when is "channel busy" detected in FSM_Cs?
-
10 April 2012 3:52
In 802.11a, the preamble constists of 10 short traning sequences, 2 long traning sequences and a guard interval for carrier, timing synchroization and channel estimation, which lasting 16 us in IEEE 802.11a protocol.
My question is: Does Sora FSM_Cs know that channel is busy until preamble has been completed? that means Sora MAC konw the channel is busy maybe behind 16 us when antenna recieves this electromagnetic signal, am I right?
If the real situation does not like this, how does it work in Sora?
Many Thanks
Semua Balasan
-
13 April 2012 0:26
I'm looking forward to comment from you!!!!!
many thanks
-
16 April 2012 3:33Pemilik
Hi,
No. The carrier sensing returns when the autocorrelation is higher than a threshold. So it will consume some samples in preamble but not all of them. After that, the state-machine goes to try to sync on the preamble and get all channel estimations.
Please refer the code for the detailed number.
Thanks,
- Kun
-
16 April 2012 13:06Moderator
We don't need to detect the whole preamble to determine whether the channel is busy, energy detection is enough for this purpose.
I guess you are asking the delay from the channel becomes busy and Sora detects it. This delay can be divided into two part. The first part is caused by transferring digital samples from radio hardware to PC memory, it's arround 1~2 us. The second part is the energy detection algorithm, basically we calculate average energy from certain number of samples. This part depends on the number of samples we used to calculate an average, you can find the number from source code and also you can modify the code to control the delay.
The delay should be much less than 16us.
-Jiansong
-
16 April 2012 14:22
We don't need to detect the whole preamble to determine whether the channel is busy, energy detection is enough for this purpose.
I guess you are asking the delay from the channel becomes busy and Sora detects it. This delay can be divided into two part. The first part is caused by transferring digital samples from radio hardware to PC memory, it's arround 1~2 us. The second part is the energy detection algorithm, basically we calculate average energy from certain number of samples. This part depends on the number of samples we used to calculate an average, you can find the number from source code and also you can modify the code to control the delay.
The delay should be much less than 16us.
-Jiansong
Thanks very much, but do you have mearsured the delay exactly?- Ditandai sebagai Jawaban oleh Jiansong zhangMicrosoft Employee, Moderator 30 April 2012 3:04
-
16 April 2012 14:31
Many thanksHi,
No. The carrier sensing returns when the autocorrelation is higher than a threshold. So it will consume some samples in preamble but not all of them. After that, the state-machine goes to try to sync on the preamble and get all channel estimations.
Please refer the code for the detailed number.
Thanks,
- Kun
- Ditandai sebagai Jawaban oleh Jiansong zhangMicrosoft Employee, Moderator 30 April 2012 3:04
