Identified and resolved issue where a 2-stream polyphase synthesizer was being updated with an incorrectly designed synthesis filter which was causing massive droop on the second half of the reconstructed channel. Oddly enough, this didn't seem to impact the P25 decoders, but it may have impacted other decoding.
Identified and resolved issue with concurrent modification of the modules list in the processing chain associated with recent addition of the channel FFT view in the channel power tab.
Co-authored-by: Dennis Sheirer <dsheirer@github.com>
Updates HackRF native byte buffers with automatic DC removal.
Creates new half-band channel source to replace the cic-based channel source in the heterodyne channelizer. Resolves issue in RTL native buffer processing. Creates a CACHED thread pool and reduces the number of threads for the SCHEDULED thread pool to 2. Changes each of the channel source implementations to use a dedicated thread (Dispatcher class) instead of the scheduled thread pool, to reduce the amount of buffered/retained samples objects in memory. Lots of other small changes and tweaks.
Removes last of Reusable buffer code. Updates DFTProcessor to use a single thread scheduled executor. Updates Polyphase channelizer to use threads instead of an executor. Removes a number of unused classes.
Commenting out window calibration. Scalar is faster in both implementations.
Adds vector and scalar implementations for windows. Repackages window and window type classes.
WIP adds vector implementations for Airspy non-interleaved buffer iterator and calibration plugins for both interleaved and non-interleaved buffer iterators.
WIP adds vector implementations for Airspy interleaved buffer iterator.
Work on developing scalar and vector implementations of airspy sample conversion and airspy native buffer processing.
Introduces new INativeBuffer framework where tuner raw samples are retained in their native (compressed) format until when they're needed and they are converted on-demand into smaller complex sample buffers to aid garbage collection.
Removes all usage of ReusableComplexBuffers and ReusableChannelREsultsBuffers. Updates polyphase channelizer to be more efficient in output channel processing.
Updates ViterbiDecoder to change Map usage to array for memory efficiency.
Updates channel processors to used scalar implementation for gain and mixer, until the polyphase channel results can be fixed to be a multiple of the SIMD lane width.
Resolves final changes for the conversion from interleaved sample arrays to complexsamples buffers for distributing baseband samples to channels and processing chains.
Updates tuner channel sources and all decoders to receive new complex samples buffers, removing the previous reusable complex buffers as the default baseband samples buffer object.
Still to resolve is the ComplexMixer implementations. Needs validation and a calibration implementation with an update to the complex mixer factory.
Updates the DMR decoder to use optimal (scalar vs vector) implementations of filter and AGC controls.
Updates the AM demodulator for separated I&Q float sample processing.
Code review and testing changes.
updates build script for runtime configuration.
Integrates calibration manager with main sdrtrunk application. Updates NBFMDecoder to use new optimized FIR and Decimation filter options.
Updates real half-band decimation filters to select the optimal (scalar vs vector) implementation.
Adds calibration manager plugins.
Adds calibration manager and plugins to determine the optimal (scalar vs vector) implementation type to use for various classes of dsp components.
Adds vector squelching FM demodulator and cleans up the IIR filter and power squelch to use all floats.
Adds vector implementation of FM demodulator. This introduces a new interface that accepts i/q samples as two separate float arrays, versus an interleaved array, which enables more efficient processing. This new complex sample interface will be used across the codebase eventually.
Adds ComplexOscillator and a SIMD version VectorComplexOscillator. Deprecates the LowPhaseNoiseOscillator because the ComplexOscillator is about 35% faster. The vectorized version is about 43% faster when SIMD lane width is 8x.
Adds new vector complex filters and new real oscillators. Note: the RealOscillator using straight Java math sin() function is 3x slower than the existing legacy implementation. Will replace this with a complex oscillator that only uses the real part.
Updates ComplexFIRFilter to be 50% more efficient.
Adds 23 tap filter implementations. Will not add 63-tap implementations because they are slower than the default vector implementation.
Adds 15-tap customized vector half band filters.
Adds initial custom complex vector half-band decimation filters for 11-tap and 15-tap decimation filters..
Adds vector complex half band filter implementations. Note: performance of the generic 64-bit 2-lane filter is horrible and is not included.
Optimizes complex half-band filter to be 27% more efficient.
Implements vectorized real half-band filter implementations.
Implements a new FIR filter class that processes sample buffers instead of single sample at a time.
Refactored all of the calibratons to be time-duration based. Testing on windows i7 gen 3 CPU showed that it took 1000x longer for one of the calibrations to run, versus my AMD 5800x.
Changed gradle jvm args so that the windows plaf export is always applied.
Updates gradle build script to ensure incubator module is included and to cleanup javafx library management for building runtime images.
Testing tweaks and changes.
Adds logging limiter to dispatcher for buffer overflows
Adds User Preferences editor for performing Vector/SIMD calibrations and new startup dialog prompting the user to perform calibrations.
Updates calibration manager and all calibration implementations to add warm-up and test phases and to reduce the overall time it takes to calibrate a system.
Co-authored-by: Dennis Sheirer <dsheirer@github.com>
There may be two separate formats that are used for LRRP position fields where one treats the value as a two's complement field and the other as a 1-bit hemisphere indicator followed by a 31-bit positive integer. If so, we'll need to expose a user-selectable flag to toggle one format or the other.
For now, this change won't impact any Northern hemisphere users.
Co-authored-by: Dennis Sheirer <dsheirer@github.com>
* Add failing test for toCSV()
* Add missing " to timeslot value
* Add failing test for quotes and commas in details
* Separate comma and quote tests
* Escape toCSV cells using utility
* Use CSVFormatter instead of escape utils
Allows us to control exactly how we want to format a CSV-valid row.
* Never have a null cell
Instead of using a stream map, opting for an explicit handling so the logger can explode when an unexpected null shows up. Prevents the situation where things are mysteriously blank, forcing developers to handle known null scenarios.
* Don't sort backing SortedList while AliasBulkEditor is applying changes.
SortedList takes too long to sort and can cause large slowdowns and UI
freezes while many changes to Aliases are applied at once via the
AliasBulkEditor.
Updates gradle build system to address classes that are now hidden in Java 17 and no longer exported from the base modules.
