| 摘要 |
Mobile broadband traffic has been exploding in wireless networks resulting in an increase of interferences and reduced operator control. Networks are also becoming more heterogeneous putting additional demand in interference management. Scheduler schedules uplink transmissions from UEs based on a load prediction algorithm that typically assumes worst case. However, UEs do not always use full power granted, and thus, much of granted radio resources are wasted. To address these and other issues, techniques to accurately predict/estimate other cell interferences and thermal noise separately and to accurately predict/estimate load utilization probabilities, rise-over-thermals, and load factor biases are described. The described techniques are applicable in a radio network node with any number of antenna branches. The techniques can be used to schedule UEs to more fully utilize available radio resources. Extended Kalman filtering can be adapted for use in estimation providing low order computational complexity. |
| 主权项 |
1. A method performed by at least one processor of a radio network node associated with a cell of interest in a wireless network, the at least one processor operating as a noise and interference predictor, wherein the radio network node comprises a total of M antenna branch(es), M being a positive integer, the method comprising:
receiving a scheduled load factor Lown(to) applicable at time t=to; receiving a used load factor Lown(to) applicable at time the t=to; receiving, for each antenna branch m, m=1 . . . M, a corresponding total wideband power PRTWPm(to) applicable at the time t=to; determining, by the at least one processor, a load utilization probability estimate {circumflex over (p)}load(t1) and an other cell interference estimate {circumflex over (P)}other(t1), both applicable at time t=t1, based on the scheduled load factor Lown(to), the used load factor Lown(to), and the M total wideband power(s) PRTWPm(to), wherein t1−to=T>0; for each antenna branch m, determining by the at least one processor a corresponding thermal noise estimate {circumflex over (P)}Nm(t1) and a corresponding rise-over-thermal estimate {circumflex over (R)}oTm(t1), both applicable at the time t=t1, based on the corresponding total wideband power PRTWPm(to), and making, by the at least one processor, a scheduling decision for network traffic based at least in part on at least one of the load utilization probability estimate {circumflex over (p)}load(t1) and the other cell interference estimate {circumflex over (P)}other(t1), the corresponding thermal noise estimate {circumflex over (P)}Nm(t1) and the corresponding rise-over-thermal estimate, and wherein Lown(t) represents resource grants scheduled for use at time t to one or more cell terminals in the cell of interest, Lown(t) represents resource grants used at time t by the same cell terminals, and each PRTWPm(t) represents the total wideband power present at time t at the antenna branch m; wherein {circumflex over (p)}load(t) represents an estimate of the load utilization probability pload(t) which expresses a probability of resources grants for use at time t actually being used at time t; wherein {circumflex over (P)}other(t) represents an estimate of the other cell interference Pother(t) which expresses a sum of interferences present in the cell of interest at time t due to wireless activities applicable in one or more cells other than in the cell of interest, wherein each {circumflex over (P)}Nm(t) represents an estimate of the corresponding thermal noise {circumflex over (P)}Nm(t) which expresses a sum of undesired signals present at time t at the antenna branch m excluding the other cell interference Pother(t), and wherein each {circumflex over (R)}oTm(t) represents an estimate of the corresponding rise-over-thermal RoTm(t) present at time t at the antenna branch m, the rise-over-thermal being defined asRoTm(t)=PRTWPm(t)PNm(t). |
