seconds, but can be up to two hours old. It’s typically valid for up to 4 hours, but may change should orbital maneuvers be required.

Clock Errors

Even the highly accurate atomic clocks aboard the satellites are susceptible to noise and drift. Ground stations will calculate corrections and upload these to the satellites for inclusion in the GPS message.

Accuracy

The positional accuracy will highly depend on the scenario and environment. Terrain, satellite visibility, satellite count, etc… can all affect positional accuracy. Under ideal conditions, civilian systems can expect somewhere around 3-5 meters or better.

Circular Error Probability (CEP)

Defined as the radius of a circle starting from the true location within which 50% (unless otherwise specified) of all measurements fall. For example, a CEP of 3 meters means that 50% of the time, the position will be within a 3 meter radius of the true location.

Often used as a metric for accuracy for receivers.

Geometric Dilution of Precision (GDOP)

This is the term to the error caused by the clustering of satellites. The closer the position of the satellites in the sky, the worse the error in position due to the individual uncertainties of each signal determination. It’s often broken down into individual (or grouped) metrics e.g. Horizontal (HDOP), Vertical (VDOP), Position/3D (PDOP), Time (TDOP).

Generally, a DOP of less than 5 is appropriate for an accurate fix.

Using Two Signal Bands

So why do GPS satellites transmit signals across more than one signal band? Redundancy and resistance to jamming. Additionally, a dual frequency receiver can use the second frequency to eliminate ionospheric error. It’s important to note that the further separation in frequencies, the more pronounced the difference in ionospheric delay, the more accurately the receiver can eliminate ionospheric error.

Why use multiple GNSS systems?

Using more than one GNSS system gives receivers access to more available satellites. This is particularly important in terrain-challenged scenarios such as mountains or urban canyons (e.g. buildings). Not only does this improve reliability (chance of having at least 4 satellites available), but it can improve accuracy and reduce TTFF, allowing the receiver to select the strongest signals with the least noise and lowest configuration for dilution of precision.

For more data on expected improvement, see U-blox’s article on using four GNSS constellations.

Power Management

Acquisition vs. Tracking

Acquisition (obtaining a fix) will typically consume more power than tracking (once a fix has been obtained). Minimizing power consumption depends on the application but will involve balancing the increased power consumption from switching off for longer periods of time and re-acquiring a fix vs. shorter off-times and longer tracking times.

Update frequency

Naturally, the more frequently navigational solutions are demanded, the more resources required. Care should be taken to balance power consumption with the need for timely position and time data.

Multi-Constellation Tracking

While tracking multiple GNSS constellations can consume more power, the increase may only be a fraction of what’s required for tracking a single GNSS constellation. Always check the receiver datasheet for power consumption data.

Satellite Based Augmentation Systems (SBAS)

SBAS are the generic name for a number of augmentation systems used on a regional and continental level. The main ones and their coverage include:

• WAAS – United States
• EGNOS – Europe
• ANGA – Africa
• GAGAN – India
• SDCM – Russia
• BDSBAS – China
• KASS – Korea
• QZSS – Japan
• SouthPAN – Australia

These systems provide increased accuracy to GNSS for the region they support.

How do SBAS work?

Generally, all SBAS operate under the same principle. The SBAS uses ground-based reference stations with known locations (using surveying and other high precision methods) to measure errors from GNSS signals (most often due to ionospheric delays). This data is processed at control centers and is then transmitted to typically geosynchronous satellites. These SBAS satellites (separate from the particular GNSS) then broadcast the correction data for receivers to use.

It’s important to note that SBAS correction data is intended to be used on a regional basis, since the ground stations making the corrections are in specific regional locations.

Output Format

Once the receiver has a fix, how does it output this data? This will depend on the receiver, but it’s typically extracted from one of a few message protocols that the receiver communicates to the host (MCU). STRDC products use NMEA and UBX.

NMEA

Published by the National Marine Electronics Association (NMEA), NMEA 0183 is a closed standard, messaging format for displaying GNSS data. Though NMEA 0183 is outdated by the newer NMEA 2000 standard, it’s still used in commercial applications.

Without delving too much into the details, NMEA messages include:

• A Talker ID that identifies which GNSS the message data concerns
• A message ID that indicates the contents of the message
• The message data that includes any number of parameters (e.g. Latitude, Longitude, Height above mean sea level, course, etc…)

Messages in NMEA are ASCII-coded which can require extra resources to parse.

UBX

A U-blox proprietary standard, operates similarly to NMEA and comes standard in all U-blox GNSS receivers. UBX messages are binary and are thus easier to parse than NMEA messages. However, the NMEA protocol is more widely used.

Which GNSS message protocol to use?

Deciding which GNSS message protocol to use will depend on the application and required data. Many modules that rely on GNSS input expect NMEA messages. An application may require data that is included in a single UBX message where it would require two separate NMEA messages. U-blox receivers often can output both message protocols if desired.

For more information on both NMEA and UBX protocols, U-blox provides detailed information in many of their GNSS receiver integration manuals.

See detailed implementations of both protocols in the strdc-sdk module GNSS-UBLOX.

Conclusion

Knowing the basics, understanding GNSS receiver parameters becomes a little easier. Ready to get started? Checkout our GNSS receivers. Contact our technical support for any questions related to our products.

Additional Reading

Penn State’s Department of Geography provides an excellent in-depth look into GPS and GNSS.