Updated build system to specifically call out JavaFX OS-specific libraries and target these to each of the OS-specific build products.
Removes jcenter as a maven repo since it's now deprecated.
Co-authored-by: Dennis Sheirer <dsheirer@github.com>
* Add type parameters to TableColumns to fix compiler warnings
* Only update AliasViewByIdentifierEditor list when the editor is selected
This prevents the editor's list from being rebuilt in the background as
aliases are updated. This caused a massive slowdown when editing many
aliases with the AliasBulkEditor.
Bulk editing 1000 aliases apply time reduced from over a minute to under
5 seconds.
Fixes issue #884 (see issue for more info).
AliasItemEditor.save() modifies objects that are observed by the
TableView's underlying SortedList. Thus, the TableView's selection can
change by calling save(), which means the ListChangeListener can be
fired.
This is a problem for two reasons.
1. The alert dialogue that prompts the user to save a modified alias
will show when the user presses the "save" button, because saving
modifies the selected alias and causes the list to update.
2. setAliases() calls Alert.showAndWait(). From the JavaDoc:
"It must not be called during animation or layout processing."
This statement is likely there because it can cause issues like the one
here: TableView's layout processing (and the list updating) is paused
and the list is further modified (by calling save()). Thus, when the
processing is resumed the list is in a different and unexpected state.
Calling setAliases() from Platform.runLater() will allow the SortedList,
TableView, and any other involved objects to finish processing changes
before more modifications are made. Additionally, it allows save(),
when called by pressing the save button, to complete and set the
modifiedProperty to false, so setAliases() does not display an
additional save prompt.
Updating the service library prompted some minor changes with the way that county info objects are handled for each site.
Co-authored-by: Denny <denny@denny-desktop>
* #775 Adds frequency column to channel editor
* #775 Resolves issue with changing channel processing flag on FX event thread
* #775 Removes auto-detect channel settings for P25 Phase 2 configurations.
* #775 Playlist editor cleanup actions. Changes style of segmented buttons. Fix issue with channel set processing not on FX thread. Removes P25P2 auto-detect settings feature. Adds frequency column to channel editor.
Co-authored-by: Denny <denny@denny-desktop>
* WIP - Adds support for aliasing P25 Phase 2 tones.Playlist editor - adds editor support for alias actions
* Playlist editor implementation and integration into application is complete. Adds JavaFX window monitoring to persist window size/location across application runs.
Co-authored-by: Denny <denny@denny-desktop>
* Misc minor performance improvements
* Use more concise loop syntax
* Use more concise Collection sort syntax
* Use StringUtils pad methods
* Optimize some loops and double-map-lookups
* Optimize array instantiation
There are two styles to convert a collection to an array: either
using a pre-sized array (like c.toArray(new String[c.size()])) or
using an empty array (like c.toArray(new String[0]).
In older Java versions using pre-sized array was recommended,
as the reflection call which is necessary to create an array
of proper size was quite slow. However since late updates of
OpenJDK 6 this call was intrinsified, making the performance
of the empty array version the same and sometimes even better,
compared to the pre-sized version. Also passing pre-sized array
is dangerous for a concurrent or synchronized collection as
a data race is possible between the size and toArray call
which may result in extra nulls at the end of the array,
if the collection was concurrently shrunk during the operation.
* Cleanup string concatenation
* Use EnumMap where possible
* Use FastMath instead of Math
* Initialize complex things only once
* Reduce method size to aloow more JIT inlining
* remove map of short->float
* Move field to local variable
* Cleanup Map lookup
Co-authored-by: Rob Ekl <xwq876@zebra.com>
* Include other member variables in copy constructor
* Add break to switch cases
* Implement equals() in classes extending Comparable
If equals() is not overridden, the equals() implementation is not consistent
with the compareTo() implementation. If an object of such a class is added
to a collection such as java.util.SortedSet, this collection will violate
the contract of java.util.Set, which is defined in terms of equals().
Co-authored-by: Rob Ekl <xwq876@zebra.com>
-Separates USB timeout event processing to a dedicated thread
-Updates USB transfer buffer processor to better detect errors and auto-restart on error (max 3 times)
-Updates Tuner to add an error state and SourceManager to not select tuners with an error state
-Updates ChannelProcessingManager to auto-shutdown channels sourced from tuners set with error state
-Improves logging of USB related errors.
* #576 Updates P25 activity summaries to list values in hex and decimal. Resolves issue with P25 explicit channel uplink frequency calculations. Resolves issue with Motorola base station id messages and traffic channel messages. Resolves issue with RFSS link control message control channel frequency calculations. Resolves issue with error when using recording tuners.
* #576 Updates P25 activity summaries to list values in hex and decimal. Resolves issue with P25 explicit channel uplink frequency calculations. Resolves issue with Motorola base station id messages and traffic channel messages. Resolves issue with RFSS link control message control channel frequency calculations. Resolves issue with error when using recording tuners.
Updates radio reference library to 15.1.2 to resolve issue with logback config collision.
Updates OS build names to better reflect processor types: x86_64 (64-bit) or x86_32 (32-bit)
Resolves issue with the multi-frequency tuner source and P25/MPT1327 (#603) rotating channel frequency configurations failing to maintain lock
on the channel.
Converted the ChannelRotationMonitor from using identifier updates to
now using DecoderStateEvents so that the ChannelState can supply a
steady stream of channel state events to the monitor.
radio reference client library and adds support for storing user
credentials in user preferences.
Removes legacy apache axis SOAP and WSDL libraries calls to radio
reference web services -- not required by the new library.
- Clears stale patch groups
- Reworks the audio playback panel label update code to properly update
talkgroups and aliases in concert with the audio playback.
- Removes code that was truncating playback aliases at 32 characters.
* Resolves#502
Resolves issue with connecting to icecast http streaming server.
The issue stems from Icecast HTTP (2.4+) servers not sending a full
HTTP OK response (missing header tokens) to a connection request.
Prior to the upgrade to apache mina 2.0.19, we received a specific
exception and handled it and set the status to connected.
After the upgrade, apache mina is now sending an array index out of
bounds exception when it decodes the incomplete http response.
So, we now catch that specific exception, inspect the http response
hex bytes and set the status to connected.
