Bounds on the capacity of FCFS multiple-access algorithms (Q2276862)

A channel is randomly accessed by multiple broadcasts. The flow of messages has Poisson behaviour and the transfer is being realized in subsequent unit time intervals (windows). In all the windows all the broadcasts know if the message has been throughput, or more than one message have met each other, or there has been no transfer at all. In the second case the message is sent repeatedly. Random multiple access algorithms are dedicated to successfully transfer all the information so that the numbers of lost messages is finite. The class of random multiple access algorithms with the FCFS rule is investigated. For certain subclasses, the capacity is estimated. A hypothesis is formulated as to the value of the capacity on the whole class. Some of the results have previously been proven by \textit{R. Z. Khas'minskij} [Dokl. Akad. Nauk SSSR 301, 1057-1060 (1988; Zbl 0666.94001); English translation in Sov. Math., Dokl. 38, No.1, 184-187 (1989)].