To better understand how developing mobile location services applications differ from traditional wireline applications, it is important to understand the basic principles of wireless networks. This section provides a brief and simplified overview of mobile network architectures. The discussion includes the basics of radio spectrum, cellular networks, and wireless data.
Wireless networks are based on radio principles that are now more than 100 years old. Radio signals are electromagnetic radiation, a category that includes light and infrared waves as well. Radio signals are considered transverse waves, which means they have wavelength and frequency (see Figure 2.1).
The wavelength is the distance between the peaks of sequential waves and the frequency is the number of cycles per second (Hz). A transverse wave's speed can be calculated by multiplying the wavelength and the frequency, but all radio waves travel at the speed of light. When waves pass through solid material they are slowed down, but even in wireless communication systems the waves that pass through air and clouds are still traveling at about 300,000 kilometers per second.
Because the speed of travel is constant, increasing frequency shortens wavelength in a mathematical pattern, and usually only one is specified. The amplitude of the wave is its height, or distance from the axis to the peak. As waves move away from their source, and spread over a wider area, the amplitude decreases. This process of losing energy is called attenuation.
The entire range of electromagnetic radiation is called the spectrum. Electromagnetic waves that are well suited to communications are known as radio, and have a lower frequency and longer wavelength than other forms of radiation (see Figure 2.2).
Generating electromagnetic waves requires accelerating a moving electrical charge by either changing its speed or direction. Radio transmitters work by vibrating electrons, the charged particles that surround all atoms. The faster the electrons vibrate, the higher the frequency of the resulting radio wave. A radio receiver uses the same process in reverse. The radio waves stir up electrons in the antenna, which create electric currents.
Radio signals with a high frequency have a much shorter range than radio signals with a low frequency because shorter wavelengths suffer greater attenuation. Different applications require different types of radio spectrum. Because spectrum is a scarce resource and is not unlimited, its use is subject to government licensing. Radio signals used for wireless communications are referred to as microwaves because of their very small wavelengths. The microwave spectrum ranges from .4 GHz to 100 GHz, and their high bandwidth makes them ideal for communications.
Traditional radio systems were designed to transmit sound. Sound is an analog signal that can be represented by a continuous wave similar to electromagnetic radiation. Simple analog broadcasting or cellular telephony systems convert sound waves to radio waves and then back to sound waves again. However, wireless networks are being increasingly used for data as much as for voice. Data is inherently digital, and a digital radio wave encodes data (the 0s and 1s used by computer systems) to radio waves and back again.
Digital radio has a number of significant advantages over analog radio. Digital systems allow a radio receiver to distinguish between static interference and the signal, so the static can be ignored. Digital signals are often also encoded with additional checksum data that allows the receiver to perform a mathematical calculation to make sure it received the transmission correctly. If not, it can be sent again. Digital signals can be compressed to use spectrum more efficiently and encrypted to prevent eavesdropping (see Chapter 7 for a more detailed analysis of security concerns in wireless networks and mobile location services). Finally, digital signals can take advantage of timing techniques to share communications channels in bandwidth-efficient ways.
Whether for analog voice, digital voice, or digital data, information has to be converted to radio waves before it can be transmitted. The process of altering a radio wave of a specific frequency so that useful information can be extracted from it is called modulation. The two primary methods of encoding a radio signal are amplitude modulation (AM) and frequency modulation (FM). AM encodes a radio signal by varying the height of the waves in accordance with an information signal. Because this method uses bandwidth inefficiently, it is rarely used in modern wireless systems.
FM keeps the amplitude constant and instead alters the frequency and wavelength of the radio signal. Because the amplitude is constant, the FM transmitter can operate at full power all the time and efficiently use the full spectrum allocated to it. Most digital wireless systems encode data using phase modulation, which is a special form of FM. Instead of just changing frequency and wavelength of a radio signal, phase modulation also quickly moves them to different points in their cycle, which is useful for encoding wireless data.
Mobile phone systems are comprised of a network of cells, each with a powerful radio transmitter at its center. This is both because the radio signals most effective for carrying digital voice and data are short ranged, and because the cellular design is modular and can provide redundancy and failover capabilities. The base stations are typically connected to each other via high-speed fiber, and then to the public phone system and the Internet. As a mobile device moves through the network it is passed from one base station to another, accessing services through the base station of the cell it is in.
Figure 2.3 shows a sample GSM network architecture. The mobile station (MS) is the mobile phone or handheld client device. The MS includes a subscriber identity module (SIM), used for authentication and security, and the hardware and software specific to the radio interface, called the mobile equipment (ME). The network switching subsystem (NSS) provides the basic switching, profile management and mobility management functions. The mobile switching center (MSC) provides the switching functions. The location of the MS is tracked by the home location register (HLR) and visitor location register (VLR). When an MS moves from its HLR, it is registered in the VLR of the system it is visiting. The HLR is then informed of the location of the MS. The NSS also manages subscriber authentication and security, which is handled in the authentication center (AuC). Connecting the MS to the NSS is the base station subsystem (BSS). The BSS consists of base transceiver stations (BTSs) and a base station controller (BSC). The BSC performs the switching functions for the BSS and is connected to the MSC. The BSC performs hand-off management and radio channel allocation and release. The BTS contains the transmitter, receiver, and signaling equipment necessary to communicate to the MS over the radio interface.
To make most efficient use of the spectrum in a cell, mobile operators need to allow it to be shared using multiple access techniques. The most common method used today (including the European GSM standard) uses what is called Time Division Multiple Access (TDMA). TDMA divides a given band into a number of time slots that correspond to a communications channel. A mobile phone will transmit and send on only one slot and then remain quiet until its next turn to communicate. In GSM a time slot is 577 microseconds, so a mobile device could miss a scheduled slot and it would still not be noticeable to a human listener. The design of TDMA systems allows for them to easily upgrade to higher speed by allowing the mobile device to receive or transmit on more than one slot at a time. This is the principle behind GPRS (General Packet Radio Service). Another multiplexing technique used to make efficient use of spectrum is CDMA (Code Division Multiple Access). In CDMA, every signal is sent at the same time, but each signal is encoded differently so receivers can understand it. This is known as spread spectrum.
Sending wireless data over a standard second-generation (2G) GSM mobile network requires full-time use of the voice channel, and allows rates of about 14.4 kbps. Third-generation (3G) mobile systems provide an always-on data connection at vastly faster rates than is possible in 2G mobile communications. An intermediate step that is a less expensive and a relatively straightforward upgrade from GSM is GPRS, sometimes also called 2.5G. GPRS is a packet-switched network, which uses bandwidth only when sending and receiving data. This allows it to be shared by numerous mobile devices at the same time, just as dial-up Internet users share one fast Internet connection from an ISP. The specification for GPRS allows it to provide up to 115 kbps.
Figure 2.4 extends the GSM architecture displayed in Figure 2.3 to include a sample GPRS architecture. The serving GPRS support node (SGSN) transmits and receives packets between the MS and the device it is communicating with over the public switched data network (PSDN). The gateway GPRS support node (GGSN) translates between the SGSN and the PSDN and supports a variety of connectionless and connection-oriented protocols, including Transmission Control Protocol/Internet Protocol (TCP/IP). The GGSN and the SGSN use the GSM location databases (HLR and VLR) to track the location of the MS to maintain connectivity as the MS moves through the mobile network.
The mobile location service infrastructure we are concerned with is accessed via the PSDN by an MS. The MS might contain proprietary client software to interact with your mobile location service, or it might use a thin client like WAP that requires no special client software. Mobile positioning technology requires additional handset or network elements and is explored in detail in Chapter 5.