Multiple Access Wireless Communications

The goals of multiple access communications systems, meaning cellular and PCS, are:

• Near-wireline quality voice service
• Near-universal geographical coverage
• Low equipment cost, both subscriber stations and fixed plant
• Mimimum number of fixed radio sites

Regulatory agencies have allocated limited bandwidth to these services, so that the solutions must achieve high spectral efficiency, measured in Erlangs per unit service area, per MHz. Cellular operators have 25 MHz each, split between the two directions of communications. The PCS service in the United States has three 30 MHz and three 10 MHz allocations, also split.

Practical implementations of cellular systems having hundreds of channels became practical with the availability of compact, low-cost frequency synthesizers. Microprocessor control permits complex control message dialogs that implement sophisticated call control protocols.

The cellular solution, originally designed by Bell Telephone laboratories in the 1970s, makes use of multiple fixed stations, or cells (the term cell sometimes refers to the equipment, sometimes to the service area). Each site services subscriber stations within a limited geographical area. When a subscriber moves between cells, over-the-air messaging is used to transfer control from the old cell to the new cell. This transfer of control is termed handoff or handover.

The original system was called the Advanced Mobile Phone System, or AMPS. It is the system we use throughout North America. Similar systems, with slight variations, are Nordic Mobile Telephone (NMT) in Scandinavia, and Total Access Communications System (TACS) used in the United Kingdom, China, and other countries. Spectral allocations are in the 800-900 MHz region.

Several hundred channels are available within the spectrum allocation. One channel of one base station is used for each conversation. Upon handoff, the subscriber station is directed via messaging to discontinue use of the old channel and tune to the new one, on which it will find the new cell.

Frequency Reuse

Central to the cellular concept is the concept of frequency reuse. Although there are hundreds of channels available, if each frequency were assigned to only one cell, total system capacity would equal to the total number of channels, adjusted for the Erlang blocking probability: only a few thousand subscribers per system. By reusing channels in multiple cells the system can grow without geographical limits.

Reuse is critically dependent upon the fact that the electromagnetic field attenuation in the cellular bands tends to be more rapid with distance than it is in free space. Measurements have shown repeatedly that typically the field intensity decays like R^-n, with 3 < n < 5. In free space n = 2. In fact, it is easily shown that the cellular concept fails completely due to interference that grows without bound if the propagation is exactly free space.

Typical cellular reuse (pre-CDMA, that is!) is easily rationalized by considering an idealized system. If we assume that propagation is uniformly R^-n, and that cell boundaries are at the equisignal points, then a planar service area is optimally covered by the classical hexagonal array of cells ...
Seven sets of channels are used, one set in each colored cell. This seven-cell unit is then replicated over the service area.

No similarly colored cells are adjacent, and therefore there are no adjacent cells using the same channel. While real systems do not ever look like these idealized hexagonal tilings of a plane, the seven-way reuse is typical of that achieved in practice.

The capacity of a K-way reuse pattern is simply the total number of available channels divided by K. With K=7 and 416 channels, there are approximately 57 channels available per cell. At a typical offered load of 0.05 Erlangs per subscriber, each site supports about 1140 subscribers.

Life is, of course, much more complicated than this simple picture, but the numbers are in the right ballpark!

Antenna Sectorization

The pictures above assume that the cells are using omnidirectional antennas. It might be expected that system capacity could be increased by antenna sectorization. Sites are in fact sectorized by the operators, usually three-ways. That is, each site is equipped with three sets of directional antennas, with their azimuths separated by 120°. Unfortunately the sectorization does not in practice lead to an increase in capacity. The reason is that the sector-to-sector isolation, often no more than a few dB, is insufficient to guarantee acceptably low interference. Only in part is this due to the poor front-to-back ratio of the antennas. The vagaries of electromagnetic propagation in the real world also conspire to mix signals between sectors. The practical result of sectorization is only an increase in coverage because of the increased forward gain of the directional antenna. Nothing is gained in reuse. The same seven-way cell reuse pattern applies in sectored cells as in omnidirectional cells. Viewed from the standpoint of sectors, the reuse is K = 7 *3 = 21, not 7.

| Index | Topics | Glossary | Standards | Bibliography | Feedback |

Copyright © 1996-1999 Arthur H. M. Ross, Ph.D., Limited