# -*- coding: utf-8 -*-
#
# DPD Calculation Engine, model implementation.
#
# http://www.opendigitalradio.org
# Licence: The MIT License, see notice at the end of this file
import datetime
import os
import logging
logging_path = os.path.dirname(logging.getLoggerClass().root.handlers[0].baseFilename)
import numpy as np
import matplotlib
matplotlib.use('agg')
import matplotlib.pyplot as plt
from sklearn.linear_model import Ridge
class Model:
"""Calculates new coefficients using the measurement and the old
coefficients"""
def __init__(self, coefs_am, coefs_pm):
self.coefs_am = coefs_am
self.coefs_history = [coefs_am, ]
self.mses = [0, ]
self.errs = [0, ]
self.coefs_pm = coefs_pm
self.coefs_pm_history = [coefs_pm, ]
self.errs_phase = [0, ]
def sample_uniformly(self, txframe_aligned, rxframe_aligned, n_bins=4):
"""This function returns tx and rx samples in a way
that the tx amplitudes have an approximate uniform
distribution with respect to the txframe_aligned amplitudes"""
txframe_aligned_abs = np.abs(txframe_aligned)
ccdf_min = 0
ccdf_max = np.max(txframe_aligned_abs)
tx_hist, ccdf_edges = np.histogram(txframe_aligned_abs,
bins=n_bins,
range=(ccdf_min, ccdf_max))
n_choise = np.min(tx_hist)
tx_choice = np.zeros(n_choise * n_bins, dtype=np.complex64)
rx_choice = np.zeros(n_choise * n_bins, dtype=np.complex64)
for idx, bin in enumerate(tx_hist):
indices = np.where((txframe_aligned_abs >= ccdf_edges[idx]) &
(txframe_aligned_abs <= ccdf_edges[idx+1]))[0]
indices_choise = np.random.choice(indices, n_choise, replace=False)
rx_choice[idx*n_choise:(idx+1)*n_choise] = rxframe_aligned[indices_choise]
tx_choice[idx*n_choise:(idx+1)*n_choise] = txframe_aligned[indices_choise]
return tx_choice, rx_choice
def get_next_coefs(self, txframe_aligned, rxframe_aligned):
tx_choice, rx_choice = self.sample_uniformly(txframe_aligned, rxframe_aligned)
# Calculate new coefficients for AM/AM correction
rx_abs = np.abs(rx_choice)
rx_A = np.vstack([rx_abs,
rx_abs ** 3,
rx_abs ** 5,
rx_abs ** 7,
rx_abs ** 9,
]).T
rx_dpd = np.sum(rx_A * self.coefs_am, axis=1)
rx_dpd = rx_dpd * (
np.median(np.abs(tx_choice)) / np.median(np.abs(rx_dpd)))
err = rx_dpd - np.abs(tx_choice)
self.errs.append(np.mean(np.abs(err ** 2)))
a_delta = np.linalg.lstsq(rx_A, err)[0]
new_coefs = self.coefs_am - 0.1 * a_delta
new_coefs = new_coefs * (self.coefs_am[0] / new_coefs[0])
logging.debug("a_delta {}".format(a_delta))
logging.debug("new coefs_am {}".format(new_coefs))
# Calculate new coefficients for AM/PM correction
phase_diff_rad = ((
(np.angle(tx_choice) -
np.angle(rx_choice) +
np.pi) % (2 * np.pi)) -
np.pi
)
tx_abs = np.abs(tx_choice)
tx_abs_A = np.vstack([tx_abs,
tx_abs ** 2,
tx_abs ** 3,
tx_abs ** 4,
tx_abs ** 5,
]).T
phase_dpd = np.sum(tx_abs_A * self.coefs_pm, axis=1)
err_phase = phase_dpd - phase_diff_rad
