Make the nonlinear beamformer steerable

webrtc/modules/audio_processing/beamformer/nonlinear_beamformer.cc

 /*
  *  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.
  */
 
@@ skipping 11 matching lines @@
 #include "webrtc/modules/audio_processing/beamformer/covariance_matrix_generator.h"
 
 namespace webrtc {
 namespace {
 
 // Alpha for the Kaiser Bessel Derived window.
 const float kKbdAlpha = 1.5f;
 
 const float kSpeedOfSoundMeterSeconds = 343;
 
-// For both target and interference angles, PI / 2 is perpendicular to the
-// microphone array, facing forwards. The positive direction goes
-// counterclockwise.
-// The angle at which we amplify sound.
-// TODO(aluebs): Make the target angle dynamically settable.
-const float kTargetAngleRadians = static_cast<float>(M_PI) / 2.f;
-
 // The minimum separation in radians between the target direction and a
 // interferer scenario.
 const float kMinAwayRadians = 0.2f;
 
 // The separation between the target direction and the closest interferer
 // scenario is proportional to this constant.
 const float kAwaySlope = 0.008f;
 
 // When calculating the interference covariance matrix, this is the weight for
 // the weighted average between the uniform covariance matrix and the angled
@@ skipping 143 matching lines @@
     for (size_t j = i + 1; j < array_geometry.size(); ++j) {
       float distance = Distance(array_geometry[i], array_geometry[j]);
       if (distance < mic_spacing) {
         mic_spacing = distance;
       }
     }
   }
   return mic_spacing;
 }
 
+float DotProduct(std::vector<float> a, std::vector<float> b) {

[Andrew MacDonald, 2015/10/14 22:12:30]

const references.

[Andrew MacDonald, 2015/10/20 03:00:08]

Looks like you missed the remainder of the comments in this file.

[aluebs-webrtc, 2015/10/21 01:41:40]

I totally missed them, sorry about that. Addressing them now.

[aluebs-webrtc, 2015/10/21 01:41:40]

Done.

+  RTC_DCHECK_EQ(a.size(), b.size());
+  float dot_product = 0.f;
+  for (size_t i = 0u; i < a.size(); ++i) {
+    dot_product += a[i] * b[i];
+  }
+  return dot_product;
+}
+
+float NormalizedDotProduct(std::vector<float> a, std::vector<float> b) {

[Andrew MacDonald, 2015/10/14 22:12:31]

const references.

[aluebs-webrtc, 2015/10/21 01:41:40]

Removed this function.

+  const float kMinNorm = 1e-6f;
+  float norm_a = DotProduct(a, a);

[Andrew MacDonald, 2015/10/14 22:12:30]

You're repeatedly computing this for first_pair_direction. Can we improve that?

[aluebs-webrtc, 2015/10/21 01:41:40]

I removed the NormalizedDotProduct and did the normalization outside, so only
calculate the first norm once. It is a little messier, but more efficient.

+  float norm_b = DotProduct(b, b);
+  if (norm_a > kMinNorm && norm_b > kMinNorm) {
+    return DotProduct(a, b) / std::sqrt(norm_a * norm_b);
+  } else {
+    return 0.f;
+  }
+}
+
+bool IsGeometryLinear(std::vector<Point> array_geometry) {
+  const float kMinDotProduct = 0.9999f;
+  bool is_geometry_linear = true;
+  std::vector<float> directiona;

[Andrew MacDonald, 2015/10/14 22:12:31]

first_pair_direction

[aluebs-webrtc, 2015/10/21 01:41:39]

Done.

+  directiona.push_back(array_geometry[1].x() - array_geometry[0].x());
+  directiona.push_back(array_geometry[1].y() - array_geometry[0].y());
+  directiona.push_back(array_geometry[1].z() - array_geometry[0].z());

[Andrew MacDonald, 2015/10/14 22:12:31]

Write a helper for this, something like:
std::vector<float> PairDirection(const Point& a, const Point& b) {
  std::vector<float> direction;
  direction.reserve(3);
  direction.push_back(...);
  ...
  return direction;
}

const auto first_pair_direction = PairDirection(...);
...


