Author_Institution :
Dept. of Electr. Eng., California Inst. of Technol., Pasadena, CA, USA
Abstract :
Studies the performance of channel assignment algorithms for “channelized” (e.g., FDMA or TDMA) cellular telephone systems, via mathematical models, each of which is characterized by a pair (H,p), where H is a hypergraph describing the channel reuse restrictions, and p is a probability vector describing the variation of traffic intensity from cell to cell. For a given channel assignment algorithm, the authors define T(r) to be the amount of carried traffic, as a function of the offered traffic, where both r and T(r) are measured in Erlangs per channel. They show that for a given H and p, there exists a function TH,p(r), which can be computed by linear programming, such that for every channel assignment algorithm, T(r)⩽TH,p(r). Moreover, they show that there exist channel assignment algorithms whose performance approaches TH,p (r) arbitrarily closely as the number of channels increases. As a corollary, they show that for a given (H,p) there is a number r0 , which also can be computed by linear programming, such that if the offered traffic exceeds r0, then for any channel assignment algorithm, a positive fraction of all call requests must be blocked, whereas if the offered traffic is less than r0, all call requests can be honored, if the number of channels is sufficiently large. The authors call r0, whose units are Erlangs per channel, the capacity of the cellular system
Keywords :
cellular radio; frequency allocation; frequency division multiple access; graph colouring; linear programming; mobile radio systems; telecommunication traffic; telephone traffic; time division multiple access; FDMA; TDMA; call requests; channel assignment algorithm; channel assignment algorithms; channel reuse restrictions; channelized cellular telephone systems; hypergraph; linear programming; offered traffic; performance limits; probability vector; traffic intensity; Channel capacity; Communication industry; Frequency division multiaccess; Linear programming; Mathematical model; Radio spectrum management; Telecommunications; Telephony; Time division multiple access; Traffic control;