simple synced capture

Signed-off-by: PopPaul2021 <Paul.Pop@analog.com>

Author: PopPaul2021

examples/sharkbyte_example.py

@@ -21,21 +21,21 @@
 print("uri: " + str(my_uri))
 
 # Individual device access (for first plot demonstration)
-hmcad15xx_dev1 = adi.hmcad15xx(uri=my_uri,device_name="axi_adc1_hmcad15xx")
-hmcad15xx_dev2 = adi.hmcad15xx(uri=my_uri,device_name="axi_adc2_hmcad15xx")
-ad5696_dev     = adi.ad5686(uri=my_uri)
+hmcad15xx_dev1   = adi.hmcad15xx(uri=my_uri,device_name="axi_adc1_hmcad15xx")
+hmcad15xx_dev2   = adi.hmcad15xx(uri=my_uri,device_name="axi_adc2_hmcad15xx")
 gpio_controller  = adi.one_bit_adc_dac(uri=my_uri, name="one-bit-adc-dac")
+ad5696_dev       = adi.ad5686(uri=my_uri)
+tddn             = adi.tddn(my_uri)
 
 tddn_channel_on_raw = 0
 tddn_channel_off_raw = 10
 tddn_channel_polarity = 0
 tddn_channel_enable = 1
-
-tddn = adi.tddn(my_uri)
 tddn.burst_count = 0
 tddn.startup_delay_ms = 0
 tddn.frame_length_ms  = 1
 
+
 tddn.enable = 0
 
 tddn.channel[0].on_ms    = 0
@@ -79,34 +79,6 @@
 ramp_pattern         = 0x40
 pattern_disabled     = 0x00
 
-hmcad15xx_dev1.rx_buffer_size = 2**18
-hmcad15xx_dev2.rx_buffer_size = 2**18
-
-hmcad15xx_dev1.rx_enabled_channels = [0]
-hmcad15xx_dev2.rx_enabled_channels = [0]
-
-print("RX1 rx_enabled_channels: " + str(hmcad15xx_dev1.rx_enabled_channels))
-print("RX2 rx_enabled_channels: " + str(hmcad15xx_dev2.rx_enabled_channels))
-
-hmcad15xx_dev1.hmcad15xx_register_write(0x42, 0x0000)  # Phase DDR
-hmcad15xx_dev2.hmcad15xx_register_write(0x42, 0x0000)  # Phase DDR
-hmcad15xx_dev1.hmcad15xx_register_write(0x26, custom_pattern)
-hmcad15xx_dev2.hmcad15xx_register_write(0x26, custom_pattern)
-hmcad15xx_dev1.hmcad15xx_register_write(0x25, ramp_pattern)
-hmcad15xx_dev2.hmcad15xx_register_write(0x25, ramp_pattern)
-
-#input_select_available value: IP1_IN1 IP2_IN2 IP3_IN3
-
-hmcad15xx_dev1.channel[0].input_select = "IP1_IN1"
-hmcad15xx_dev1.channel[1].input_select = "IP1_IN1"
-hmcad15xx_dev1.channel[2].input_select = "IP1_IN1"
-hmcad15xx_dev1.channel[3].input_select = "IP1_IN1"
-
-hmcad15xx_dev2.channel[0].input_select = "IP1_IN1"
-hmcad15xx_dev2.channel[1].input_select = "IP1_IN1"
-hmcad15xx_dev2.channel[2].input_select = "IP1_IN1"
-hmcad15xx_dev2.channel[3].input_select = "IP1_IN1"
-
 # Enable TDD and configure sync for synchronized multi-ADC capture
 tddn.enable = 1
 tddn.sync_external = False
@@ -120,7 +92,7 @@
 
 # Configure buffer size and enabled channels
 multi.rx_buffer_size = 2**17
-multi.rx_enabled_channels = [0]  # Apply to both devices
+multi.rx_enabled_channels = [3]  # Apply to both devices
 
 print("Multi-ADC rx_enabled_channels: " + str(multi.rx_enabled_channels))
 
@@ -154,205 +126,10 @@
 plt.title('ADC Data - Channel 0 from Both Devices (TDD Synchronized)')
 plt.legend()
 plt.grid(True)
-plt.savefig("Data channels 0 of both devices - Synchronized")
+plt.show()
 plt.close()  # Close the figure to free memory
 
 print("Synchronized data capture complete!")
 
