Copyright © 2006 Penton Media, Inc., All rights reserved.
Printing of this document is for personal use only.
Reprints

Wherever You Go, There You Are. That just happens to be the title of Jon Kabat-Zinn's 1994 book on Buddhist meditation. However, you could also apply that description to the U.S. Air Force's Global Positioning System (GPS). This technology knows exactly where you are, even if you don't.

Also referred to as Navstar, GPS has been in operation since the early 1990s. It proved invaluable initially in the Gulf War and in all military endeavors since then many times over. Not only that, GPS has become an essential part of many commercial and personal products that rely on location, position, and navigation.

Every year, GPS goes through constant updates, making it more accurate than ever before. And many low-cost GPS radio chips and navigation receivers have become available to everyone. In fact, one standout trend is to add GPS to as many handheld products as possible.

GPS IN ACTION
The GPS system comprises a constellation of 24 operational satellites, plus at least three spares, that orbit the earth at 12,548 miles or 10,898 nautical miles (20,200 km) up with an inclination of 55° to the equator. There are six orbits with four satellites each. The rotational period is just two minutes short of 12 hours per orbit.

Each satellite carries four atomic clocks (two cesium-based and two rubidium-based) that generate dead-on accurate timing pulses. These are used as the basis for generating the signals sent to receivers on earth. Each satellite contains its own unique pseudorandom code (PRC) for differentiating itself from its neighbors.

Also, each satellite transmits what is called ephemeris information, which defines precisely where it is in orbit. Such information is translated into a ground track on earth that will identify its latitude and longitude, providing the requested location. The earth station updates the ephemeris data daily.

Each satellite transmits its PRC and ephemeris data in the microwave L band at 1575.42 MHz. This is called the L1 signal. The receiver can recognize each individual satellite by its unique PRC, just as in other direct-sequence spread-spectrum systems. The PRC is transmitted at a 1-Mbit/s rate using binary phase-shift keying (BPSK). Repeating every 1023 bits, this is called the coarse-acquisition (C/A) code.

Figure 1 shows how this code is used as the chipping code for the navigation data, which occurs at a 50-bit/s rate—yes, 50 bits per second! The navigation code contains the ephemeris data. The overall L1 signal occupies a bandwidth of about 1 MHz.

Each satellite additionally transmits an L2 signal at 1227.6 MHz. The L2 uses another 1023-bit PRC called the P-code. It occurs at a 10.23-Mbit/s rate and is used to chip the 50-bit/s navigation data. The P-code also may be encrypted, in which case it's called the Y-code. The resulting signal then modulates the 1227.6-MHz carrier and the L1 signal as well. The L2 signal is strictly for military use.

Back on Earth, a receiver picks up the signals, does a tricky triangulation calculation, and spits out time, altitude, and position data. The position information is in the form of latitude and longitude. Since time is available, it also can figure velocity. Receiver manufacturers call it PVT, or position-velocity-time output. Using fancy software and map overlays, you can generate a display that shows where you are on a detailed map, much like that used in those 1960s James Bond movies.

The most important issue in getting a GPS fix is being able to "see" the satellites. Given that they're over 12,000 miles away, you need all the signal you can get plus a good antenna and a super-sensitive radio. The only real way to get a signal is to have the antenna in clear view of the satellites. If you go inside, you'll lose the signal. That's why GPS radios only work outside or in a vehicle with a window.

Once you have a good view of the satellites, the receiver takes several minutes to lock on to one of the satellites. It then extracts the data and is passed off to another satellite in view, and again the data is taken. Next, it locks onto a third satellite, and so on. Latitude and longitude data requires three satellites. Altitude and speed calculations require four satellites.

The receiver measures the signal's time of travel from the satellite to the receiver. Knowing the speed of light or radio waves in space (slightly less than 300 million meters/s) and the precise time, it can calculate the distance to the satellite. That distance value is used in the calculations along with the other data from the satellite.

