Common use of Scheduler Performance Clause in Contracts

Scheduler Performance. The SINR values are measured at the output of the cooperation area scheduler for all resource blocks and all UEs. Then, by using the SINR to throughput mapping from [KPK+09] as described in ▇▇▇▇▇▇▇▇ ▇▇-▇, summing up all contributions and dividing the troughput by the number of cells, the average cell throughput without overhead is obtained. To obtain the resulting spectral efficiency (SE) one has to multiply with (1-0.43)=0.57, which includes the 17 This is an important difference to many known schedulers that use very restrictive selection criteria for CoMP UEs, based on small threshold values. overhead for synchronization channels, guard intervals and bands, reference signals and control channels as known from LTE Release 8. For the example according to results from ▇▇▇▇▇▇▇▇ ▇▇-▇, the attained spectral efficiency was:   ▇ ▇▇▇▇ ▇▇▇▇▇▇▇▇▇▇▇▇ ,UEk  = 6.5 bits/s/Hz/cell. This corresponds to a brutto data rate TPbrutto (excluding all overhead) of the transmission system of TPbrutto = 6.5/0.57 = 11.4 bits/s/Hz/cell. Compared to the values reported in [3GPPTR1], this would be an improvement of a factor of 2.1 over Rel10 MU MIMO and of 1.6 over the best JT CoMP results in [3GPPTR1] so far. However, it should be noted that due to the limited statistics of these system level simulations, further validation of the results are needed. In addition, further optimizations are possible so that the final performance might differ slightly from the preliminary one presented here.

Scheduler Performance. The SINR values are measured at the output of the cooperation area scheduler for all resource blocks and all UEs. Then, by using the SINR to throughput mapping from [KPK+09] as described in ▇▇▇▇▇▇▇▇ ▇▇-▇, summing up all contributions and dividing the troughput by the number of cells, the average cell throughput without overhead is obtained. To obtain the resulting spectral efficiency (SE) one has to multiply with (1-0.43)=0.57, which includes the 17 This is an important difference to many known schedulers that use very restrictive selection criteria for CoMP UEs, based on small threshold values. overhead for synchronization channels, guard intervals and bands, reference signals and control channels as known from LTE Release 8. For the example according to results from ▇▇▇▇▇▇▇▇ ▇▇-▇, the attained spectral efficiency was:   ▇ ▇▇▇▇ ▇▇▇▇▇▇▇▇▇▇▇▇ SE = 0.57 ∑∑ K NPRE Bits(SINR(PRBi ,UEk  )) = 6.5 bits/s/Hz/cell. This corresponds to a brutto data rate TPbrutto (excluding all overhead) of the transmission system of TPbrutto = 6.5/0.57 = 11.4 bits/s/Hz/cell. Compared to the values reported in [3GPPTR1], this would be an improvement of a factor of 2.1 over Rel10 MU MIMO and of 1.6 over the best JT CoMP results in [3GPPTR1] so far. However, it should be noted that due to the limited statistics of these system level simulations, further validation of the results are needed. In addition, further optimizations are possible so that the final performance might differ slightly from the preliminary one presented here.