Develops a first-order recursive retransmission control policy for a two-station slotted-ALOHA network. Under this policy, terminals adjust their retransmission probability at the end of each slot based on feedback from the current slot. The fundamental difference between a single-station and a two-station network is the interaction between the terminal groups. This interaction occurs because some terminals are in transmission range of both stations; when these terminals transmit to their station they also interfere with the other station. The retransmission control keeps the offered traffic levels near the maximum throughput point and divides the channel capacity fairly between the two terminal groups. The control is intuitively appealing: to acquire channel capacity in proportion to its arrival rate, each group balances efforts to increase its throughput with a certain backoff that permits channel use by the other group.>

A first-order recursive retransmission control policy for a two-station slotted-Aloha network is presented. Under this policy, terminals adjust their retransmission probability at the end of each slot based on feedback from the current slot. The fundamental difference between a single-station and a two-station network (packet radio) is the interaction between the terminal groups, which occurs because some terminals are in transmission range of both stations, resulting in interference. Retransmission control keeps the offered traffic levels near the maximum throughput point and divides the channel capacity fairly between the two terminal groups. The control is intuitively appealing: to acquire channel capacity in proportion to its arrival rate, each group balances efforts to increase its throughput with a certain backoff that permits channel use by the other group.>

A packet radio network with two nodes, both of which access a single radio channel using the slotted ALOHA protocol, is discussed. This model results in a network of two interfering queues. Networks of interfering nodes normally do not have partial balance and thus seldom have product-form solution (PFSs). However, when certain assumptions are made on the network parameters, PFS-controls can be found that yield a PFS for the interfering two-node network. Three such cases are described. In the first two cases, PFS controls are found that control the new packet arrival rate. When operating under these PFS controls, the two-mode network has a simple, closed-form solution of the equilibrium probabilities, the expected throughput, and the expected packet delay. The throughput-delay performance closely approximates that of an exact two-node packet radio model. In the third case, a PFS control is found on one node's transmission probabilities. When operating under this PFS control, the transmission probabilities approach asymptotic values as the packet backlog increases. The network tends to adjust its transmission probabilities to reduce large unequal packet backlogs.>

An optimal routing algorithm for slotted Aloha PRNs (packet radio networks) which minimizes the average packet delay is developed. The network topology and traffic are arbitrary but fixed. The packet radio nodes serve as sources (and sinks) of traffic as well as repeaters which forward packets to other nodes. The optimal routing algorithm captures the important PRN features discussed and avoids routes that result in high levels of interference and delay. Numerical results which show that optimal routing can result in dramatic performance improvements are presented.>

A packet radio network is considered with two nodes; both nodes access a single radio channel using the slotted-ALOHA protocol. This model results in a queuing network with two interfering queues. Networks of interfering queues such as this seldom have closed-form expressions for the equilibrium probabilities; however, when the new-packet arrival rate is properly controlled, the network will have a product-form solution. This control is called PFS-control. When operating under PFS-control, the interfering queuing network has a simple, closed-form expression for the equilibrium probabilities, the expected throughput, and the expected packet delay. Necessary and sufficient conditions for network stability are derived.>