close
Provided by NCTUns
user: janice
The NCTUns
simulator adopted a simpler but fairer scheduling design whereby during each
bandwidth request (three-way handshake), the node requests for a fixed number of
mini-slots. The default setting is 20 mini-slots and is specified in the source
file ( ../wimax/mac/mesh/sch/data/scheduler.h, the parameter:
REQUEST_SIZE).
Therefore, you can
easily calculate the utilized channel bandwidth based on the 1) requested
mini-slots size, 2) modulation scheme and 3) the symbol time. Herewith, I
present a simple calculation for your reference:
Default Settings
in the simulator are the followings:
1) Request_size =
20 mini-slots
2) Modulation
scheme = QAM 64, coded at 3/4 coding rate
3) Total
mini-slots per frame = 221 mini-slots
4) the Symbol time
= 13.8875 us
5) Number of coded
bits per symbol time = 1152 (../wimax/phy/channel_coding.cc, the parameter:
Ncbps)
For QAM 64 , the
raw channel capacity is calculated as :
Raw channel capacity = 1152
bits/symbol time / 13.8875us = 82.952 Mbps
For QAM 64 encoded
at 3/4, the raw channel capacity = 82.952 Mbps * 3/4 = 62.214 Mbps.
For QAM 64 encoded
at 3/4, 20 mini-slots will utilize a total raw channel capacity of
= (62.241
Mbps/ 221) * 20 = 5.63 Mbps.
The calculated
5.63 Mbps includes the overhead of both the physical layer and MAC layer. As
such, the data throughput will be definitely slightly lower than the raw channel
capacity.
In order to
increase the overall data throughput, you can modify the relevant network
parameters as mentioned above.
全站熱搜
留言列表