@ea4gpz profile picture
Daniel Estévez @ea4gpz

Everything space & RF. Amateur radio operator (EA4GPZ / M0HXM). PhD in Mathematics from @Matematicas_UAM. he/him

Joined October 2016
  • Tweets 8,935
  • Following 399
  • Followers 9,955
1,514 Photos and videos
New blog post: PLL coefficients for rate-only feedback. A couple years ago I wrote a post that extends a paper of Stephens and Thomas where they model PLLs in discrete time rather than transforming a continuous-time design.
1
1
18
1,384
more replies
Explicit formulas can be obtained by applying a Padé approximant to this degree 6 equation. This leads to designs that have a very small error in the loop bandwidth and have the correct damping.
A plot that shows the loop bandwidth relative error in terms of the normalized loop bandwidth. The error tends to zero when the loop bandwidth tends to zero, and it is already around 0.01% for a loop bandwidth of 0.1.

ALT A plot that shows the loop bandwidth relative error in terms of the normalized loop bandwidth. The error tends to zero when the loop bandwidth tends to zero, and it is already around 0.01% for a loop bandwidth of 0.1.

1
5
448
Read more: destevez.net/2026/06/pll-coe…
3
231
Daniel Estévez retweeted
AMSAT-DL
@amsatdl
Jun 10
Tianwen-2 at #Asteroid (469219) Kamoʻoalewa now transmitting at High Data Rate!! Receiving with the 20m @SternwarteBO dish on X-Band.
3
15
110
4,026
Daniel Estévez retweeted
Dwingeloo Telescoop@radiotelescoop
Jun 8
We think a Tianwen-2 manoever happened on 7 June, as expected. With the Bochum telescope (@amsatdl) and the Dwingeloo telescope, we observe that a) Tianwen-2 is close to the asteroid on the sky and b) the line-of-sight velocity change (Doppler) now almost matches the asteroid's.
Sky plot showing the location where the Dwingeloo telescope measured the Tianwen-2 spacecraft, with a 0.1 degree field of view. On 7 June, the spacecraft is apparently within 0.1 degree of the asteroid.

ALT Sky plot showing the location where the Dwingeloo telescope measured the Tianwen-2 spacecraft, with a 0.1 degree field of view. On 7 June, the spacecraft is apparently within 0.1 degree of the asteroid.

Frequency residual w.r.t. Kamo'oalewa, with an estimated base freq of 8428.201 MHz. On 4-6 June the residual is between 13800 and 14300 Hz, increasing. From 7 June, the line is almost flat, changing only a few Hz in hours. This shows the change in line-of-sight velocity is now close to that of Kamo'oalewa.

ALT Frequency residual w.r.t. Kamo'oalewa, with an estimated base freq of 8428.201 MHz. On 4-6 June the residual is between 13800 and 14300 Hz, increasing. From 7 June, the line is almost flat, changing only a few Hz in hours. This shows the change in line-of-sight velocity is now close to that of Kamo'oalewa.

3
22
90
5,815
Daniel Estévez retweeted
AMSAT-DL
@amsatdl
Jun 7
Just a quick update on Tianwen-2. We saw several events before LOS @SternwarteBO : At 23:06:10 UTC the data rate was reduced, probably in preparation for some trajectory maneuvers. At 23:37:00 UTC the signal suddenly faded away. Before it completely disappeared, some sudden frequency jumps in the doppler were noticeable. More later...
AMSAT-DL
@amsatdl
Jun 2
Tianwen-2 is a Chinese mission that will return samples from the Earth quasi-satellite #Asteroid 469219 Kamoʻoalewa. Today the angular distance of TW-2 from AST469219 is 0.295°. This is reducing at a rate of -0.0559°/day. At that rate the angular spacing will reach zero on Jun 07 03:20 utc, when TW-2 is planned to perform its orbital insertion at Kamoʻoalewa on 7 June 2026. 📡〰️〰️@SternwarteBO
5
13
77
13,037
Daniel Estévez retweeted
AMSAT-DL
@amsatdl
Jun 2
Tianwen-2 is a Chinese mission that will return samples from the Earth quasi-satellite #Asteroid 469219 Kamoʻoalewa. Today the angular distance of TW-2 from AST469219 is 0.295°. This is reducing at a rate of -0.0559°/day. At that rate the angular spacing will reach zero on Jun 07 03:20 utc, when TW-2 is planned to perform its orbital insertion at Kamoʻoalewa on 7 June 2026. 📡〰️〰️@SternwarteBO
2
14
47
16,086
Daniel Estévez retweeted
pi9rd@PI9CAM
May 31
Tracking #tianwen2 for a few days, we see that the Chinese spacecraft is clearly getting closer to the asteroid Kamo'oalewa. According to online rumours (see x.com/AJ_FI/status/191065037… from @AJ_FI), it would arrive 7 June 2026. That's consistent with our observations so far.
Dwingeloo observations of Tianwen-2. A star plot (DSS background) with a blue line showing the coordinates of Kamo'oalewa over time and red dots showing the Dwingeloo radio telescope localizations of the spacecraft. A fit through the localizations shows that the spacecraft is getting closer to the asteroid.

ALT Dwingeloo observations of Tianwen-2. A star plot (DSS background) with a blue line showing the coordinates of Kamo'oalewa over time and red dots showing the Dwingeloo radio telescope localizations of the spacecraft. A fit through the localizations shows that the spacecraft is getting closer to the asteroid.