The receiver itself is the usual superhet or direct-conversion type with DSP demodulation and baseband recovery in an on-chip or external CPU. The processor, usually quite powerful, typically is a 32-bit CPU with floating point so it gets the required accuracy.

ENHANCED GPS
While GPS is pretty accurate, it doesn't quite meet the grade for many applications. The military can get precision of less than 1 meter, but such results aren't available to commercial applications or individuals. Current precision from a commercial receiver is less than about 10 m.

Several augmented versions of GPS have been developed for commercial applications that need greater precision.

The USCG builds fixed basestations along the U.S. coast and major waterways where fixed positions are precisely known. These basestations monitor the satellites and compare this information with the station's own precisely known location. The system quantifies any difference or error and then transmits that error by a separate transmitter operating in the 285- to 325-kHz range to nearby DGPS radios.

The DGPS radios feature the GPS receiver as well as a receiver for the DGPS signal. The error data is then used in the processor to compute a more accurate location. Resulting accuracy is better than 5 m. DGPS receivers are widely available, but they're only effective near coastal areas where the signals are available. Boats and ships really benefit from the more precise navigation. The railroad industry has a similar system.

The Federal Aviation Administration (FAA) developed WAAS to permit blind instrument landings of aircraft. Even the DGPS level of precision is a bit too loose to determine the exact location of runways and their physical bounds.

The FAA built about 25 ground stations around the U.S., some on the coast, with precisely known positions. These stations monitor the GPS satellites and determine errors. The errors are then broadcast to two geosynchronous satellites that rebroadcast the error signals to special WAAS GPS radios in the planes. With this system, location can be predicted to within about 3 feet.

Assisted GPS works with cell phones. Many cell phones, specifically those CDMA handsets used with Verizon and Sprint Nextel systems, contain a GPS receiver that helps to meet the government's mandate for E911 capability on all cell phones. Therefore, a cell phone's position can be accurately determined if a 911 call is made from it, enabling emergency personnel to quickly respond and locate the caller even if no directions are provided.

Incidentally, phones that work with AT&T (previously Cingular) and T-Mobile don't use GPS. They employ another location technique based on the position of three cell towers that can receive the phone's signals and triangulate the locations. In any case, A-GPS only works on GPS-enabled phones if it's incorporated.

A-GPS brings the cell-phone network into the picture to provide location information when the cell phone goes inside, at which point the antenna cannot see the satellites. In the network, a location server monitors the satellites and knows the precise location. That server communicates the information to the cell phones via the network.

One frustration with GPS is the long acquisition time required for a receiver to lock on to the satellite signal and receive the position data. The "time to first fix" (TTFF) is many minutes from a cold receiver start. That's because the navigation data is transmitted at 50 bits/s.

Emergencies require a faster lock time. One way to get a faster lock is to have the cell-phone GPS receiver get its current location using position assistance from the location server. This lets the GPS radio know where to start, reducing the start time for an initial calculation from minutes to seconds. Then, even if the cell phone loses contact with the satellites, it still has some location data from the server, making it possible to still determine location.

The GPS receiver in the cell phone has the last known position. Communicating back and forth with the location server, the phone can figure out pretty closely where it is and give that information back to the network.

Most A-GPS is contained in CDMA phones and implemented by Qualcomm's gpsOne chip sets. A-GPS also can be used with GSM/WCDMA phones. However, that isn't happening at the present.

There's some movement to eventually include GPS in AT&T and T-Mobile cell phones. A-GPS is a TIA/EIA and ANSI standard. Other A-GPS standards include Radio Resource Control (RRC), Radio Resource Location Protocol (RRLP), and Secure User Plane Location (SUPL).

GLOBAL NAVIGATION SATELLITE SYSTEMS (GNSS)
While GPS is the most widely used satellite navigation system, it's not alone. Russia's Global Navigation Satellite System (GLONASS) has been around for nearly as long as GPS. Comprising 12 satellites, it performs essentially the same functions as GPS. GLONASS is used primarily in Russia, Northern Europe, and Canada.

