...Private Transmissions
- Cellular in Code System launch commenced in August this year First stage AMPS closures are expected to occur on 31st December and the new system should be fully operational by the end of 2000. One of the attractions of CDMA is that it is designed to share the AMPS spectral allocation in the 800 MHz band (dividing the radio spectrum into carriers 1.25MHz wide) and with the development of dual AMPS/CDMA handsets that automatically adjust to suit incoming signals, it means minimum disruption to customers during the changeover period. Being on the same frequency, antennas designed to optimise 800MHz cellular signals, like the Moonraker CEL800, will also improve performance and range with CDMA. While CDMA technology is brand new, like many other invention - the linear motor to name but one - it has had to wait its time for the promise of what it has to offer to be achievable. The origins of CDMA date back to the end of World War II and the 1950s when spread spectrum technology was first invented. Initially its use was confined to small aperture satellite services and to military applications, where it was used to avoid jamming and detection of sensitive transmissions. However, for personal communications, CDMA had to await the development of a way for the base station to control the transmission power of the mobiles. In the late 1980s power control was added to the GMS system to reduce average interference levels. This made CDMA feasible as it was now possible to eliminate narrowband interference which could prevent correct demodulation of the wanted signal (near-far effect), and to ensure that all accessing signals could be received at a similar amplitude. Economically too, prospects were more attractive as low cost high density digital integrated circuits, reducing the size, weight and cost of subscriber stations to an acceptably low level, were now available. In AMPS technology FDMA (frequency division multiple access) is used to allocate a single channel to each call. With the advent of GSM, TDMA (time division multiple access) made it possible for more calls to share channels by allocating time slots to 8 users over a 200 kHz bandwidth divided into 25 kHz channels with 8 users per channel.
 CDMA is radically different in that noise-like carrier waves are used to enable all of the users to have access to all of the bandwidth all of the time. Calls are separated by allocating each receiver a unique code which it shares with the base station. All signals then share the available bandwidth and are heard as a low noise. The reverse process is applied at the receiver end, to separate the calls. Thus there is no need for complex systems to regulate time slots as in GMS. Base stations are synchronised to a common time reference (GPS) and guard bands are used to prevent interference. Pseudo-noise is generated digitally rather than using true thermal noise so that high density digital devices can be employed simplifying transmitters and receivers. With all users occupying all of the same spectrum at the same time, the receiver is able to adjust the signal to noise ratio at the detector to allow for all noise extraneous to the wanted signal and there is no worst case scenario to take into account.
Channels are no longer separated by analog filtering in the frequency domain but by means of a pseudo-random modulation, applied and removed in the digital domain. The 9.6 kilobits p/s signal is spread to a transmission rate of 1.23 Megabits p/s and digital codes are applied. The signal-to-noise ratio is enhanced by narrowband filtering that rejects most of the interference power. Modulation, either BPSK (binary phase shift keying) or more usually QPSK (quadrature), as in cdmaOne, takes place in the binary domain and transmitted signals are carefully band limited. The pulse timing of the modulation in cdmaOne is more generally random than GMS causing less problems with hearing aids. cdmaOne is able to overcome the potential propagation path loss difference between callers located close in and at cell boundaries, the 'near-far' problem mentioned last month, by controlling transmitter power so that received powers from all users are roughly equal. To explain the significance of the problems, a 30dB difference between the largest and smallest path losses means a 60dB difference in the signal-to-noise ratio of the closest and farthest user because these are the received powers. With FDMA a signal can be detected simply by using an off-the-shelf scanner. TDMA detection is more difficult but not impossible for the expert. Obviously CDMA privacy is much better. It is very difficult to locate calls in the noise stream let alone crack the code to gain illegal access to the data you want. Range is another area of gain. The restricted 35km cells of GSM networks are designed to ensure than signals in one frequency cell do not interfere with another frequency cell. With CDMA restriction is no longer necessary as all cells use the same frequency.
 While in the majority of cases CDMA cells will have a 35km range, this will not always apply, especially where network boundaries are concerned. Generally, cells are designed for a maximum 60km range but it is also possible to have 'boomer' cells with up to 140km range, depending on terrain characteristics. Some of the new mountain top cells in Australia have a 120km range. This flexibility makes it possible to achieve better coverage in more remote locations. Handover at cell boundaries no longer presents problems. The gaps in handover in the AMPS analog system preventing data communications was improved in GSM digital technology through monitoring bit error rate to determine when handover occurs. In cdmaOne a call can be monitored by more than one base station at a time ensuring a seamless handover. This opens up future opportunities for data transmissions like internet, radio and video. Multipath interference has plagued cellular mobile communications from its outset, especially in built-up locations. With AMPS being an analog signal, only fading normally occurs, but with the GSM digital signal, there are often significant gaps in transmission, and the higher the frequency, the greater the incidence. In these circumstances, CDMA is better equipped to cope, as when the multi-path signal takes longer than the predetermined time for a signal to reach the receiver, it is discarded as noise. Only when it is within this limitation and cannot be separated does flat fading occur. With larger cell capacity and less signal disruption from multi-path interference, it becomes easier to site base stations and the number of actual cells required is reduced. The cost of installation becomes more economical with less environmental impact. Because the effective noise is the sum of all other-user signals, cell capacity is now determined by the balance between the required signal to noise ratio for each user and the spread spectrum processing gain. In effect, above a certain level of users, signal quality becomes unacceptable. Looking to the future, the developers of cdmaOne have been keen to build a system with a capacity for expansion and growth, with the emphasis on compatibility, looking towards cdma2000, rather than continue down the path of system replacement. However, W-CDMA (wideband) which has a larger chip rate 4.096 Mcps (may be revised to 3.84) compared to cdmaOne's 3.6864 Mcps is also in the background. But that's some time in the future... |
|