Multiple Acess Categories in Wireless Systems

Depending on how much coordination is required to access the shared resources, the access methods may fall into the following categories:

Scheduled Multiple-Access Methods. The access methods in this category make use of the signal insulation principle in which the information of different sources is transmitted on nonoverlapping channels. In this case, collision does not occur. The methods providing for scheduled multiple-access capabilities in the frequency domain, time domain, code domain, and space domain are, respectively, frequency division (FDMA), time division (TDMA), code division (CDMA), and space division multiple access (SDMA).

Random Multiple-Access Methods. The access methods in this category make use of a contention scheme in which no or little coordination is provided. The information of different sources is cast onto a common channel, and collision may occur. Collision is usually detected by an acknowledgment mechanism in which the terminal sending the packet waits for a confirmation of the successful reception of it. The widely used methods providing for random multiple-access capabilities are Pure ALOHA, Slotted ALOHA, Tree Algorithm, FCFS Algorithm, CSMA (nonslotted, slotted, 1-persistent. p-persistent), CSMA/BT, CSMA/DT, CSMA/CD, and CSMA/CA.

Controlled Multiple-Access Methods. The access methods in this category make use of a deterministic or contentionless strategy in which, by means of a control signal, permission to send is granted to terminals individually, so that only one terminal is allowed to access the medium at a time. The main access schemes here are polling, token ring, and token bus.

Hybrid Multiple-Access Methods. The access methods in this category make use of a combination of scheduled, contention, and deterministic schemes in which some degree of coordination is included into the random-access mechanisms or some other more sophisticated deterministic methods are employed. Examples of such schemes include R-ALOHA,OriginalPRMA,PRMA++,AggressivePRMA,FastPRMA, Multirate PRMA, PRMA/DA, PRMA/ATDD, DQRUMA, DSA++, DTDMA/PR, MASCARA, DTDMA/TDD, Pure RAMA, T-RAMA, F-RAMA, and D-RAMA.

For stream, steady-flow traffic, the available resources are more efficiently utilized if allotted on a scheduled access basis. This is because contention in this case may lead to a significant amount of packet loss. For intermittent, burst (random) traffic, better efficiency can be achieved if the available resources are seized on a random-access basis. This is because scheduled allocation in this case may lead to an inefficient use of the resources. For intermittent, random traffic if a high incidence of packet collision occurs, the use of controlled-access methods is recommended. For stream traffic and bursty traffic simultaneously sharing the same resource, hybrid access methods may be applied.

In a multiuser system, capacity is certainly a sensitive issue. The efficiency with which the available resources are used determines the number of users the system can support. Higher efficiency is achieved if the resources are made available to all users and they are assigned on a demand basis. This characterizes what is known as demand-assigned multiple-access (DAMA) or, simply, multiple access. The assignment of the resources on a demand basis suggests that some initial protocol be established before the resource is actually assigned. Handshaking protocols usually typify bursty traffic, whereas payload information can be characterized by either bursty or steady-flow traffic.

Certainly, the ultimate aim of any communication network is provision of resources to convey payload information. Overhead traffic, although necessary, should be kept to a minimum to increase efficiency. The initial access protocols may use the same physical resources as those used to convey payload information. However, splitting the resources into two distinct groups—one for initial access purposes and another for payload information purposes—is a practice common to networks where a great amount of signaling is required. And this is true of wireless systems. The proportion of channels dedicated to each function varies according to the needs.