| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
| |
Make the FEC code just set the CRC_OK bit like the cc1111 radio does;
eliminates a bunch of weird conventions across the FEC API.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Integrate interleaving, CRC and padding within the decode/encode
functions.
Provide for ISR priorities so that the 1120 RX interrupt takes
precedence over the other interrupts or we risk losing bits.
Optimize the viterbi decoder a bit (goes from 10ms per packet to 7ms
per packet).
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
| |
Keep it parallel with ao_fec_tx.c
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
| |
Decode radio input one interleave block at a time. This overlaps the
decode computation with the packet reception, leading to lower latency
in an attempt to keep up with the transmitter.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
| |
Interrupt-per-bit, but it seems to work
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
| |
Integrate interleaving and whitening into encode and decode steps.
Add CRC checking function for receive.
Make ao_fec_test program round-trip the data and verify correctness.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
| |
I think I understand how it works now. It's not exactly speedy, and it
uses a lot of memory.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
| |
Add CRC, whitening, FEC and interleaving routines for transmission
path to allow cc1120 to send telem packets to cc1111.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
| |
To be used with the MS5607 which generates data in calibrated units.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
This prevents any apogee detection from occurring until the specified
number of seconds after boost. This also prevents the switch from
accel+baro to baro only mode in the Kalman filter.
The test frame work is also changed to look for Apogee lockout: in the
eeprom input file.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
| |
The flight test scripts were using invalid bash syntax. Clean that up
and also switch the default flight directory to my new
~/misc/rockets/flights directory.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
| |
Support multiple serial ports more cleanly
Split out parts of ao.h into separate feature header files
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
v1.0 boards have noisy accelerometer data caused by interactions
between RF transmission and the accelerometer measurements; this noise
generates a negative bias in the accelerometer readings. The net
effect is that the estimated speed is lower than the actual speed,
causing early an apogee estimate.
By increasing the sigma value for accelerometer data, the kalman
filter 'trusts' the acceleration data less, putting more weight on the
barometer data. This causes the estimated time of apogee to be closer
to the correct value.
This reduces the response to changes in acceleration.
This new value is applied solely to TeleMetrum v1.0 boards. v1.1
boards correct for this error, and hence can use the correct sigma
value for the accelerometer.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These are significantly smaller than the general pointer versions from
libc on the cc1111.
Signed-off-by: Keith Packard <keithp@keithp.com>
Conflicts:
src/Makefile.proto
src/cc1111/ao_adc.c
src/cc1111/ao_packet_master.c
src/core/ao.h
Fix up the new makefiles
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Split out sources into separate directories:
core: architecture and product independent bits
cc1111: cc1111-specific code
drivers: architecture independent drivers
product: product-specific sources and Makefile fragments
util: scripts for building stuff
This should have no effect on the built products, but testing is encouraged
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
| |
Need to rebuild ao_flight_test when any of the flight sources change
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| | |
|
| |
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
| |
This automates flight code testing by reporting mis-detected apogee or
main events.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This changes the full telemetry stream to include kalman data instead
of the old ad-hoc flight data. It's compatible in that the packet
sizes are the same so teledongle can receive either and figure out
which it has received.
A few plotting and testing tools are added to make validating the new
code easier.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
|
| |
This runs a long list of flights (there's a user-specific path
pointing at the flights) and squawks if the baro and dual flight
computers don't match.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| | |
|
| |
|
|
|
|
| |
This builds the flight code in baro-only mode for testing.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| |
|
|
|
|
|
| |
Yes, it would be nice to automate dependency generation here, but I
can't be bothered.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
| | |
|
|
|
This creates per-product subdirectories and recompiles everything for
each product, allowing per-product compile-time changes for things
like peripheral pin assignments and attached serial devices.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