Going further, is it really worth transforming these into vectors of floats? All
it does is save you one line in DotProduct. Instead you could have PairDirection
return a Point:

Point PairDirection(...) {
  return {b.x() - a.x(), b.y() - a.y(), b.z() - a.z()};
}

[aluebs-webrtc, 2015/10/21 01:41:40]

I though about using Point, but I thought it was stretching the Point class,
since they are directions and not Points, but I guess you are right. Done.

+  for (size_t i = 2u; i < array_geometry.size(); ++i) {
+    std::vector<float> directionb;
+    directionb.push_back(array_geometry[i].x() - array_geometry[i - 1].x());

[Andrew MacDonald, 2015/10/14 22:12:31]

pair_direction

[aluebs-webrtc, 2015/10/21 01:41:40]

Done.

+    directionb.push_back(array_geometry[i].y() - array_geometry[i - 1].y());
+    directionb.push_back(array_geometry[i].z() - array_geometry[i - 1].z());
+    is_geometry_linear &=
+        std::abs(NormalizedDotProduct(directiona, directionb)) > kMinDotProduct;

[Andrew MacDonald, 2015/10/14 22:12:31]

Instead, short circuit:

  if (std::abs(NormalizedDotProduct(...) < kMinDotProduct) {
    return false;
  }
}
return true;

[aluebs-webrtc, 2015/10/21 01:41:40]

I believe that function with only one return statement are less bug prone and
easier to understand, but I don't have a too strong opinion, so I changed it.
Done.

+  }
+  return is_geometry_linear;
+}
+
 }  // namespace
 
 // static
 const size_t NonlinearBeamformer::kNumFreqBins;
 
 NonlinearBeamformer::NonlinearBeamformer(
-    const std::vector<Point>& array_geometry)
+    const std::vector<Point>& array_geometry,
+    float target_angle_radians)
     : num_input_channels_(array_geometry.size()),
       array_geometry_(GetCenteredArray(array_geometry)),
-      mic_spacing_(GetMinimumSpacing(array_geometry)) {
+      mic_spacing_(GetMinimumSpacing(array_geometry)),
+      target_angle_radians_(target_angle_radians) {
   WindowGenerator::KaiserBesselDerived(kKbdAlpha, kFftSize, window_);
 }
 
 void NonlinearBeamformer::Initialize(int chunk_size_ms, int sample_rate_hz) {
   chunk_length_ =
       static_cast<size_t>(sample_rate_hz / (1000.f / chunk_size_ms));
   sample_rate_hz_ = sample_rate_hz;
-  InitFrequencyCorrectionRanges();
 
   high_pass_postfilter_mask_ = 1.f;
   is_target_present_ = false;
   hold_target_blocks_ = kHoldTargetSeconds * 2 * sample_rate_hz / kFftSize;
   interference_blocks_count_ = hold_target_blocks_;
 
   lapped_transform_.reset(new LappedTransform(num_input_channels_,
                                               1,
                                               chunk_length_,
                                               window_,
                                               kFftSize,
                                               kFftSize / 2,
                                               this));
   for (size_t i = 0; i < kNumFreqBins; ++i) {
     time_smooth_mask_[i] = 1.f;
     final_mask_[i] = 1.f;
     float freq_hz = (static_cast<float>(i) / kFftSize) * sample_rate_hz_;
     wave_numbers_[i] = 2 * M_PI * freq_hz / kSpeedOfSoundMeterSeconds;
   }
 
-  // Initialize all nonadaptive values before looping through the frames.
-  InitInterfAngles();
-  InitDelaySumMasks();
-  InitTargetCovMats();
-  InitInterfCovMats();
-
-  for (size_t i = 0; i < kNumFreqBins; ++i) {
-    rxiws_[i] = Norm(target_cov_mats_[i], delay_sum_masks_[i]);
-    rpsiws_[i].clear();
-    for (size_t j = 0; j < interf_angles_radians_.size(); ++j) {
-      rpsiws_[i].push_back(Norm(*interf_cov_mats_[i][j], delay_sum_masks_[i]));
-    }
-  }
+  InitLowFrequencyCorrectionRanges();
+  InitDifuseCovMats();

[Andrew MacDonald, 2015/10/14 22:12:31]

Diffuse

[aluebs-webrtc, 2015/10/21 01:41:40]

Done.