* Stores minimal main application window location and size/layout settings
to allow application to be restored to the same size/location on
restart.
* Removes talkgroup swapping from the patch group manager which may have
been causing talkgroup <> patch group cross-linking on certain systems.
* Stores minimal main application window location and size/layout settings
to allow application to be restored to the same size/location on
restart.
* Adds column width monitor to all tables in application that stores the
column width settings in the user preferences so that on restarting
the application the table column widths are restored to the previous
widths.
* Resolves issue with non-aliased talkgroup values not displaying in
audio playback (left/right/mono) windows.
* Adds buildSdr build target back into gradle script.
* Updates gradle to 5.1.1
* Updated to run application from intellij using installed Java 11.
Resolves issues with Java 9 deprecations.
* Updates gradle build dependencies and other deprecation issues.
* Updates gradle to produce OS-specific builds.
* Updates the beryx runtime plugin to 1.1.1 to resolve issue with zipping
each of the target distributions.
* Import cleanup
* Resolves issue with Look and Feel error on Max OSX
Resolves issue with building OSX package
Resolves issue with excessive logging about JMBE library.
* Updates Talkgroup and Talkgroup ranges and editors to support formatted
talkgroup values.
* Updates talkgroup format masks to allow spaces (ie incomplete values)
in talkgroup values.
so that users can set the library path once and it will persist across
application upgrades so that they do NOT have to copy the library to
their application install with each upgrade.
* Initial work on add support for rotating control channels.
* Creates multiple frequency editor control.
* Updates channel source editor to allow user to select multi-frequency
configuration and add multiple frequencies.
* Adds full support for rotating control channels to the P25 channelizer.
Users define multiple frequencies in the channel source configuration.
The decoder monitors the state of the channel for a CONTROL state. If
the state does not stay in the CONTROL state for a user-specified delay
value, then the decoder rotates to the next frequency in the list. This
behavior continues until the decoder state once again changes to CONTROL.
This is NOT a 'scanner' feature.
we can account for multi-alias identifiers like patch groups and ensure
that patch group alias intentions are carried throughout the system for
recording, streaming, etc.
Updates all consumers of alias list data to ensure it can operate with
a list of aliases versus an assumed single alias.
Refactors APCO25 unit/user status identifiers from a base status class
so that alias list can support status identifiers.
* Updates playlist to fully clean aliases from the alias list when deleted
and to perform a delete and re-add on any alias that changes, to ensure
that the alias list is kept fully up to date.
Removed site id as an alias identifier.
Add support for converting ltr-net uid's to talkgroups in playlist
updater and updated playlist to version 3 to ensure anyone using current
codebase will have their playlists fully updated.
Moved MIN, UniqueID and Site ID to the legacy package and updated
label descriptions to signify that they are no longer supported.
* Refactors DirectoryPreference from PLaylistPreference.
Adds editor to PreferenceEditor for directories.
Organizes preference editor tree.
* Adds recording directory preference to editor and updates recording
manager to dynamically use this property for recordings.
* Adds event logs directory preference to editor and updates event log
manager to dynamically use this property for event logging.
* Adds application log directory preference. Updates logging to use user
preference for application log directory. Moves logging initialization
into the main application class and removes file based logging configs
from the logback.xml configuration file.
* Refactors application logging into its own class and adds prefeence
update monitoring to dynamically stop/start the logger in the new
directory location.
* Adds streaming and screen capture directories to the user preferences
editor and updates streaming and screen captures to dynamically adapt
to user changes.
* WIP on new P25 message framer.
* WIP on new P25 message framer.
* Adds new 1/2 and 3/4 rate trellis decoders. Still trying to chase
down why there is a sync pattern mismatch in the NID buffer queue.
* WIP - new p25 message framer works correctly now. PDU's are dectected
and decoded using the 1/2 and 3/4 rate trellis decoders. Packet
Sequences are assembled. Need to add a new packet sequence message
factory that can detect Motorola Auto Registration Service (ARS)
formatted UDP/ID packets and SNDCP command packets.
* WIP - removes nullify on channel results queue.
* WIP - updates message framer for TSBK 1/2/3 processing.
* WIP removing P25 sync pattern and nid bits from all message parsers.
* WIP - merges identifier framework branch changes into this branch.
* WIP on updating link control words to new format and identifier framework.
* WIP - implements rest of link control words for extended formats
* WIP - initial work on ISP/OSP TSBK message restructuring.
* WIP - adds more TSBK ISP and OSP messages
* WIP - finished updating most of the TSBK ISP and OSP messages. Converted
all messages to no longer include the NID (nac + dataunitID) so that
each message can simply hold the parsed values. Removed AliasList as
an argument to each message and updated all messges to use the new
IIdentifer and IAPCO25Channel interfaces.
Removed legacy TDU and LDU link control messages and replaced with a
single LinkControlWord class hierarchy.
P25 decoder state is not yet updated for the new message structure.
PDU message parsing is not fully implemented.
* WIP - adds some Multi-Block trunking control messages with Alternate
and Unconfirmed formats.
* WIP - Fully implements AMBTC and UMBTC PDU message parsers.
Some package restructuring.
* WIP - Resolves Link control parsing in LDU1 and 2 messages.
* WIP - finishes updating TSBK and Link Control messages. Adds support
for moto link control messages for patch groups.
* WIP - finishes updating TSBK and Link Control messages. Adds support
for moto link control messages for patch groups. Adds support for moto
ARS service, IPV4 decoding, and UDP/IP packets.
* WIP - finishes SNDCP packet data with header parsing. Compressed IP or
UDP headers are not yet supported, but are logged as such.
* WIP - converts all talkgroup identifiers to integers and updates
alias list to create a per-protocol map of talkgroup id's to alias.
Inserts a legacy talkgroup (ltr, p25, passport, mpt1327, mdc, etc.)
conversion utility to convert existing legacy (string) talkgroups to
new integer talkgroup format.
Creates an audio packet manager to insert metadata into audio packets
before streaming packets for playback, recording, and/or streaming.
Creates identifier collection to manage current identifiers in each of
the channel and decoder states.
Creates simplified channel metadata to maintain the now playing display.
Moves channel attributes (alias list, system, site, name) as configuration
identifiers in the channel state.
Many other refactoring and updates.