Currently under construction, the European Union's Galileo satellite navigation system resembles and even operates on the same L1 frequency. It will use 29 satellites with orbits that fill in the gaps where GPS satellites aren't positioned. A receiver with both GPS and Galileo receive capability would get better coverage and more precise positioning information.

Only one Galileo satellite is in orbit, but a program in place now will look to launch more over the next few years. Full operational status is scheduled for 2010 to 2012.

DESIGNING IN GPS
So far, the industry has focused on personal navigation devices (PNDs) and automotive accessories using GPS. Handheld GPS radios are widely available, and virtually every car and truck now offers a GPS navigation system as an option. Today's trend is to add GPS to other devices, especially portable and handheld devices like cell phones, PDAs, and even digital cameras to location-stamp the pictures.

The small size and lower power consumption of the newer devices makes this possible. Any new design requires the addition of a GPS receiver chip, a processor, and, of course, the all-important antenna. Nearly a dozen receiver chip choices hail from companies like Atmel, Global Locate, Glonav, Infineon, Maxim, SiGe Semiconductor, SiRF, STMicroelectronics, and Texas Instruments.

Positioned to exploit the handheld GPS trend, Nemerix SA of Switzerland was founded to address the low-power and small-size needs of portable and mobile devices. Its NX3 receiver claims one of the best, if not the best, power-consumption figures available today. Requiring only about 35 mW, it achieves one of the industry's top receiver sensitivities at 158 dBm.

The NX3 comes in a 6- by 6-mm package and uses stacked die inside the silicon-germanium (SiGe) receiver front end along with a low-noise amplifier (LNA) and a CMOS baseband processor. The output is a serial port to the host processor, usually an ARM7 or ARM9 with the 5 to 10 MIPS needed to run the GPS software. The NX3 is fully A-GPS capable as well.

Nemerix also recently announced its NeX extended ephemeris solution, which is a software component that works with A-GPS to provide regular predictive ephemeris updates every seven days to the receiver. This reduces the need for the receiver to download the full ephemeris data from the satellites, which normally takes minutes. In fact, NeX shrinks TTFF to about 5 to 15 seconds.

One of the smallest of the new batch of GPS receiver chips is SiGe Semiconductor's SE4120S (Fig. 2). With dimensions of 4 by 4 mm, it's ideal for embedding in cell phones and other portable devices. As a bonus, it's a fully operational Galileo receiver, too.

Its on-chip local-area network (LAN) has a gain of 18 dB with a noise figure of 1.6 dB, giving the receiver exceptional sensitivity. It can acquire satellite signals as low as 160 dBm and track signals as weak as 170 dBm. An adjustable on-chip IF filter optimizes the performance to GPS or Galileo. A GPS-only version called the SE4110S comes in a 2.2- by 2.2- by 0.3-mm chip-scale package.

Another Galileo-ready receiver is the u-blox 5 from Swiss-based u-blox. It features 160-dBm sensitivity and is A-GPS ready. The u-blox 5, which comes in several versions, boasts a 1-second TTFF. Consumption runs less than 50 mW.

RF Micro Devices, a longtime player in GPS, acquired IBM's GPS business about five years ago. Dave Lyon of RFMD indicates that the company's latest contribution is the RF8110 receiver and GPS Software-Based Solution. The receiver possesses 146-dBm acquisition sensitivity and 154-dBm tracking sensitivity. It supports WAAS and EGNOS (European Geostationary Navigation Overlay Service).

Designed to work with an external processor running the RFMD software, it works with ARM, Freescale i.MX, Marvell X-scale, and TI OMAP processors. The RFMD solution performs most of the GPS functions in software. Digital IQ samples created from the GPS RF signal are streamed into host memory.

