Implement new version of the NonlinearBeamformer

webrtc/modules/audio_processing/beamformer/nonlinear_beamformer.h

 /*
  *  Copyright (c) 2014 The WebRTC project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS.  All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
 #ifndef WEBRTC_MODULES_AUDIO_PROCESSING_BEAMFORMER_NONLINEAR_BEAMFORMER_H_
 #define WEBRTC_MODULES_AUDIO_PROCESSING_BEAMFORMER_NONLINEAR_BEAMFORMER_H_
 
 #include <vector>
 
 #include "webrtc/common_audio/lapped_transform.h"
 #include "webrtc/common_audio/channel_buffer.h"
 #include "webrtc/modules/audio_processing/beamformer/beamformer.h"
 #include "webrtc/modules/audio_processing/beamformer/complex_matrix.h"
+#include "webrtc/system_wrappers/interface/scoped_vector.h"
 
 namespace webrtc {
 
 // Enhances sound sources coming directly in front of a uniform linear array
 // and suppresses sound sources coming from all other directions. Operates on
 // multichannel signals and produces single-channel output.
 //
 // The implemented nonlinear postfilter algorithm taken from "A Robust Nonlinear
 // Beamforming Postprocessor" by Bastiaan Kleijn.
-//
-// TODO(aluebs): Target angle assumed to be 0. Parameterize target angle.
 class NonlinearBeamformer
   : public Beamformer<float>,
     public LappedTransform::Callback {
  public:
-  // At the moment it only accepts uniform linear microphone arrays. Using the
-  // first microphone as a reference position [0, 0, 0] is a natural choice.
   explicit NonlinearBeamformer(const std::vector<Point>& array_geometry);
 
   // Sample rate corresponds to the lower band.
   // Needs to be called before the NonlinearBeamformer can be used.
   void Initialize(int chunk_size_ms, int sample_rate_hz) override;
 
   // Process one time-domain chunk of audio. The audio is expected to be split
   // into frequency bands inside the ChannelBuffer. The number of frames and
   // channels must correspond to the constructor parameters. The same
   // ChannelBuffer can be passed in as |input| and |output|.
@@ skipping 15 matching lines @@
                          int num_input_channels,
                          size_t num_freq_bins,
                          int num_output_channels,
                          complex<float>* const* output) override;
 
  private:
   typedef Matrix<float> MatrixF;
   typedef ComplexMatrix<float> ComplexMatrixF;
   typedef complex<float> complex_f;
 
+  void InitInterfAngles();
   void InitDelaySumMasks();
-  void InitTargetCovMats();  // TODO(aluebs): Make this depend on target angle.
+  void InitTargetCovMats();
   void InitInterfCovMats();
 
-  // An implementation of equation 18, which calculates postfilter masks that,
-  // when applied, minimize the mean-square error of our estimation of the
-  // desired signal. A sub-task is to calculate lambda, which is solved via
-  // equation 13.
+  // Calculates postfilter masks that minimize the mean squared error of our
+  // estimation of the desired signal.
   float CalculatePostfilterMask(const ComplexMatrixF& interf_cov_mat,
                                 float rpsiw,
                                 float ratio_rxiw_rxim,
-                                float rmxi_r,
-                                float mask_threshold);
+                                float rmxi_r);
 
   // Prevents the postfilter masks from degenerating too quickly (a cause of
   // musical noise).
   void ApplyMaskTimeSmoothing();
   void ApplyMaskFrequencySmoothing();
 
   // The postfilter masks are unreliable at low frequencies. Calculates a better
   // mask by averaging mid-low frequency values.
   void ApplyLowFrequencyCorrection();
 
@@ skipping 32 matching lines @@
   size_t high_mean_start_bin_;
   size_t high_mean_end_bin_;
 
   // Quickly varying mask updated every block.
   float new_mask_[kNumFreqBins];
   // Time smoothed mask.
   float time_smooth_mask_[kNumFreqBins];
   // Time and frequency smoothed mask.
   float final_mask_[kNumFreqBins];
 
+  // Angles of the interferer scenarios.
+  std::vector<float> interf_angles_radians_;
+
   // Array of length |kNumFreqBins|, Matrix of size |1| x |num_channels_|.
   ComplexMatrixF delay_sum_masks_[kNumFreqBins];
   ComplexMatrixF normalized_delay_sum_masks_[kNumFreqBins];
 
   // Array of length |kNumFreqBins|, Matrix of size |num_input_channels_| x
   // |num_input_channels_|.
   ComplexMatrixF target_cov_mats_[kNumFreqBins];
 
   // Array of length |kNumFreqBins|, Matrix of size |num_input_channels_| x
-  // |num_input_channels_|.
-  ComplexMatrixF interf_cov_mats_[kNumFreqBins];
-  ComplexMatrixF reflected_interf_cov_mats_[kNumFreqBins];
+  // |num_input_channels_|. ScopedVector has a size equal to the number of
+  // interferer scenarios.
+  ScopedVector<ComplexMatrixF> interf_cov_mats_[kNumFreqBins];
 
   // Of length |kNumFreqBins|.
-  float mask_thresholds_[kNumFreqBins];
   float wave_numbers_[kNumFreqBins];
 
   // Preallocated for ProcessAudioBlock()
   // Of length |kNumFreqBins|.
   float rxiws_[kNumFreqBins];
-  float rpsiws_[kNumFreqBins];
-  float reflected_rpsiws_[kNumFreqBins];
+  // The vector has a size equal to the number of interferer scenarios.
+  std::vector<float> rpsiws_[kNumFreqBins];
 
   // The microphone normalization factor.
   ComplexMatrixF eig_m_;
 
   // For processing the high-frequency input signal.
   float high_pass_postfilter_mask_;
 
   // True when the target signal is present.
   bool is_target_present_;
   // Number of blocks after which the data is considered interference if the
   // mask does not pass |kMaskSignalThreshold|.
   size_t hold_target_blocks_;
   // Number of blocks since the last mask that passed |kMaskSignalThreshold|.
   size_t interference_blocks_count_;
 };
 
 }  // namespace webrtc
 
 #endif  // WEBRTC_MODULES_AUDIO_PROCESSING_BEAMFORMER_NONLINEAR_BEAMFORMER_H_