Andrew Jones@AJ_FI
11 Apr 2025
Tianwen-2 unverified mission timeline: 🚀 Launch: May 29, 2025 🛰️ Arrival at asteroid Kamoʻoalewa: July 4, 2026 👋 Departure: April 24, 2027 🌏 Reentry capsule landing: Nov. 29, 2027 ☄️Arrival at comet 311P: Jan. 24, 2035 Source: m.weibo.cn/detail/5154207293…
12
37
36,601
New blog post: An update about Tianwen-2 telemetry. This is a quick update about Tianwen-2 telemetry decoding. First, I've figured out the format of the timestamps included in AOS frames. I explain what was confusing me yesterday.
A plot of space packet APIDs received over time by Bochum

ALT A plot of space packet APIDs received over time by Bochum

2
6
18
1,442
Second, @amsatdl has decoded around 12 hours of telemetry. I run all this telemetry through the same analysis, showing that there are no relevant differences compared to the telemetry received by @PI9CAM. Read more: destevez.net/2026/05/an-upda…
2
12
953
New blog post: Decoding Tianwen-2. I analyze the telemetry decoded from recent @PI9CAM recordings of Tianwen-2. Compared to Tianwen-1, there is not much in the telemetry. There are not that many fields and most are mainly static.
A post of a GNU Radio flowgraph GUI demodulating the signal from Tianwen-2, which has rather high SNR.

ALT A post of a GNU Radio flowgraph GUI demodulating the signal from Tianwen-2, which has rather high SNR.

A raster map of one of the decoded Space Packet APIDs. We can see a few fields changing values, but most of the contents are fairly static.

ALT A raster map of one of the decoded Space Packet APIDs. We can see a few fields changing values, but most of the contents are fairly static.

2
5
37
2,052
Read more: destevez.net/2026/05/decodin…

1
5
12
1,093
Daniel Estévez retweeted
AMSAT-DL
@amsatdl
May 26
Tianwen-2 is a Chinese asteroid sampling mission. Asteroid 469219 Kamo'oalewa Orbit insertion is planned for June 7, 2026. 📡〰️Received with our 20m @SternwarteBO #Tianwen2 #Asteroid
pi9rd@PI9CAM
May 25
With the 25m Dwingeloo radio telescope, we just tracked #Tianwen-2 on X-band. It was at an offset of 1.1 degrees from asteroid Kamo'oalewa, which it is going to visit. We recorded IQ data, maybe others can decode this (@ea4gpz, @r2x0t, @aang254?). Data at data.camras.nl/satellites/ra….
Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.

ALT Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.

4
31
90
6,793
Daniel Estévez retweeted
Aang254 (🦋 aang23.satdump.org)@aang254
May 26
CCSDS 8192-bits CADUs, RSS 223/255 I=4
pi9rd@PI9CAM
May 25
With the 25m Dwingeloo radio telescope, we just tracked #Tianwen-2 on X-band. It was at an offset of 1.1 degrees from asteroid Kamo'oalewa, which it is going to visit. We recorded IQ data, maybe others can decode this (@ea4gpz, @r2x0t, @aang254?). Data at data.camras.nl/satellites/ra….
Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.

ALT Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.

2
6
45
2,541
Daniel Estévez retweeted
pi9rd@PI9CAM
May 25
With the 25m Dwingeloo radio telescope, we just tracked #Tianwen-2 on X-band. It was at an offset of 1.1 degrees from asteroid Kamo'oalewa, which it is going to visit. We recorded IQ data, maybe others can decode this (@ea4gpz, @r2x0t, @aang254?). Data at data.camras.nl/satellites/ra….
Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.

ALT Screenshot showing a spectrum of Tianwen-2. Left shows SatDump with a clean constellation, using the Tianwen-1 decoder. Top right shows a graph of power vs time, which was used to find the right pointing. Bottom right shows another spectrum.
5
24
10,156
More fun with the Ryzen AI NPU. I'm working on a more complex algorithm, and I'm seeing memory stalls that make me lose one cycle in each iteration. It turns out that the data memory of compute tiles is organized as 8x 256-bit wide 8 KiB memory banks.
A trace shown in the Perfetto UI that shows that the processor performs 8 cycles of vector instructions and then stalls waiting for memory for one cycle.

ALT A trace shown in the Perfetto UI that shows that the processor performs 8 cycles of vector instructions and then stalls waiting for memory for one cycle.

A figure from the AIE-MLv2 documentation that shows the memory module organized as 8 banks.

ALT A figure from the AIE-MLv2 documentation that shows the memory module organized as 8 banks.

2
13
2,019
Each bank supports one simultaneous access, and every two banks are interleaved to form a 512-bit wide 16 KiB virtual bank. The two load units of the processor can perform simultaneous 512-bit loads only if these target different 16 KiB banks.
1
2
654
This is somewhat hidden to the programmer because the linker script tries to place the buffers in different banks when possible. However, it's relevant when there are more buffers than banks or when trying to access multiple elements from the same buffer, which is what I ran into
1
2
558
New blog post: Getting peak TOPS on a Ryzen AI 7 350 NPU. This is an introduction to low-level programming on AMD NPUs using mlir-aie. I build an example that demonstrates 56 TOPS, very close to the max theoretical performance. These NPUs are identical to Xilinx AIE-MLv2 engines.
1
3
22
1,440
more replies
Finally, I show how to use tracing to measure the performance of the NPU workload execution, and check that it matches the understanding we had obtained by analyzing the assembly code.
A screenshot of the Perfetto UI that shows an execution trace. We measure how many cycles it takes to execute each C kernel call.

ALT A screenshot of the Perfetto UI that shows an execution trace. We measure how many cycles it takes to execute each C kernel call.

1
2
394
This post is an ideal self-contained introduction if you want to learn how NPUs work from a low-level perspective. Read more: destevez.net/2026/05/getting…
6
352
Load more