* WIP - numerous updates to push use of identifier collection throughout
the system. Updates the channel metadata and audio packet processing
logic. Implements a large portion of the P25 decode state processing.
Updates traffic channel management. Adds user preferences for timestamp
and identifier formatting, however doesn't add a user interface yet.
* WIP - initial implementation of user preferences editor window.
* WIP - updates preferences editor window to close and dispose window but
retain javafx event thread running.
* WIP - fully updates p25 decoder state and traffic channel processing to
use the new identifier framework.
Cleans up channel event processing and removes channel model as the sole
broadcaster of channel events. Channel processing manager now listens
for all start/stop events and requestors dispatch event directly to the
channel processing manager.
Resolves issue in airspy that was causing user interface to lock-up. The
issue appears to occur when the dongle is plugged into a usb port that
can't fully support the data rate. sdrtrunk now recognizes these errors
and warns the user that they should use another usb port.
* WIP - updates alias action manager and adds new talkgroup range alias identifier.
Updates playlist to incorporate a version number so that the playlist
updater only runs when needed, to update a playlist to the correct
version format.
Updates user preferences to use the google event bus to broadcast
preference update notifications to remove the need for listeners to
register directly on the user preferences instance.
* WIP - updates alias id editors to flag invalid values and moves legacy
alias ids to legacy package.
Updates event logging configuration to add traffic channel decode events
and decoded message logging. Removes binary message logging since that
is now replaced with demodulated bitstream recording.
* Removes system properties from user preferences. Preferences are stored
in the OS-dependent settings store (windows - registry)
* Updates MapService to use new PlottableDecodeEvent instead of monitoring
all of the decoded messages.
Updates Fleetsync decoder and related messages to identifier framework.
* Updates LTR-Net, Fleetsync and LJ1200 decoders to new identifier
framework and adds support for user preferences for formatting of
LTR-Net and fleetsync talkgroup identifiers.
Breaks LTR-Net messages into various message types and uses a new
message factory.
Removes alias list from message processing for ltr-net, fleetsync and
lj1200.
* Updates LTR-Standard messages to Identifier framework.
* Removes Plottable and IPlottable from codebase.
* wip
* WIP on MPT-1327 messages.
* Updates MPT1327 decoder to identifier framework and new traffic channel
manager. This decoder still needs to be updated with a new message
parser factory and a sync detector like what is now used for P25.
* Updates MDC1200 decoder and message to identifier framework.
* Updates Tait1200 and Passport decoders to identifier framework.
* Merges changes from master branch.
* Removes legacy alias support classes. Updates passport message processor
to only pass valid messages. Adds support to playlist updater for
converting passport mobile ids to become passport talkgroup values.
* Updates P25 decoder state and channel state to better track allocated
traffic channels. Resolves issue in timestamp manager for trackig
message timestamps. Resolves issue in airspy tuner that was causing
the tuner to stop sending samples to the application.
* Update P25 Decoder state to process SNDCP management messages and
PDU messages.
Updates P25 activity summary to display network and site details.
Resolves issue with overlay panel that prevented enabled channels from
rendering correctly when zoomed in.
* Creates a new FilePreferences as part of the user preferences to begin
the transition of SystemProperties over to user preferences and consolidate
file preferences in one location.
Updates playlist manager to use the new file preferences and to use
default.xml as the new default playlist name, following conversion of
a playlist_v2.xml format to the new format.
In the near future, the user will be able to edit file locations and to
create/use other playlists.
Resolved an issue with all decoder states where not all identifiers were
being included in decode events.
* Fix for issue #441.
Method doc reads that the method is thread-safe, but there exist no locking to ensure thread safety on call into recycle method. Modifying the decrementUserCount method to block other callers to ensure thread-safe access.
This appears to be a hot area of code and there may be a performance hit by doing this, please double check my work here.
* Moving synchronized to recycle method as it's accessed by two different methods that could potentially cause an unsafe access race condition.
* Revert - synchronizing recycle does not protect against multiple recycle calls on the same object.
* Fixing issue #353
Moved replaceIllegalCharacters method into StringUtils.
Modified classes that write out to disk to call this method when saving a file.
* Implementing suggested changes from comments in PR #448
Updated audio packet wave recorder to correctly dequeue audio packets
to ensure that the overflowable buffer queue accounting process was
working correctly.
Updated overflowable buffer queue's poll() method to correctly decrement
the queue size counter.
Resolved issue where the mutable metadata was being prematurely reset
and causing the end audio packet to be mis-identified. This resulted
in the recording manager not stopping the recording on command with the
timeout mechanism kicking in after a brief delay.
so that the modules are reset after they are shut down, in order to
prepare for processing chain reuse.
Previously, the modules were reset upon startup, but after any traffic
channel metadata had already been transferred to the channel state,
causing the metadata to be lost and allowing initial audio packets to
stream without any metadata and related streaming/monitoring rules.
only update the center frequency when needed and to place the set of
decoding channels adjacent to the right side of the tuner's center
frequency when the complete set of channels fit into the usable half
bandwidth of the tuner.
Adds 500 ms delay to initial registering for samples from all tuners to
feed the spectral display/waterfall. The existing approach of immediately
registering for samples was causing the airspy tuner to lockup
aperiodically. Investigation via debugger didn't show any application
thread locks ... usb master processor was continuing to service libusb
timeouts. Unable to debug internal airspy processes. Adding the delay
to startup seems to be a usable solution.
Reverts PLL gain values to version 0.3.4beta2 values to restore P25
decoder performance.
This change in conjuction with recent channel sample rate change should
restore P25 channel decoding to comparable levels from 0.3.4b2.
Updates decoder types to specify channel sample rate and pass/stop
frequencies for requested channels. Updates tuner channel source
managers to accept a channel specification for channels that are
allocated.
channel manager so that channel source requests can have a start
timestamp specified and the channel can be preloaded with samples from
the delay buffer. If no timestamp is specified, current time is assumed.
This feature requires corresponding updates to any decoder that wishes
to implement the timestamped sample preloading functionality.
reusable byte buffers out to the processing chain. This enables the
processing chain to broadcast these buffers back to the decoder for
message assembly and to any running binary recorder.
