General Assumptions

  • Random access preambles are not divided into set-A and set-B preambles. Currently, all of the preambles are assumed to belong to set-A.
  • The modulation curve models used by the LTE physical layer represent SNR versus block error rate (BLER). The block size assumed by those models is one LTE allocation block (i.e., two LTE resource blocks). The drop probability of a packet that is transmitted over multiple allocation blocks is computed by combining the individual block error rate of the blocks composing that transmission.
  • The MIMO antenna diversity implementation assumes no antenna correlation; therefore there will be no difference in performance between transmit diversity with two antennas and four antennas.
  • The length of sequence numbers in RLC AM PDUs is always 10 bits, but specifications provide an option between 5-bit and 10-bit sequence numbers for RLC UM PDUs. The model ignores this option and assumes that sequence numbers in RLC PDU headers are always 10 bits whether it is AM or UM.
  • While evaluating GBR-type bearers, admission control does not take RLC and MAC overheads into account. It assumes that the requested uplink and downlink GBR values for a bearer already account for those overheads.
  • The MPDU processing delay and the ACK/NACK processing delay for both the uplink and downlink is assumed to be exactly three subframes each (adjusted for the propagation delays). For FDD, this translates to the HARQ round trip being exactly 8 ms, and to sustain uninterrupted communication, exactly eight HARQ processes are defined for both uplink and downlink. For TDD, the number of HARQ processes is given in Table 147. These are derived from the same requirement, in that uninterrupted communication with the UE is desired. The HARQ round trip time is not a constant for TDD neither in uplink or downlink (except in some special cases) and is rather a function of the current subframe.
  • Handover of a UE between eNodeBs supporting FDD and TDD is allowed.
  • Only DCI format 1A or DCI format 2A is used for downlink scheduling assignments, which are assumed to be contiguous. DCI format 2A is assumed to be used only for spatial multiplexing transmissions.
  • The HARQ blanking mechanism is assumed to be used for multi-codeword spatial multiplexing transmissions. In the HARQ blanking mechanism, new transmissions will not begin until the HARQ process for both of the codewords are completed.
  • Interference on data channels is produced by other data channels and control channels
  • Interference effects for control channels are based on a probability distribution function
  • PUSCH and PDSCH do not interfere with each other. LTE designates non-overlapping uplink and downlink frequency band ranges for each given geographical region. Therefore the model will not account for interference between uplink and downlink channels regardless of their frequency configuration.
  • In a given LTE PHY profile, the guard bands for uplink and downlink channels are indicated in the table below. The same guard band value is applied at the beginning and at the end of the channel.

Table 149: Guard Band Parameters

Channel BW (MHz) 1.4 3.0 5.0 10.0 15.0 20.0
Guard Band (KHz) 160.0 150.0 250.0 500.0 750.0 1000.0