With the RFMD software, the applications processor takes this information and calculates a position either autonomously or with the use of assistance data. The result is a very scalable and flexible product with a very low parts count. Like most other receivers, the RFMD receivers use the National Marine Electronics Association (NEMA) interface to the display.

Also designed for handsets and other portable devices, GloNav's GNS4540 includes a GPS RF chip and a baseband chip in a single 9- by 6-mm or 6- by 4-mm package. It has 157-dBm acquisition and 159-dBM tracking sensitivity.

Its proprietary DynaTrak feature implements multipath algorithms for robust low-dropout tracking indoors or in very low-signal and rapid-signal transition environments. Designed to work with an external processor via an SPI or UART serial interface, the GNS4540 supports A-GPS on CDMA, WCDMA, and GSM. Power consumption is a low 30 mW in tracking mode.

GPS chip supplier Global Locate recently announced its Hammerhead II GPS receiver, which was jointly developed with Infineon. Like most other new GPS chips, it targets cell phones and other mobile devices. It measures 3.74 by 3.59 by 0.6 mm in a single die, making it one of the smallest. On top of that, sensitivity is –160 dBm.

The Hammerhead II works with an external host processor. Global Locate supplies the software. NXP Semiconductors recently included the Hammerhead II in its GSM/EDGE/UMTS WCDMA cellular reference design. Hammerhead chips are also used in TomTom's PND products.

Interestingly, Bluetooth chip leader CSR recently acquired NordNav Technologies AB and Cambridge Positioning Systems Ltd. Combined, these companies will let CSR provide software-based, low-cost GPS for cell phones and PNDs. Along with CSR's wireless technologies, the company looks to offer a complete GPS solution for less than $1. Products are expected later in the first half of this year.

Texas Instruments also has a new GPS product, the NaviLink 5.0, for mobile phones and other portable devices. This chip supports A-GPS, and it's designed to work with TI's OMAP and OMAP-Vox processors. It also interfaces seamlessly with the TI 2.5G and 3G cell-phone chip sets.

If you have limited experience with RF but want to include GPS in a product, consider DeLorme's new GPS module. Based on STMicroelectronics' STA2056 GPS receiver chip, this 25- by 25-mm module includes the entire GPS position engine, including the low-noise amplifier (LNA), surface-acoustic-wave (SAW) filter, GPS baseband receiver, oscillators, and power conditioning. Just add an external antenna, serial data interface, and power source. DeLorme also offers the Street Atlas USA and Topo USA travel and routing software.

Much of your software will come from your chip vendor, but there are other choices. Rand McNally and TeleCommunications Systems offer Navigator, which includes in-application speech recognition for address entry along with voice output for directions. Navigator provides voice-enabled, hands-free, turn-by-turn directions from a wireless phone or smart phone and includes real-time traffic info.

TeleCommunications Systems' Xypoint SUPL Server conforms to the SUPL standard for IP-based (Internet Protocol) A-GPS location as defined by the Open Mobile Alliance (OMA). The OMA standard uses a network's existing IP infrastructure to pass data between the mobile handset and an AGPS server to support next-generation location-based services (LBS), including navigation, fleet tracking, and mobile marketing.

Another choice is TruePosition's SUPL Server for A-GPS. Wireless operators can deploy this high-accuracy technology to enable high-performance location services, such as local search and navigation, and bring location capabilities to social networking applications. TruePosition has partnered with mobile technology provider GPShopper to offer a location-aware shopping service for carrier-branded and white-labeled services to meet local search and mobile shopping needs.

When adopting A-GPS, you need a method for testing it. One hardware/ software solution comes by way of Rhode & Schwarz. Its complete set of test cases has been verified by the Global Certification Forum (GCF) under Work Item Wi-015. Also, it's designed to work with the company's CRTU-W protocol tester. GCF is a partnership between network operators and mobile handset manufacturers that provides an independent program to ensure global interoperability for 2G and 3G phones.

For more, see "Top 10 Applications For GPS".