Updated P25Decoder to be a reusable byte buffer listener. The decoder
will now pass reusable byte buffers down to the message framer for
message processing. This should make the code more efficient since it
can process byte arrays versus handling each individual dibit.
The P25MessageFramer has been updated to be a listener of reusable byte
buffers instead of Dibits, but the actual processing of byte arrays
has to be coded.
Added max recording size to the binary recorder so that it cycles at
500 kB max file size.
buffers. None of the decoders currently support this yet, but the P25
will be updated to provide bitstream output for recording.
Renames the ReusableBuffer class to ReusableFloatBuffer for
consistency.
actively decoding, but no call or data is being transmitted. This is
primarily used for P25 channels when TDU Link Control messages are
being transmitted in between voice or data transmissions on a channel.
Updated several of the P25 decoder state message processing sections to
keep the channel state UI updated with current call identifiers and
states.
Hard-coded channel state colors to remove inefficiency in the way that
channel state coloration was being handled.
the lost connection timeout from default of 60 seconds down to 5 seconds
so that the connection state updates more quickly. Removed complex
error handling and replaced with a simpler approach to catch an error
and disconnect and then attempt a reconnect.
Added new 'Network Unavailable' error state that detects when the network
connection is unavaialable.
Updated the icecast metadata updater to close the TCP session after
successfully writing metadata to the socket.
and auto-correction for R820T tuners.
-Updated TunerModel and TunerViewPanel to add 2 new columns to
display the current tuner's PPM setting and (when available) the
measured frequency error averaged over 5 second intervals provided by
certain decoders (e.g. P25 C4FM and LSM). Frequency error value from
P25 decoders is provided from the Costas Loop PLL once 2 or more P25
message sync events have been observed in a short period.
-Updated R820T tuner to auto-correct PPM when measured frequency error
exceeds the +/- 0.4PPM threshold for a period of 30 seconds or more.
Feature can be enabled/disabled by user in the tuner editor panel.
Resolved several issues with P25 C4FM/LSM decoders that was impacting
performance related to channel misalignment from tuner PPM offsets.
Resolves an issue with the previous implementation where the tuner
editor was being refreshed repeatedly to update the lock state for the
tuner, rendering the editor almost completely unusable.
Now updates only the frequency control and the sample rate control to
reflect the tuner's lock state.
Adds capability to disable spectral display.
Adds right-click context
menu items to disable display and/or reenable display by selecting
another tuner.
Adds view menu items to disable spectral display or show a tuner.
Persists spectral display enabled state across application run sessions.
The C4FM and LSM decoders were issuing a PLL reset with each new
channel that was allocated, causing any existing decoding channels to
be disrupted. Decoding re-sync delays or inability to recapture the
signal for decoding would likely result as the tuner's overall PPM
caused misadjustment in the channel.
Resolved a NPE that was occurring in the FrequencyController for
sample rate changes.
sample rates.
Resolves issue in PolyphaseChannelManager responding to sample rate
changes in the tuner. The manager now recreates the channelizer when
needed if the tuner's sample rate has changed since the last time the
channelizer was used.
Updates the Tuner selection view to reflect tuner lock state. Updates
the tuner editors to enable/disable frequency controller and sample
rate controls according to the tuner's lock state.
Refactors a MixerReader instance from the RealMixerSource so that it
can be reused for both real and complex data outputs.
Updates RealMixerSource to use refactored MixerReader and use a default
sample rate of 8 kHz.
Updates the ComplexMixer to use the refactored MixerReader.
Updates the FCD 1/2 tuners for compatibility with the new ComplexMixer.
Removes several unused classes.
Fleetsync II decoder to use it. Adds instrumented version of the decoder
and a new Fleetsync2 FX pane to the DemodulatorViewerFX tool.
Note: the remaining AFSK-1200 based decoders will all be ported to use
this new AFSK1200 decoder (MDC1200, Tait1200, LowJack1200).
Deprecates the Listener<Boolean> interface method in the MessageFramer
to optimize for primitive booleans and reduce impact on memory. Will
remove this interface once the AFSK1200 decoders are updated.
channelizer sub-system is functionally complete.
Updated the channelizer and synthesizer filter design methods to use a
Kaiser windowed Nyquist filter. Updated the synthesizer filter design
to use a new cutoff frequency of 105% of the channel bandwidth to
resolve a slight dip in the synthesized channel gain at the channel
union region.
Updated the 2 channel output processor to use a low-pass filter that
is designed for the runtime channel sample rate versus hard-coded 12.5
khz.
Testing and redesign continues ...
filter structure. Updated the channel results array buffer to produce
reusable buffers.
This has dropped the overall memory usage considerably when profiling.
Resolved issue in the processing chain where polyphase channel sources
were not being stopped and were stacking up in memory.
that is currently up/down converting the signal of interest, but the
synthesizer is not correctly combining the original two channels.
Some experimentation with the tap count for the channelizer indicates
that we can cut the tap count in half and still have decent decoding
on the P25 decoder when paired to an aligned channelized output. The
2-channel synthesized output decodes partially, but needs more work.
Resolved issue with channel gain in the channelizer that was making the
sample values so small that they were registering as zero when converted
from float to 16-bit int value in the complex wave recorder.
samples prior to interpolation.
Resolves issue with using incorrect quadrature index when pulling a
sample from the interpolating buffer.
Resolves issue where the instrumented version of the interpolating
buffer was being constructed in the non-interpolated decoder component.
was not computationally feasible with my 3rd Gen i7 processor running
and airspy at 10 MHz. Restructured the polyphase channelizer to use
vector iteration structures that the JDK 9 JIT compiler can translate
into SIMD instructions. Separated the IFFT processing from the filter
process so that they can be executed on separate threads and this allows
the processing to occur real time.
P25 C4FM decoding at 25 kHz channel sample rate using an airspy
10MHz sample stream seems to be working nicely with ~25% CPU usage and
~300 to 500 MB memory usage. Incremental (parallel) channel decoding
should have a minimal increase in CPU loading since all of the 12.5 kHz
channels are available in the polyphase channelizer and the only
additional workload is the decoding of the 25 kHz sample stream.
