====== AR9130/AR9160 radar detection ======

===== Overview =====

The AR9130/AR9160 radar reports include much more information than the previous generation.

===== PHY error frame format =====

The PHY error length is now (at least) three bytes, with the following information (for a frame 'n' bytes in length):

  * optional FFT payloads, if the pulse is long enough
  * byte n-3: Pulse length in the primary channel
  * byte n-2: Pulse length in the extension channel
  * byte n-1: flags

Flags contains:

  * 0x01: Pulse in primary channel
  * 0x02: Pulse in extension channel
  * 0x04: "Early extension event" found (TBD: more explanation)

The stronger of the RSSIs in the primary and extension channel should be used.

===== FFT information =====

An optional FFT report is available for frames which meet a certain set of parameters. This is primarily designed to assist in chirp detection, where a radar pulse moves in frequency over time. These pulses are normally detected as long pulses (45-90 uS).

Each FFT sample corresponds to a 3.2uS window (40MHz/80MHz clock) or 2.9uS window (44MHz/88MHz clock).

HT20:

| Byte | Bit field | Description |
| 0 | [3:0]: | bin -28  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin -27  {bitmap, quantized magnitude[2:0]} |
| 1 | [3:0]: | bin -26  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin -25  {bitmap, quantized magnitude[2:0]} |
| 2 - 25 |  | ... |
| 26 | [3:0]: | bin  24  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin  25  {bitmap, quantized magnitude[2:0]} |
| 27 | [3:0]: | bin  26  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin  27  {bitmap, quantized magnitude[2:0]} |
| 28 | [7:0]: | all bins {max_magnitude[1:0], bitmap_weight[5:0]} |
| 29 | [7:0]: | all bins  max_magnitude[9:2] |
| 30 | [7:0]: | all bins {max_index[5:0], max_magnitude[11:10]} |

Dynamic HT40 mode (regardless of whether the received data frames are HT20 or HT40; this is the mode the PHY is in):

| Byte | Bitfield | Description |
| 0 | [3:0]: | bin -64  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin -63  {bitmap, quantized magnitude[2:0]} |
| 1 | [3:0]: | bin -62  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin -61  {bitmap, quantized magnitude[2:0]} |
| 2 - 61 |  | … |
| 62 | [3:0]: | bin  60  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin  61 {bitmap, quantized magnitude[2:0]} |
| 63 | [3:0]: | bin  62  {bitmap, quantized magnitude[2:0]} |
|  | [7:4]: | bin  63  {bitmap, quantized magnitude[2:0]} |
| 64 | [7:0]: | lower bins {max_magnitude[1:0], bitmap_weight[5:0]} |
| 65 | [7:0]: | lower bins  max_magnitude[9:2] |
| 66 | [7:0]: | lower bins {max_index[5:0], max_magnitude[11:10]} |
| 67 | [7:0]: | upper bins {max_magnitude[1:0], bitmap_weight[5:0]} |
| 68 | [7:0]: | upper bins  max_magnitude[9:2] |
| 69 | [7:0]: | upper bins {max_index[5:0], max_magnitude[11:10]} |

The radar FFT analysis provides a variety of information:

  * The max_magnitude is calculated, as the value of |I| + |Q|
  * the FFT bin index of the max_magnitude field is populated in max_index
  * for each bin, if the magitude is greater than a pre-programmed threshold (**TODO**: document this register field), the bit in the bitmap field is set;
  * the quantized magnitude is then calculated for all bins, calculated as floor(bin power * 8 / max_magnitude)

===== Notes / Limitations =====

  * The FFT report data may be truncated if the incoming signal level drops before an FFT is completed.  Thus, don't trust that the FFT data will be complete.
  * There are some known BB hangs with DFS on the AR9130/AR9160. The BB hang code in the ath_hal should detect it as a DFS BB hang and reset things. (TODO: check whether the chip needs to be "fixed" by going into another operational mode and bcak again, rather than just a straight reset.)
  * There are some corruption issues, where the MAC mis-classifies the larger PHY error frames as data frames and re-aligns some portion of the payload. This only affects frames with FFT payloads and not the short PHY error frames (3 bytes.)