[Hackrf-dev] The whole 20MHz LTE signal is decoded by HACKRF 19.2Msps!

[Jiao Xianjun]

HI,

In previous work, I just got raw bits from PDSCH. Now, ASN1 decoder is
built based on lameditor and asn1c. That means we have almost full stack
from A/D samples to PDSCH output and ASN1 parser for LTE Downlink RRC
messages.

A article on "How to decode LTE RRC/SIB/ASN1 messages by lameditor and
asn1c" is written: http://sdr-x.github.io/LTE-SIB-decoding-by-asn1c/

Have fun.

(I plan to move this Matlab-C mixed program to pure C, and accelerate it by
fixed-point/parallel to achieve real-time processing LTE Downlink signal.
Anyone who is interested in this project, feel free to join!)

BR

Jiao Xianjun



On Tue, May 27, 2014 at 9:13 AM, Jiao Xianjun <putaoshu at gmail.com> wrote:

> Hi,
>
> I have decode the whole 20MHz LTE signal by HACKRF board with 19.2Msps
> sampling rate! See all Matlab codes and captured signal here:
> https://github.com/JiaoXianjun/LTE-Cell-Scanner     (Thanks again to
> hackrf.net for lending me a board to do experiment!)
>
> Maximum sampling rate of HACKRF is about 20Msps, but 20MHz LTE needs
> 30.72Msps rate according to LTE spec. A sampling rate converter is designed
> to get 30.72Msps signal from 19.2Msps HACKRF captured signal. (19.2*8/5 =
> 30.72)
>
> By this way, those channels in the whole 20MHz LTE bandwidth are detected,
> such as PBCH, PCFICH, PHICH, PDCCH, PDSCH(SIB1, SIB2, etc).
>
> We are lucky for two reasons:
> 1. Most 20MHz LTE signals here in China are 100RB configuration. That
> means actual bandwidth is 100*180kHz = 18MHz
> 2. USB transmission between HACKRF board and computer supports 19.2Msps
> sampling rate without packets loss. Higher rate (for example 23Msps) may
> lead to packets loss.
>
> Some figures from Matlab are attached.
>
> Figure 1 info: time-frequency resource grids of a TDD radio frame (Y:
> sub-carrier; X OFDM symbol)
>
> Matlab outputs and these match the Figure 1:
>
> TDD SFN-310 ULDL-2-|D|S|U|D|D|D|S|U|D|D| CID-30 nPort-1 CP-normal
> PHICH-DUR-normal-RES-1
>
> SF5 PHICH1 PDCCH3 RNTI: SI-RNTI SI-RNTI
> No.0 4CCE: Localized VRB from RB0 to RB6 MCS-2 HARQ-0 NEWind-0 RV-1 TPC-1
> DAI-0
> No.0 8CCE: Localized VRB from RB0 to RB6 MCS-2 HARQ-0 NEWind-0 RV-1 TPC-1
> DAI-0
>
>
> Figure 2 : Constellation of SIB1 in PDSCH of subframe 5 in above radio
> frame
>
>
> Figure 3 info: time-frequency resource grids of another TDD radio frame
> (Y: sub-carrier; X OFDM symbol)
>
> Matlab outputs and these match the Figure 3:
>
> TDD SFN-900 ULDL-2-|D|S|U|D|D|D|S|U|D|D| CID-216 nPort-2 CP-normal
> PHICH-DUR-normal-RES-1
> SF4 PHICH1 PDCCH1 RNTI: SI-RNTI SI-RNTI
> No.0 4CCE: Localized VRB from RB0 to RB5 MCS-2 HARQ-0 NEWind-0 RV-0 TPC-1
> DAI-0
> No.0 8CCE: Localized VRB from RB0 to RB5 MCS-2 HARQ-0 NEWind-0 RV-0 TPC-1
> DAI-0
> SF5 PHICH1 PDCCH1 RNTI: SI-RNTI SI-RNTI
> No.0 4CCE: Localized VRB from RB0 to RB5 MCS-2 HARQ-0 NEWind-0 RV-3 TPC-1
> DAI-0
> No.0 8CCE: Localized VRB from RB0 to RB5 MCS-2 HARQ-0 NEWind-0 RV-3 TPC-1
> DAI-0
>
> Videos:
>
> in China:
> http://v.youku.com/v_show/id_XNzE3NDYwMDgw.html
>
> outside China:
> http://youtu.be/2JH_EGdHyYE
>
> Next step: decoding PDSCH and porting all these into C/C++ in
> LTE-Cell-Scanner
>
> BR
> Jiao Xianjun
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://nine.pairlist.net/pipermail/hackrf-dev/attachments/20140707/e210332f/attachment.html>