There is still an issue with stopping channels and the channelizer keeps
processing, and occasionally when starting traffic channels something
deadlocks and causes out of memory.
adds Feed forward gain support to the C4FM DQPSK demodulator which seems
to improve output from the interpolating sample buffer at each symbol
period. Updates the symbol timing gain for C4FM to decrease the gain a
bit since the feed forward gain is providing a better interpolated symbol.
updates the LSM decoder to follow the same design pattern. Adds LSM
decoder support to the demodulator viewer application. The current
symbol timing gain for LSM is set too high -- is using the C4FM gain
and causes too much fluctuation in the timing error.
error calculator. After increasing the gain for symbol timing, this
new symbol evaluator class seems to track the signal very accurately
despite lots of noise and ISI.
timing error calculation with symbol decision. In the process of
changing over to this new class and remove the previous implementations
for the 3 values. This commit includes the partial implementation
so I can access from another dev machine.
based on the P25 LSM decoder using a costas loop and a gardner detector.
Updates the Demodulator Viewer FX application to add in a symbol chart
and a phase error chart. The decoder won't lock on yet ... still a wip.
ordering to each channel configuration to provide the user with greater
control over initial application startup configuration.
Adds new dialog window that displays on startup that allows the user to
invoke/cancel auto-start, or simply allow the timer to expire whereby
the channels will be auto-started.
Updates channel table model to include an auto-start column and relabels
the 'Enabled' column to read 'Playing'.
Updates channel editor to change the 'Enable/Disable' button to read
'Start/Stop'.
Changes the 'enabled' property of the channel configuration to indicate
the current 'processing' status - isProcessing().
Channel's XML configuration re-purposes the 'enabled' property to indicate
auto-start and adds a new (optional) auto-start 'order' attribute.
so that on startup, the application attempts to source the channel from
the preferred tuner.
If the tuner is currently not available, or if the tuner cannot source
the channel, then the next available tuner is selected.
when the tuner controller indicates that there is a center DC spike
blocked region or when the tuner has clipped outer spectral regions
resulting in less than 100% usable bandwidth.
produces a significant reduction in memory usage, especially for tuners
like the airspy and hackrf that produce 10/20 MSPS sample rates. This
is observed to be around 66% reduction in peak memory usage and provides
a more stable memory utilization across time reducing stress on the
garbage collector.
Use of these reusable memory buffers will be pushed further down into
the processing chain modules to ensure we keep new object creation
to the lowest levels possible.
ChannelizerViewer so that we can compare the two-channel synthesis
operations. The SynthesizerViewer is successfully down-sampling and
synthesizing two 25.0 kHz channels into a combined 25.0 kHz channel with
no noticeable drop/nulling in the joined region. However, the channelizer
viewer's synthesized channels definitely have a null in the joined
region. I suspect that the channelizer is not quite where it needs to
be and am going to start experimenting there.
and translating it by -FS/4 to correctly center the synthesized channel.
The low-pass filter is currently disabled to better troubleshoot/view the
state of the synthesized channel. It appears that the synthesizer is
not correctly joining the channel ... instead of down-sampling and then
synthesizing the two channels, it appears to be synthesizing and then
down-sampling. The synthesized channel has a null in the middle which
would render it unusable.
TestTuner and TestTunerController implementations. This implementation
only has support for channels that are completely centered within a
channelizer output. Two output slice stitching to form a composite
channel will be implemented next.
Initial profiling indicates that the FFT function within the polyphase
channelizer is now the dominant processing workload - 28% cpu loading
when channelizing a 10 MHz sample rate input into 800 channels running
at 25.0 kHz each. This is ideal for offloading to an OpenCL kernel and
running on a GPU.
complex buffer listeners. Removes complex buffer broadcast support from
the source manager and renames it to channel source manager to convey
that it's only responsible for providing channels from the tuner.
calculator must be updated immediately any time the tuner's center
frequency changes so that a channel can be allocated immediately. So,
I changed source event handling to two parts where the channel calculator
is immediately updated and the channelizer configuration is deferred so
that it can happen on the buffer processor execution thread and we can
avoid any sync() locking on the channelizer or polyphase channel
sources.
messages and events can be accurately timestamped, and more importantly,
allocated traffic channels can request to be pre-loaded with buffered
samples to ensure that traffic channels don't miss any samples that
are currently missed due processing latency inherent in current buffer
sizing. This should reduce the clipping of starting messages or voice
on the traffic channels.
Restructures processing path for tuner frequency and sample rate change
events. The tuner now embeds these events in the sequence of buffers
that are generated immediately after the event occurs since those are
the buffers that will be impacted by the change. Previously, the change
event would be broadcast throughout the system and would impact any
buffered/cached buffers. This also reduces the number of source event
change listeners that have to be managed. Note: this change is not yet
implemented in all tuners ... only in the TestTuner implementation.
Updates the polyphase channelizer to now buffer channel results and then
dispatch the buffer to the channel output processors to improve efficiency
of using float arrays versus single value complex float processing.
center frequency each time we allocate a new DDC/channel. Adds
support for broadcasting channel count change source events so that the
tuner can accurately convey sourced channel counts via tuner events.
Completes changeover of polyphase channel manager to support arbitrary
bandwidth channels. The new approach provides greater compatibility
across any tuner sample rates. Likewise, all decoders will have to be
updated to adapt to any channel sample rate, although channel bandwidths
will be at least 12.5 kHz and likely no more than 25.0 kHz, with a 2x
oversampling that produces channel sample rates in range 25.0 - 50.0
kHz.
Updates wave writer to add the metadata LIST and ID3 chunks to the end
of the recording so that the file contains the final/complete set of
metadata that was accumulated over the call lifecycle.
controllers manager sample buffer processing and channel management.
General cleanup across all modules to remove the executor from each
of the start() methods. Removed the B100 tuner class since it's not
yet implemented.
controllers manager sample buffer processing and channel management.
General cleanup across all modules to remove the executor from each
of the start() methods. Removed the B100 tuner class since it's not
yet implemented.
Updates Icecast TCP broadcaster to correctly handle content type:text/html
message that normally follows mount point in use error received from
broadcastify.
Changes mount point in use broadcast state to be a temporary error so that
the stream will attempt to reconnect after receiving this error.
