Wireless Enters the 4th(G) Dimension

Wireless mobile-communications systems are uniquely identified by "generation" designations. Introduced in the early 1980s, first-generation (1G) systems were marked by analog-frequency modulation and used primarily for voice communications. Second-generation (2G) wireless-communications systems, which made their appearance in the late 1980s, were also used mainly for voice transmission and reception. The major change between 1G and 2G systems was that 2G systems are digitally based, employing TDMA (time-division multiplexing access), CDMA (code-division multiplexing access), or FDM (frequency-division multiplexing) techniques. 2G digital-cellular systems support voice services at rates up to around 10 kbps.

The wireless system in widespread use today goes by the name of 2.5G—an "in-between" service that serves as a stepping stone to 3G. Whereby 2G communications is generally associated with Global System for Mobile (GSM) service, 2.5G is usually identified as being "fueled" by General Packet Radio Services (GPRS) along with GSM.

General Packet Radio Services (GPRS), for 2.5G communications systems, is based on GSM. GPRS is packet-based and promises real-world data rates from 56 to 114 Kbps (the theoretical maximum is 144 to 170 Kbps) with a continuous Internet connection for mobile phone and computer users. The higher data rate lets users take part in video conferences, receive streaming video for news and entertainment, and interact with multimedia Web sites and applications using mobile handheld devices and portable computers.

Theoretically, a packet-based service such as GPRS should be less expensive than a circuit-switched (2G) service, since you use a communications channel on a shared-use, as-packets-are-needed basis rather than have a channel dedicated only to one user at a time. However, it will probably be awhile before service providers adapt this business model, since margins are greater using the service-limiting, "one channel per user" model.

GPRS should also simplify making applications available to mobile users because the faster data rate means that the middleware you currently use to adapt applications to the slower speed of wireless systems is no longer needed. For example, as GPRS becomes available, mobile users of a virtual private network (VPN) will be able to access the network continuously rather than through a dial-up connection.

The performance of 2.5G (GPRS) exceeds that of Wireless Access Protocol (WAP). Currently, you can use WAP-enabled 2G phones to access the web and to send and receive data. However, performance suffers from sluggish performance due to slow and dropped connections, high airtime cost, and limited content.

3G systems, making their appearance in late 2002 and in 2003, are designed for voice and paging services, as well as interactive-media use such as teleconferencing, Internet access, and other services. The problem with 3G wireless systems is bandwidth—these systems provide only WAN coverage ranging from 144 kbps (for vehicle mobility applications) to 2 Mbps (for indoor static applications). Segue to 4G, the "next dimension" of wireless communications.

If you are looking for a hard and fast definition for 4G wireless, you won't find it. Not expected for deployment until at least 2006, and possibly not until 2010, the 4G feature set and underlying technologies are not yet fixed. Even so, we can still compare some probable 4G vs. existing 3G parameters in Table 1.

One of 4G's technology drivers is orthogonal frequency-division multiplexing (OFDM). Patented over thirty years ago, only recently has OFDM gained widespread use. OFDM, a form of multi-carrier modulation, works by dividing the data stream for transmission at a bandwidth B into N multiple and parallel bit streams, spaced B/N apart (Figure 1). Each of the parallel bit streams has a much lower bit rate than the original bit stream, but their summation can provide very high data rates. N orthogonal sub-carriers modulate the parallel bit streams, which are then summed prior to transmission.

Figure 1:  OFDM modulation provides efficient use of bandwidth and decreased inter-channel interference

An OFDM transmitter accepts data from an IP network, converting and encoding the data prior to modulation. An IFFT (inverse fast Fourier transform) transforms the OFDM signal into an IF analog signal, which is sent to the RF transceiver. The receiver circuit reconstructs the data by reversing this process. With orthogonal sub-carriers, the receiver can separate and process each sub-carrier without interference from other sub-carriers. More impervious to fading and multi-path delays than other wireless transmission techniques, ODFM provides better link and communication quality.

4G's error-correction will most likely use some type of concatenated coding and will provide multiple Quality of Service (QoS) levels. Forward error-correction (FEC) coding adds redundancy to a transmitted message through encoding prior to transmission. The advantages of concatenated coding (Viterbi/Reed-Solomon) over convolutional coding (Viterbi) are enhanced system performance through the combining of two or more constituent codes (such as a Reed-Solomon and a convolutional code) into one concatenated code. The combination can improve error correction or combine error correction with error detection (useful, for example, for implementing an Automatic Repeat Request if an error is found). FEC using concatenated coding allows a communications system to send larger block sizes while reducing bit-error rates.

• Support for interactive multimedia, voice, streaming video, Internet, and other broadband services
  
  
• High speed, high capacity, and low cost-per-bit
  
  
• Global access, service portability, and scalable mobile services
  
  
• Seamless switching, and a variety of QoS-driven services
  
  
• Better scheduling and call-admission-control techniques
  
  
• Ad-hoc and multi-hop networks (the strict delay requirements of voice make multi-hop network service a difficult problem)
  
  
• Better spectral efficiency
  
  
• Seamless network of multiple protocols and air interfaces (since 4G will be all-IP, look for 4G systems to be compatible with all common network technologies, including 802.11, WCDMA, Bluetooth, and HyperLAN)
  
  
• An infrastructure to handle pre-existing 3G systems along with other wireless technologies, some of which are currently under development.

UWB
Currently under severe FCC low-power restrictions, UWB uses its power differently from other, more traditional transmission techniques. A UWB transmitter spreads its signal over a wide portion of the RF spectrum, generally 1 GHz wide or more, above 3.1GHz. The FCC has chosen UWB frequencies to minimize interference to other commonly used equipment, such as televisions and radios. This frequency range also puts UWB equipment above the 2.4 GHz range of microwave ovens and modern cordless phones, but below 802.11a wireless Ethernet, which operates at 5 GHz.