+  SteerBeam(target_angle_radians_);
 }
 
-void NonlinearBeamformer::InitFrequencyCorrectionRanges() {
+// These bin indexes determine the regions over which a mean is taken. This is
+// applied as a constant value over the adjacent end "frequency correction"
+// regions.
+//
+//             low_mean_start_bin_     high_mean_start_bin_
+//                   v                         v              constant
+// |----------------|--------|----------------|-------|----------------|
+//   constant               ^                        ^
+//             low_mean_end_bin_       high_mean_end_bin_
+//
+void NonlinearBeamformer::InitLowFrequencyCorrectionRanges() {
+  low_mean_start_bin_ = Round(kLowMeanStartHz * kFftSize / sample_rate_hz_);
+  low_mean_end_bin_ = Round(kLowMeanEndHz * kFftSize / sample_rate_hz_);
+
+  RTC_DCHECK_GT(low_mean_start_bin_, 0U);
+  RTC_DCHECK_LT(low_mean_start_bin_, low_mean_end_bin_);
+}
+
+void NonlinearBeamformer::InitHighFrequencyCorrectionRanges() {
   const float kAliasingFreqHz =
       kSpeedOfSoundMeterSeconds /
-      (mic_spacing_ * (1.f + std::abs(std::cos(kTargetAngleRadians))));
+      (mic_spacing_ * (1.f + std::abs(std::cos(target_angle_radians_))));
   const float kHighMeanStartHz = std::min(0.5f *  kAliasingFreqHz,
                                           sample_rate_hz_ / 2.f);
   const float kHighMeanEndHz = std::min(0.75f *  kAliasingFreqHz,
                                         sample_rate_hz_ / 2.f);
-
-  low_mean_start_bin_ = Round(kLowMeanStartHz * kFftSize / sample_rate_hz_);
-  low_mean_end_bin_ = Round(kLowMeanEndHz * kFftSize / sample_rate_hz_);
   high_mean_start_bin_ = Round(kHighMeanStartHz * kFftSize / sample_rate_hz_);
   high_mean_end_bin_ = Round(kHighMeanEndHz * kFftSize / sample_rate_hz_);
-  // These bin indexes determine the regions over which a mean is taken. This
-  // is applied as a constant value over the adjacent end "frequency correction"
-  // regions.
-  //
-  //             low_mean_start_bin_     high_mean_start_bin_
-  //                   v                         v              constant
-  // |----------------|--------|----------------|-------|----------------|
-  //   constant               ^                        ^
-  //             low_mean_end_bin_       high_mean_end_bin_
-  //
-  RTC_DCHECK_GT(low_mean_start_bin_, 0U);
-  RTC_DCHECK_LT(low_mean_start_bin_, low_mean_end_bin_);
+
   RTC_DCHECK_LT(low_mean_end_bin_, high_mean_end_bin_);
   RTC_DCHECK_LT(high_mean_start_bin_, high_mean_end_bin_);
   RTC_DCHECK_LT(high_mean_end_bin_, kNumFreqBins - 1);
 }
 
-
 void NonlinearBeamformer::InitInterfAngles() {
   const float kAway =
       std::min(static_cast<float>(M_PI),
                std::max(kMinAwayRadians,
                         kAwaySlope * static_cast<float>(M_PI) / mic_spacing_));
 
   interf_angles_radians_.clear();
-  // TODO(aluebs): When the target angle is settable, make sure the interferer
-  // scenarios aren't reflected over the target one for linear geometries.
-  interf_angles_radians_.push_back(kTargetAngleRadians - kAway);
-  interf_angles_radians_.push_back(kTargetAngleRadians + kAway);
+  if (IsGeometryLinear(array_geometry_)) {
+    if (target_angle_radians_ - kAway >= 0.f) {

[Andrew MacDonald, 2015/10/14 22:12:30]

These values depend on the array being parallel with the x-axis (so that the
angle of reflection is the x-axis). What if that's not the case?