Updates logging to include memory allocation/usage metrics with each
logging entry in order to support debugging issues related to high
memory usage situations. Updates initial application logging to better
identify the host system operating environment.
occured when the ChannelProcessionManager was allocating a new channel
causing the tuner to invoke the updateLOFrequency() method that was
trying to find a center tune frequency that would accommodate all of the
channels, specifically in the instance where the distance between the
highest channel and the middle channel was exactly one-half of the
tuner's bandwidth. This situation would cause the calculated adjustment
to be zero and would cause an infinite loop, locking all channels from
accessing the any channel startup/teardown events and causing running
channels to block with buffer overflows. Added 1 to the adjustment to
ensure that the initial adjustment placed the center overlapping channel
fully to the left of the center DC spike so that on the next iteration
that channel would no longer be overlapping.
Updates the build script for conditional compilation using either Java
8 or Java 9 options. Java 9 deprecated the java.xml.bind package and
removed it as a default Java SE package, so we have to reference it via
the --add-modules argument.
Updates the linux/mac launch script to check the java runtime version
and use either a Java 8 or Java 9 launch command.
state and replaces it with a processing chain heartbeat that is generated
by the sample source.
Previously, the state monitor was running on a separate thread and this
was causing multi-threading issues. With this update, all channel state
operations (and all module operations) occur on the source's buffer
processing thread.
Tweaked buffer size limits and buffer handling to attempt to avoid overflow situations in the first place.
Resolved an issue where the tuner processor was not deregistering correctly from the LIBUSB tuner processor during tuner shutdown.
detect encrypted audio state so that encrypted audio is not (incorrectly)
decoded and played back to the user.
Processes HDU and LDU2 messages to identify encrypted call state. Once
established, all subsequent LDU1/2 messages are decoded only if the call
is unencrypted. If an LDU1 message is received out-of-sequence, meaning
that the initial HDU message was missed, then the LDU1 is cached until
we receive an LDU2 message and correctly determine the call encryption
state.
A) streaming connection errors - detects errors and if currently connected, resets the connection.
B)duplicate calls being streamed - issue with comparison of broadcast channels allowed the same broadcast channel to be queued two or more times for the list of channels to stream a recording.
Issue identified with Patch Group management and circular dependencies that can develop because of stale patch groups. Refactored all patch group management into a new PatchGroupManager. Patch groups are now automatically deleted if we don't receive any update messages within 5 seconds. The PatchGroupAlias was also updated to not roll-up attributes from subordinate patched group aliases when those aliases are also patch group aliases -- this eliminates the circular dependency chaining.
Added IOverflowListener interface and updated Sources to support overflow listeners. Updated ChannelState to register as an overflow listener so that it can visually depict to the user when the channel enters/exits source overflow state. Added new color pattern for OVERFLOW with yellow text on red.
Resolves issue with MP3 framing where the remote server was disconnecting us due to bad MP3 sync. The LAME encoder was sending partial 20 byte MP3 frames initiall, then following with a 144 byte array which contained the remainder of the first frame and the header of the next frame. Adjusted the MP3SilenceGenerator to only emit complete MP3 frames.
This should be the final commit for this feature, pending merge back into master branch.
Note: Shoutcast V1 streaming to a Shoutcast V2.5.x server seems to be problematic. The newer server tracks MP3 boundaries in order to insert advertising and this appears to have problems keeping sync with the MP3 stream. I spent significant time eliminating MP3 framing errors in my code. You can see these 'Bad frame' errors in the sc_serv.log and the shoutcast server will eventually reset the connection after a period of time or after a number of framing errors occur. I anticipate that streaming to a Shoutcast V2 server using the ultravox (ie V2.x) protocol may resolve this issue.
Updated code to show dialog to user whenever the tuning limit has been exceeded. Added new InvalidFrequencyException that includes the requested invalid frequency and the valid frequency limit that was exceeded
Change code to no longer assume that 10 MHz is an available default sample rate. The mini is only capable of 6 and 3 MHz.
Note: only the 6 MHz sample rate option should be used at this time. 6 MHz can be decimated to 48 kHz using an integer decimator. sdrtrunk doesn't support the 3 MHz option, because it can't decimate that rate to 48 kHz until the new channelizer is developed.
for Hilbert and Differentiator FIR filter designers may be added in the
future.
Adds developer filter viewer tool written in JavaFX to visualize the
frequency and impulse response of FIR filters. This application is not
intended to be a user application and is specifically for the developer
to analyze filters and filter specifications that will be used in the
code.
NOTE: Java 8 JDK is now required to build this project for version 0.3.x
and forward. The JavaFX visualizer application and future user
interfaces will use the Java 8 version of JavaFX
Changes to model, view and controller components and updates all gui
editors. Removes the configuration panel and tree and replaces with 3
new panels. Updates playlist.xml to a new playlist_v2.xml format.
Removed frequency change listener framework from tuner so that the tuner
controller handles all frequency change interaction.
Updated tuner view to allow selecting tuners for main spectral display
and/or launching a new spectral display by clicking in the tuner table
cell.
configuration and tuner configuration assignment editors. Beginning of
tuner editor panel rework. Removes most of the remaining JTree
configuration pieces. Updates the settings manager to load and save
tuners and refactors all of the tuner configuration management out into
the tuner configuration model.
number of unused classes. Removed playlist manager from a number of
classes and replaced with the alias, channel and channel map models.
Updated playlist manager to no longer retain a playlist except for when
loading to models or when storing from models. Extracted version 1 to
version 2 playlist conversion and changed playlist manager to keep the
original version 1 playlist and create a new version 2 (playlist_v2.xml)
file. On startup, try to load a v2, and if not, look for a v1 to
convert. Otherwise, don't create a playlist and one will be created
once the user adds items to the models.
multi-select editor.
Changes example hyperlink to 'Help'.
Ensures editor only allows 1 instance of call priority or non-recordable
can be added to any alias.
better handle processing and shutdowns.
-Reuses transfers between tuner start/stop invocations
-Moves libusb transfer status direct byte buffer allocation out of
buffer processing thread and reuses it, saving a lot of memory
allocations over time.
the spectral display to be slow to reflect changes to tuner settings.