-num_captures = 1000
-phase_shifts_samples = []
-phase_shifts_degrees = []
-correlation_peaks = []
-
-# Phase Shift Analysis using Cross-Correlation
-print("\n--- Phase Shift Analysis ---")
-print(f"Performing {num_captures} captures for cross-correlation analysis...")
-
-sampling_rate = multi.dev1.sampling_rate
-print(f"Sampling rate: {sampling_rate} Hz")
-
-for i in range(num_captures):
-    if (i + 1) % 10 == 0:
-        print(f"Progress: {i + 1}/{num_captures} captures")
-
-    # Capture synchronized data
-    data1, data2 = multi.rx()
-
-    # Normalize signals (remove DC offset and normalize amplitude)
-    data1_norm = (data1 - np.mean(data1)) / np.std(data1)
-    data2_norm = (data2 - np.mean(data2)) / np.std(data2)
-
-    # Compute cross-correlation using FFT (much faster for large arrays)
-    # This is equivalent to np.correlate(data1_norm, data2_norm, mode='full')
-    # but uses FFT for O(N log N) complexity instead of O(N^2)
-    fft1 = np.fft.fft(data1_norm, n=2*len(data1_norm))
-    fft2 = np.fft.fft(data2_norm, n=2*len(data2_norm))
-    correlation = np.fft.ifft(fft1 * np.conj(fft2)).real
-
-    # Rearrange to match 'full' mode output (shift zero-lag to center)
-    correlation = np.concatenate([correlation[len(data1):], correlation[:len(data1)]])
-
-    # Find the lag with maximum correlation
-    max_corr_idx = np.argmax(correlation)
-    lag = max_corr_idx - (len(data1) - 1)  # Offset to center
-
-    # Store results
-    phase_shifts_samples.append(lag)
-    correlation_peaks.append(correlation[max_corr_idx])
-
-    # Calculate phase shift in degrees (if signal is periodic)
-    # For a sinusoidal signal, we can estimate phase shift
-    # Phase shift in samples / samples per period * 360 degrees
-    # We'll store the sample shift for now and convert later if needed
-
-# Convert to numpy arrays for analysis
-phase_shifts_samples = np.array(phase_shifts_samples)
-correlation_peaks = np.array(correlation_peaks)
-
-# Statistical analysis
-mean_shift = np.mean(phase_shifts_samples)
-std_shift = np.std(phase_shifts_samples)
-median_shift = np.median(phase_shifts_samples)
-
-print(f"\nPhase Shift Statistics (in samples):")
-print(f"  Mean: {mean_shift:.3f} samples")
-print(f"  Std Dev: {std_shift:.3f} samples")
-print(f"  Median: {median_shift:.3f} samples")
-print(f"  Min: {np.min(phase_shifts_samples)} samples")
-print(f"  Max: {np.max(phase_shifts_samples)} samples")
-
-# Time delay in seconds
-time_delay_mean = mean_shift / sampling_rate
-time_delay_std = std_shift / sampling_rate
-print(f"\nTime Delay:")
-print(f"  Mean: {time_delay_mean*1e9:.3f} ns")
-print(f"  Std Dev: {time_delay_std*1e9:.3f} ns")
-
-# Create comprehensive analysis plots
-fig = plt.figure(figsize=(16, 12))
-
-# Plot 1: Histogram of phase shifts
-ax1 = plt.subplot(3, 2, 1)
-plt.hist(phase_shifts_samples, bins=50, edgecolor='black', alpha=0.7)
-plt.axvline(mean_shift, color='red', linestyle='--', linewidth=2, label=f'Mean: {mean_shift:.2f}')
-plt.axvline(median_shift, color='green', linestyle='--', linewidth=2, label=f'Median: {median_shift:.2f}')
-plt.xlabel('Phase Shift (samples)')
-plt.ylabel('Frequency')
-plt.title('Distribution of Phase Shifts Across 1000 Captures')
-plt.legend()
-plt.grid(True, alpha=0.3)
-
-# Plot 2: Time series of phase shift
-ax2 = plt.subplot(3, 2, 2)
-plt.plot(phase_shifts_samples, alpha=0.6, linewidth=0.5)
-plt.axhline(mean_shift, color='red', linestyle='--', linewidth=2, label=f'Mean: {mean_shift:.2f}')
-plt.fill_between(range(num_captures),
-                  mean_shift - std_shift,
-                  mean_shift + std_shift,
-                  alpha=0.2, color='red', label=f'±1σ: {std_shift:.2f}')
-plt.xlabel('Capture Number')
-plt.ylabel('Phase Shift (samples)')
-plt.title('Phase Shift Over Time')
-plt.legend()
-plt.grid(True, alpha=0.3)
-
-# Plot 3: Correlation peak values
-ax3 = plt.subplot(3, 2, 3)