[aluebs-webrtc, 2015/10/21 01:41:40]

Great Point. Now I generalized it.

+      interf_angles_radians_.push_back(target_angle_radians_ - kAway);
+    } else {
+      interf_angles_radians_.push_back(M_PI);
+    }
+    if (target_angle_radians_ + kAway <= M_PI) {
+      interf_angles_radians_.push_back(target_angle_radians_ + kAway);
+    } else {
+      interf_angles_radians_.push_back(0.f);
+    }
+  } else {
+    interf_angles_radians_.push_back(target_angle_radians_ - kAway);
+    interf_angles_radians_.push_back(target_angle_radians_ + kAway);
+  }
 }
 
 void NonlinearBeamformer::InitDelaySumMasks() {
   for (size_t f_ix = 0; f_ix < kNumFreqBins; ++f_ix) {
     delay_sum_masks_[f_ix].Resize(1, num_input_channels_);
     CovarianceMatrixGenerator::PhaseAlignmentMasks(f_ix,
                                                    kFftSize,
                                                    sample_rate_hz_,
                                                    kSpeedOfSoundMeterSeconds,
                                                    array_geometry_,
-                                                   kTargetAngleRadians,
+                                                   target_angle_radians_,
                                                    &delay_sum_masks_[f_ix]);
 
     complex_f norm_factor = sqrt(
         ConjugateDotProduct(delay_sum_masks_[f_ix], delay_sum_masks_[f_ix]));
     delay_sum_masks_[f_ix].Scale(1.f / norm_factor);
     normalized_delay_sum_masks_[f_ix].CopyFrom(delay_sum_masks_[f_ix]);
     normalized_delay_sum_masks_[f_ix].Scale(1.f / SumAbs(
         normalized_delay_sum_masks_[f_ix]));
   }
 }
 
 void NonlinearBeamformer::InitTargetCovMats() {
   for (size_t i = 0; i < kNumFreqBins; ++i) {
     target_cov_mats_[i].Resize(num_input_channels_, num_input_channels_);
     TransposedConjugatedProduct(delay_sum_masks_[i], &target_cov_mats_[i]);
   }
 }
 
