Guard Interval
The CP" time to the inverse FFT time "T(IFFT)." The guard interval is used to eliminate inter-symbol and inter-carrier interference. A copy of the last guard interval T(GI) of the useful symbol period "T(IFFT)", termed Cyclic Prefix "CP", is used to collect multipath, while maintaining the orthogonality of the subcarriers. Each symbol is transmitted for a slightly longer time, extended symbol time T(s), than the active (or useful) symbol time T(IFFT). The extra time is the guard interval.
is the ratio of the Cyclic Prefix "- 1/8: Sets the to 1/8 (see Guard Interval Time Calculation below)
- 1/4: Sets the to 1/4 (see Guard Interval Time Calculation below)
- Other: Enables you to enter values between 0 to 1.
The (GI) is specified as a fraction (percentage) of the inverse FFT time period T(IFFT). For 802.11a, the only selection is a of 1/4 (1/8 is greyed). For HIPERLAN/2, both 1/4 and 1/8 are selections. The selection allows the input of a non-standard value between 0 and 1.
time period TGuard Interval (TGI) Time Calculation
The following table shows calculated OFDM signal:
values for a HIPERLAN/2
| Guard Interval | T(FFT) | T(GI) |
|
---|---|---|---|---|
| 1/4 | 3.2 µs | 0.8 µs |
|
| 1/8 | 3.2 µs | 0.4 µs |
|
| where: |
可以看到802.11A/G/N的GI可以分为1/4或者1/8;一个OFDM符号的时间TS=T(IFFT)+T(GI) ,所以有3.6us(短保护间隔,1/8),4us(长保护间隔,1/4)。下图Data rate中的800ns GI或者400ns GI值对应上面的T(GI).
MCS index | Spatial streams | Modulation type | Coding rate | Data rate (Mbit/s) | |||
---|---|---|---|---|---|---|---|
20 MHz channel | 40 MHz channel | ||||||
800 ns GI | 400 ns GI | 800 ns GI | 400 ns GI | ||||
0 | 1 | BPSK | 1/2 | 6.5 | 7.2 | 13.5 | 15 |
1 | 1 | QPSK | 1/2 | 13 | 14.4 | 27 | 30 |
2 | 1 | QPSK | 3/4 | 19.5 | 21.7 | 40.5 | 45 |
3 | 1 | 16-QAM | 1/2 | 26 | 28.9 | 54 | 60 |
4 | 1 | 16-QAM | 3/4 | 39 | 43.3 | 81 | 90 |
5 | 1 | 64-QAM | 2/3 | 52 | 57.8 | 108 | 120 |
6 | 1 | 64-QAM | 3/4 | 58.5 | 65 | 121.5 | 135 |
7 | 1 | 64-QAM | 5/6 | 65 | 72.2 | 135 | 150 |
8 | 2 | BPSK | 1/2 | 13 | 14.4 | 27 | 30 |
9 | 2 | QPSK | 1/2 | 26 | 28.9 | 54 | 60 |
10 | 2 | QPSK | 3/4 | 39 | 43.3 | 81 | 90 |
11 | 2 | 16-QAM | 1/2 | 52 | 57.8 | 108 | 120 |
12 | 2 | 16-QAM | 3/4 | 78 | 86.7 | 162 | 180 |
13 | 2 | 64-QAM | 2/3 | 104 | 115.6 | 216 | 240 |
14 | 2 | 64-QAM | 3/4 | 117 | 130 | 243 | 270 |
15 | 2 | 64-QAM | 5/6 | 130 | 144.4 | 270 | 300 |
16 | 3 | BPSK | 1/2 | 19.5 | 21.7 | 40.5 | 45 |
17 | 3 | QPSK | 1/2 | 39 | 43.3 | 81 | 90 |
18 | 3 | QPSK | 3/4 | 58.5 | 65 | 121.5 | 135 |
19 | 3 | 16-QAM | 1/2 | 78 | 86.7 | 162 | 180 |
20 | 3 | 16-QAM | 3/4 | 117 | 130 | 243 | 270 |
21 | 3 | 64-QAM | 2/3 | 156 | 173.3 | 324 | 360 |
22 | 3 | 64-QAM | 3/4 | 175.5 | 195 | 364.5 | 405 |
23 | 3 | 64-QAM | 5/6 | 195 | 216.7 | 405 | 450 |
24 | 4 | BPSK | 1/2 | 26 | 28.8 | 54 | 60 |
25 | 4 | QPSK | 1/2 | 52 | 57.6 | 108 | 120 |
26 | 4 | QPSK | 3/4 | 78 | 86.8 | 162 | 180 |
27 | 4 | 16-QAM | 1/2 | 104 | 115.6 | 216 | 240 |
28 | 4 | 16-QAM | 3/4 | 156 | 173.2 | 324 | 360 |
29 | 4 | 64-QAM | 2/3 | 208 | 231.2 | 432 | 480 |
30 | 4 | 64-QAM | 3/4 | 234 | 260 | 486 | 540 |
31 | 4 | 64-QAM | 5/6 | 260 | 288.8 | 540 | 600 |
32 | 1 | BPSK | 1/2 | N/A | N/A | 6.5 | 7.2 |
物理层 | 带宽(数据子载波频率的个数) | × | 空间流个数 | × | 每个子载波的数据比特 | ÷ | 每个正交频分多路复用符号的时间 | = | 物理层数据速率(bps) |
11n或11ac | 56(20MHz) | 1到4个 | 最多5/6×log2(64)=5 | 3.6us(短保护间隔) 4us(长保护间隔) |
| ||||
108(40MHz) |
| ||||||||
11ac | 234(80MHz) | 5到8个 | 最多5/6×log2(256)≈6.67 |
| |||||
2×234(160MHz) |
|
11N有效子载波数
HT20M子载波数量 56 ,其中52 个用于传输,4个用于引导帧
HT40M子载波数量 112 ,其中108 个用于传输,6个用于引导帧,64-QAM编码率5/6,数据比特率64=2^6,所以是6bit。GI时间是上面的TS时间,3.6/4us
所以11n的速度=(108*4*(5/6)*6)/(3.6) = 600 Mbp/s