-plt.plot(correlation_peaks, alpha=0.6, linewidth=0.5)
-plt.axhline(np.mean(correlation_peaks), color='red', linestyle='--', linewidth=2, label=f'Mean: {np.mean(correlation_peaks):.2f}')
-plt.xlabel('Capture Number')
-plt.ylabel('Correlation Peak Value')
-plt.title('Cross-Correlation Peak Strength')
-plt.legend()
-plt.grid(True, alpha=0.3)
-
-# Plot 4: Example cross-correlation function
-ax4 = plt.subplot(3, 2, 4)
-# Perform one final capture to show the correlation function
-data1_ex, data2_ex = multi.rx()
-data1_norm_ex = (data1_ex - np.mean(data1_ex)) / np.std(data1_ex)
-data2_norm_ex = (data2_ex - np.mean(data2_ex)) / np.std(data2_ex)
-# Use FFT-based cross-correlation for speed
-fft1_ex = np.fft.fft(data1_norm_ex, n=2*len(data1_norm_ex))
-fft2_ex = np.fft.fft(data2_norm_ex, n=2*len(data2_norm_ex))
-correlation_ex = np.fft.ifft(fft1_ex * np.conj(fft2_ex)).real
-correlation_ex = np.concatenate([correlation_ex[len(data1_ex):], correlation_ex[:len(data1_ex)]])
-lags_ex = np.arange(-len(data1_ex) + 1, len(data1_ex))
-
-# Plot only around the peak for clarity
-max_idx = np.argmax(correlation_ex)
-window = 1000  # Show ±1000 samples around peak
-start_idx = max(0, max_idx - window)
-end_idx = min(len(correlation_ex), max_idx + window)
-
-plt.plot(lags_ex[start_idx:end_idx], correlation_ex[start_idx:end_idx], linewidth=1)
-plt.axvline(lags_ex[max_idx], color='red', linestyle='--', linewidth=2, label=f'Peak at lag={lags_ex[max_idx]}')
-plt.xlabel('Lag (samples)')
-plt.ylabel('Cross-Correlation')
-plt.title('Example Cross-Correlation Function')
-plt.legend()
-plt.grid(True, alpha=0.3)
-
-# Plot 5: Example synchronized signals (zoomed in)
-ax5 = plt.subplot(3, 2, 5)
-zoom_samples = min(1000, len(data1_ex))  # Show first 1000 samples
-plt.plot(data1_ex[:zoom_samples], label='Device 1', alpha=0.7, linewidth=1)
-plt.plot(data2_ex[:zoom_samples], label='Device 2', alpha=0.7, linewidth=1)
-plt.xlabel('Sample Index')
-plt.ylabel('ADC Value')
-plt.title('Example Synchronized Signals (Zoomed)')
-plt.legend()
-plt.grid(True, alpha=0.3)
-
-# Plot 6: Phase shift distribution (box plot)
-ax6 = plt.subplot(3, 2, 6)
-plt.boxplot(phase_shifts_samples, vert=True)
-plt.ylabel('Phase Shift (samples)')
-plt.title('Phase Shift Distribution (Box Plot)')
-plt.grid(True, alpha=0.3, axis='y')
-plt.text(0.5, mean_shift, f'Mean: {mean_shift:.2f}\nStd: {std_shift:.2f}',
-         transform=ax6.get_yaxis_transform(), ha='left', va='center',
-         bbox=dict(boxstyle='round', facecolor='wheat', alpha=0.5))
-
-plt.tight_layout()
-plt.savefig("Phase_Shift_Analysis_1000_Captures")
-plt.close()
-
-print("\nPhase shift analysis complete!")
-print(f"Analysis plot saved as: Phase_Shift_Analysis_{num_captures}_Captures.png")
-
-# Summary report
-print("\n" + "="*60)
-print("PHASE SHIFT ANALYSIS SUMMARY")
-print("="*60)
-print(f"Number of captures: {num_captures}")
-print(f"Sampling rate: {sampling_rate} Hz")
-print(f"\nPhase alignment:")
-if abs(mean_shift) < 1.0:
-    print(f"  ✓ Excellent alignment (mean shift: {mean_shift:.3f} samples)")
-elif abs(mean_shift) < 5.0:
-    print(f"  ✓ Good alignment (mean shift: {mean_shift:.3f} samples)")
-else:
-    print(f"  ⚠ Measurable shift detected (mean shift: {mean_shift:.3f} samples)")
-
-print(f"\nConsistency:")
-if std_shift < 0.5:
-    print(f"  ✓ Very consistent (σ = {std_shift:.3f} samples)")
-elif std_shift < 2.0:
-    print(f"  ✓ Consistent (σ = {std_shift:.3f} samples)")
-else:
-    print(f"  ⚠ Variable phase shift (σ = {std_shift:.3f} samples)")
-
-print(f"\nCorrelation quality:")
-mean_corr = np.mean(correlation_peaks)
-print(f"  Mean correlation peak: {mean_corr:.3f}")
-if mean_corr > 0.9:
-    print(f"  ✓ Excellent correlation")
-elif mean_corr > 0.7:
-    print(f"  ✓ Good correlation")
-else:
-    print(f"  ⚠ Moderate correlation")
-print("="*60)
 
-# Cleanup buffers
 multi.rx_destroy_buffer()