UWB equipment transmits very narrow RF pulses—low power and short pulse period means the signal, although of wide bandwidth, falls below the threshold detection of most RF receivers. Traditional RF equipment uses an RF carrier to transmit a modulated signal in the frequency domain, moving the signal from a baseband to the carrier frequency the transmitter uses. UWB is "carrier-free", since the technology works by modulating a pulse, on the order of tens of microwatts, resulting in a waveform occupying a very wide frequency domain. The wide bandwidth of a UWB signal is a two-edged sword. The signal is relatively secure against interference and has the potential for very high-rate wireless broadband access and speed. On the other hand, the signal also has the potential to interfere with other wireless transmissions. In addition, the low-power constraints placed on UWB by the FCC, due to its potential interference with other RF signals, significantly limits the range of UWB equipment (but still makes it a viable LAN technology).

One distinct advantage of UWB is its immunity to multi-path distortion and interference. Multi-path propagation occurs when a transmitted signal takes different paths when propagating from source to destination. The various paths are caused by the signal bouncing off objects between the transmitter and receiver—for example, furniture and walls in a house, or trees and buildings in an outdoor environment. One part of the signal may go directly to the receiver while another, deflected part will encounter delay and take longer to reach the receiver. Multi-path delay causes the information symbols in the signal to overlap, confusing the receiver—this is known as inter-symbol interference (ISI). Because the signal's shape conveys transmitted information, the receiver will make mistakes when demodulating the information in the signal. For long-enough delays, bit errors in the packet will occur since the receiver can't distinguish the symbols and correctly interpret the corresponding bits.

The short time-span of UWB waveforms—typically hundreds of picoseconds to a few nanoseconds—means that delays caused by the transmitted signal bouncing off objects are much longer than the width of the original UWB pulse, virtually eliminating ISI from overlapping signals. This makes UWB technology particularly useful for intra-structure and mobile communications applications, minimizing S/N reduction and bit errors.

Millimeter Wireless
Using the millimeter-wave band (above 20 GHz) for wireless service is particularly interesting, due to the availability in this region of bandwidth resources committed by the governments of some countries to unlicensed cellular and other wireless applications. If deployed in a 4G system, millimeter wireless would constitute only one of several frequency bands, with the 5 GHz band most likely dominant.

Smart Antennas
A smart antenna system comprises multiple antenna elements with signal processing to automatically optimize the antennas' radiation (transmitter) and/or reception (receiver) patterns in response to the signal environment. One smart-antenna variation in particular, MIMO, shows promise in 4G systems, particularly since the antenna systems at both transmitter and receiver are usually a limiting factor when attempting to support increased data rates.

MIMO (Multi-Input Multi-Output) is a smart antenna system where 'smartness' is considered at both transmitter and the receiver. MIMO represents space-division multiplexing (SDM)—information signals are multiplexed on spatially separated N multiple antennas and received on M antennas. Figure 2 shows a general block diagram of a MIMO system. Some systems may not employ the signal-processing block on the transmitter side.

Figure 2:  A MIMO (Multi-input Multi-Output) antenna system can significantly improve the capacity of a 4G wireless system

Multiple antennas at both the transmitter and the receiver provide essentially multiple parallel channels that operate simultaneously on the same frequency band and at the same time. This results in high spectral efficiencies in a rich scattering environment (high multi-path), since you can transmit multiple data streams or signals over the channel simultaneously. Field experiments by several organizations have shown that a MIMO system, combined with adaptive coding and modulation, interference cancellation, and beam-forming technologies, can boost useful channel capacity by at least an order of magnitude.

Both service providers and users want to reduce the cost of wireless systems and the cost of wireless services. The less expensive the cost of the system, the more people who will want to own it. Competition among service providers requires a continuous drop in service cost (remember just a few years ago when cell-phone time was almost $1 per minute?). The high bandwidth requirements of upcoming streaming video necessitates a change in the business model the service providers use—from the dedicated channel per user model to one of a shared-use, as-packets-are-needed model. This will most likely be the model service providers use when 4G systems are commonplace (if not before).

Increased speed is a critical requirement for 4G communications systems. Data-rate increases of 10-50X over 3G systems will place streaming audio and video access into the hands of consumers who, with each wireless generation, demand a much richer set of wireless-system features. Power control will be critical since some services (such as streaming video) require much more power than do others (such as voice).

4G's flexibility will allow the integration of several different LAN and WAN technologies. This will let the user apply one 4G appliance, most likely a cell-phone/PDA hybrid, for many different tasks—telephony, Internet access, gaming, real-time information, and personal networking control, to name a few. A 4G appliance would be as important in home-networking applications as it would as a device to communicate with family, friends, and co-workers.

Finally, a 4G wireless phone would give a user the capability of global roaming and access—the ability to use a cell phone anywhere worldwide. At this point, the 4G wireless system would truly go into a "one size fits all" category, having a feature set that meets the needs of just about everyone.

Jim Lipman is a consultant providing marketing, writing, and other electronics industry services, specializing in EDA tools and ASIC/SoC design methodologies. His job experience includes chip-design R&D, marketing, marcom, technical editing, and on-line publishing of technical content for engineers.

Author's Note
For more information regarding evolving 4G wireless systems, visit the Wireless World Research Forum (WWRF). Founded by Alcatel, Ericsson, Motorola, Nokia, and Siemens in 2001, the forum's objective "is to formulate visions on strategic future research directions in the wireless field, among industry and academia, and to generate, identify, and promote research areas and technical trends for mobile and wireless system technologies." Forum members comprise representatives from manufacturers, network operators, universities, R&D centers, and small and medium enterprises.