+void NonlinearBeamformer::InitDifuseCovMats() {
+  for (size_t i = 0; i < kNumFreqBins; ++i) {
+    uniform_cov_mat_[i].Resize(num_input_channels_, num_input_channels_);
+    CovarianceMatrixGenerator::UniformCovarianceMatrix(wave_numbers_[i],
+                                                       array_geometry_,
+                                                       &uniform_cov_mat_[i]);
+    complex_f normalization_factor = uniform_cov_mat_[i].elements()[0][0];
+    uniform_cov_mat_[i].Scale(1.f / normalization_factor);
+    uniform_cov_mat_[i].Scale(1 - kBalance);
+  }
+}
+
 void NonlinearBeamformer::InitInterfCovMats() {
   for (size_t i = 0; i < kNumFreqBins; ++i) {
-    ComplexMatrixF uniform_cov_mat(num_input_channels_, num_input_channels_);
-    CovarianceMatrixGenerator::UniformCovarianceMatrix(wave_numbers_[i],
-                                                       array_geometry_,
-                                                       &uniform_cov_mat);
-    complex_f normalization_factor = uniform_cov_mat.elements()[0][0];
-    uniform_cov_mat.Scale(1.f / normalization_factor);
-    uniform_cov_mat.Scale(1 - kBalance);
     interf_cov_mats_[i].clear();
     for (size_t j = 0; j < interf_angles_radians_.size(); ++j) {
       interf_cov_mats_[i].push_back(new ComplexMatrixF(num_input_channels_,
                                                        num_input_channels_));
       ComplexMatrixF angled_cov_mat(num_input_channels_, num_input_channels_);
       CovarianceMatrixGenerator::AngledCovarianceMatrix(
           kSpeedOfSoundMeterSeconds,
           interf_angles_radians_[j],
           i,
           kFftSize,
           kNumFreqBins,
           sample_rate_hz_,
           array_geometry_,
           &angled_cov_mat);
       // Normalize matrices before averaging them.
-      normalization_factor = angled_cov_mat.elements()[0][0];
+      complex_f normalization_factor = angled_cov_mat.elements()[0][0];
       angled_cov_mat.Scale(1.f / normalization_factor);
       // Weighted average of matrices.
       angled_cov_mat.Scale(kBalance);
-      interf_cov_mats_[i][j]->Add(uniform_cov_mat, angled_cov_mat);
+      interf_cov_mats_[i][j]->Add(uniform_cov_mat_[i], angled_cov_mat);
     }
   }
 }
+
+void NonlinearBeamformer::NormalizeCovMats() {
+  for (size_t i = 0; i < kNumFreqBins; ++i) {
+    rxiws_[i] = Norm(target_cov_mats_[i], delay_sum_masks_[i]);
+    rpsiws_[i].clear();
+    for (size_t j = 0; j < interf_angles_radians_.size(); ++j) {
+      rpsiws_[i].push_back(Norm(*interf_cov_mats_[i][j], delay_sum_masks_[i]));
+    }
+  }
+}
 
 void NonlinearBeamformer::ProcessChunk(const ChannelBuffer<float>& input,
                                        ChannelBuffer<float>* output) {
   RTC_DCHECK_EQ(input.num_channels(), num_input_channels_);
   RTC_DCHECK_EQ(input.num_frames_per_band(), chunk_length_);
 
   float old_high_pass_mask = high_pass_postfilter_mask_;
   lapped_transform_->ProcessChunk(input.channels(0), output->channels(0));
   // Ramp up/down for smoothing. 1 mask per 10ms results in audible
   // discontinuities.
@@ skipping 11 matching lines @@
       // averaging).
       float sum = 0.f;
       for (int k = 0; k < input.num_channels(); ++k) {
         sum += input.channels(i)[k][j];
       }
       output->channels(i)[0][j] = sum / input.num_channels() * smoothed_mask;
     }
   }
 }
 
+void NonlinearBeamformer::SteerBeam(float target_angle_radians) {
+  target_angle_radians_ = target_angle_radians;
+  InitHighFrequencyCorrectionRanges();
+  InitInterfAngles();
+  InitDelaySumMasks();
+  InitTargetCovMats();
+  InitInterfCovMats();
+  NormalizeCovMats();
+}
+
 bool NonlinearBeamformer::IsInBeam(const SphericalPointf& spherical_point) {
   // If more than half-beamwidth degrees away from the beam's center,
   // you are out of the beam.
-  return fabs(spherical_point.azimuth() - kTargetAngleRadians) <
+  return fabs(spherical_point.azimuth() - target_angle_radians_) <
          kHalfBeamWidthRadians;
 }
 
 void NonlinearBeamformer::ProcessAudioBlock(const complex_f* const* input,
                                             int num_input_channels,
                                             size_t num_freq_bins,
                                             int num_output_channels,
                                             complex_f* const* output) {
   RTC_CHECK_EQ(num_freq_bins, kNumFreqBins);
   RTC_CHECK_EQ(num_input_channels, num_input_channels_);
@@ skipping 144 matching lines @@
                    new_mask_ + high_mean_end_bin_ + 1);
   if (new_mask_[quantile] > kMaskTargetThreshold) {
     is_target_present_ = true;
     interference_blocks_count_ = 0;
   } else {
     is_target_present_ = interference_blocks_count_++ < hold_target_blocks_;
   }
 }
 
 }  // namespace webrtc