Changed the max size of the sample queue based on the highest rate tuner
sample rate 10 MHz (airspy) and the lowest available spectral update
frame rate (12). Sized the buffer queue based on the maximum number of
buffers that should be available per DFT processing iteration. If the
queue backs up, new buffers will be discarded until the DFT processor
consumes the queue and catches up.
shared across multiple tuners of the same type, unless the user created
a new tuner configuration for each tuner.
Updated code to create a default tuner configuration based on the tuner
type, but name the configuration as 'Default-UNIQUETUNERID', so that any
changes to that uniquely identified tuner will persist across sessions.
Note: this does not avoid the situation where multiple tuners of the
same type are all using the same default serial number (e.g. 0000001).
In this case, they will all still share the same configuraiton. User
must change the serial number of the dongle in order for it to be
uniquely identifiable, so that we can correctly lookup the configuration
for the specific tuner.
haven't yet been initialized. This occurs when there is a channel
allocated from the tuner that is the first/only listener and then is
immediately released. Some of the buffers are in motion and the others
simply haven't yet been initialized.
Potential existed for the audio output class to become unsynchronized
with the underlying source data line and attempt to perform a blocking
write to the data line when the data line doesn't have the capacity to
accept the full buffer, and the data line is not started. Thus,
playback could potentially stay in a persitent blocked state.
Refactored Mono and Stereo audio output implementations to push the
audio buffer queue processing into the parent abstract class.
AudioManager was updated to remove the last time active timestamp out
of the audio manager and down into the audio output to remove duplicate
timestamp accounting.
and constraints on scheduled thread pool saturation. Adds a new
scheduled thread pool for each processing chain that the components in
the chain can share, versus the single shared pool exposed by the
threadpoolmanager class.
the NID crc check but uses an unrecognized data unit id, it will now be
broadcast as NID:UNKN, but allows for detection of new DUID types when
we have a large number of occurances of unknonwn duid types.
saturation.
Each tunerchannelsource was sharing a thread pool and as the processing
load increased, the thread pool was not expanding and the processing
tasks were backing up, causing the buffers to grow with unprocessed
data.
Changed so that each tuner channel source is handed a thread pool to use
for processing incoming sample data. Processing chain maintains the
thread pool and starts the channel source. This allows traffic channels
to reuse the thread pool across multiple iterations.
chain has a message broadcaster built in, so the decoder's broadcaster
is redundant. Changes Decoder(s) to only have a single registered
message listener (ie processing chain)
-Removed channel collection panel since panels will only have 1 channel
now
-Resolved issue with channel list and channel selection events
-Fixed some of the classes from the rebase to master branch
channels upon loading the playlist.
Processing manager now detects channel configuration changes and
restarts the channel if it is currently processing.
Channel editor enable button now shows green if it is processing, or
default when not processing.
functionality so that the editor starts up without a channel displayed
and as each channel row is selected in the channels table, the editor is
updated with the channel configuration.
Likewise, the channel editor is now a channel event listener on the
channel model and will maintain channel state if the user changes
channel configuration in any of the external channel controllers like
the spectral display or the channel list.
-Started work on new channel editor channel viewer/controller
-Removed a number of unused buffer package classes
-Moved automatic frequency control to be a module and removed special
handling
-Added IUnfilteredRealBufferListener & Provider interfaces
-Completely removed the ResourceManager class
AudioManager connection class monitoring of the playback state. Adds a
line listener to monitor when the line is started/stopped and responds
accordingly.
Removes squelch control from mono and stereo audio playback components
which was a holdover -- squelch state is now managed within the
audio component for each processing chain.
with TO/FROM identifiers displayed, but no audio playing. This occurred
when the decoder sent a few audio buffers which caused the display to
update from the metadata, however the decoder stopped sending buffers
and the audio start threshold is never met. Clear display is invoked
upon audio stop, so the panel was never started or stopped or cleared
until the next call came in.
Changed to store the incoming metadata and only display it once the
audio playback starts and reset the display once the audio stops.
state enumeration is designed to block these attempts which can occur
when a non-standard call sequence of events occurs, normally due to
intermittent decoding.
starting/stopping the monitor for each call to simply leaving the
monitor task running continuously. Was causing issue with scheduled
thread executor not shutting down the task in a timely manner -- new
approach should avoid start/stop related issues altogether.
Updates P25DecoderState with reordering of timing of when we send
updates for the TO/FROM identifiers so that the audio manager receives
up-to-date metadata and hopefully resolves issue with 'Do Not Monitor'
talkgroups being monitored by the audio manager.
Updates the AudioManager with a try/catch block around the processing
task so that we can catch any audio processing exceptions that occur
during the timer task and might get suppressed by the scheduled thread
executor.
Resolves issue in tuner controller where controller was not releasing
sourced tuner channels, preventing re-tuning to other frequencies.
Resolves issue with channel timeout on MPT1327 monitored traffic
channels
A cross-platform java application for decoding, monitoring, recording and streaming trunked mobile and related radio protocols using Software Defined Radios (SDR).
sb.append("This action will sound a beep according to the interval that you select.\n\n");
sb.append("Once - Beep the first time the alias is active and never again.\n\n");
sb.append("Once, Reset After Delay - Beep once and suppress subsequent beeps for the specified period in seconds. After the reset period, it will beep again when the alias is active.\n\n");
sb.append("Until Dismissed - Beep every period seconds until you click OK on the dialog that appears. Alerting is suppressed for 15 seconds after you click OK.");
sb.append("This action will play the audio clip according to the interval that you select.\n\n");
sb.append("Once - Play the first time the alias is active and never again.\n\n");
sb.append("Once, Reset After Delay - Play once and suppress subsequent plays for the specified period in seconds. After the reset period, it will play again when the alias is active.\n\n");
sb.append("Until Dismissed - Play audio clip every period seconds until you click OK on the dialog that appears. Alerting is suppressed for 15 seconds after you click OK.");
sb.append("This action will run the script according to the interval that you select.\n\n");
sb.append("Once - Run script the first time the alias is active and never again.\n\n");
sb.append("Once, Reset After Delay - Run once and suppress subsequent runs for the specified period in seconds. After the reset period, it will play again when the alias is active.\n\n");
sb.append("Until Dismissed - Run script every period seconds until you click OK on the dialog that appears. Alerting is suppressed for 15 seconds after you click OK.");
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