self.errs_phase.append(np.mean(np.abs(err_phase ** 2)))
a_delta = np.linalg.lstsq(tx_abs_A, err_phase)[0]
new_coefs_pm = self.coefs_pm - 0.1 * a_delta
logging.debug("a_delta {}".format(a_delta))
logging.debug("new new_coefs_pm {}".format(new_coefs_pm))
def dpd_phase(tx):
tx_abs = np.abs(tx)
tx_A_complex = np.vstack([tx,
tx * tx_abs ** 1,
tx * tx_abs ** 2,
tx * tx_abs ** 3,
tx * tx_abs ** 4,
]).T
tx_dpd = np.sum(tx_A_complex * self.coefs_pm, axis=1)
return tx_dpd
tx_range = np.linspace(0, 2)
phase_range_dpd = dpd_phase(tx_range)
rx_A_complex = np.vstack([rx_choice,
rx_choice * rx_abs ** 2,
rx_choice * rx_abs ** 4,
rx_choice * rx_abs ** 6,
rx_choice * rx_abs ** 8,
]).T
rx_post_distored = np.sum(rx_A_complex * self.coefs_am, axis=1)
rx_post_distored = rx_post_distored * (
np.median(np.abs(tx_choice)) /
np.median(np.abs(rx_post_distored)))
mse = np.mean(np.abs((tx_choice - rx_post_distored) ** 2))
logging.debug("MSE: {}".format(mse))
self.mses.append(mse)
def dpd(tx):
tx_abs = np.abs(tx)
tx_A_complex = np.vstack([tx,
tx * tx_abs ** 2,
tx * tx_abs ** 4,
tx * tx_abs ** 6,
tx * tx_abs ** 8,
]).T
tx_dpd = np.sum(tx_A_complex * self.coefs_am, axis=1)
return tx_dpd
rx_range = np.linspace(0, 1, num=100)
rx_range_dpd = dpd(rx_range)
rx_range = rx_range[(rx_range_dpd > 0) & (rx_range_dpd < 2)]
rx_range_dpd = rx_range_dpd[(rx_range_dpd > 0) & (rx_range_dpd < 2)]
if logging.getLogger().getEffectiveLevel() == logging.DEBUG:
logging.debug("txframe: min %f, max %f, median %f" %
(np.min(np.abs(txframe_aligned)),
np.max(np.abs(txframe_aligned)),
np.median(np.abs(txframe_aligned))
))
logging.debug("rxframe: min %f, max %f, median %f" %
(np.min(np.abs(rx_choice)),
np.max(np.abs(rx_choice)),
np.median(np.abs(rx_choice))
))
dt = datetime.datetime.now().isoformat()
fig_path = logging_path + "/" + dt + "_Model.pdf"
fig = plt.figure(figsize=(3*6, 1.5 * 6))
ax = plt.subplot(3,3,1)
ax.plot(np.abs(txframe_aligned[:128]),
label="TX sent",
linestyle=":")
ax.plot(np.abs(rxframe_aligned[:128]),
label="RX received",
color="red")
ax.set_title("Synchronized Signals of Iteration {}".format(len(self.coefs_history)))
ax.set_xlabel("Samples")
ax.set_ylabel("Amplitude")
ax.text(0, 0, "TX (max {:01.3f}, mean {:01.3f}, median {:01.3f})".format(
np.max(np.abs(txframe_aligned)),
np.mean(np.abs(txframe_aligned)),
np.median(np.abs(txframe_aligned))
), size = 8)
ax.legend(loc=4)
ax = plt.subplot(3,3,2)
ax.plot(np.real(txframe_aligned[:128]),
label="TX sent",
linestyle=":")
ax.plot(np.real(rxframe_aligned[:128]),
label="RX received",
color="red")
ax.set_title("Synchronized Signals")
ax.set_xlabel("Samples")
ax.set_ylabel("Real Part")
ax.legend(loc=4)
ax = plt.subplot(3,3,3)
ax.plot(np.abs(txframe_aligned[:128]),
label="TX Frame",
linestyle=":",
linewidth=0.5)
ax.plot(np.abs(rxframe_aligned[:128]),
label="RX Frame",
linestyle="--",
linewidth=0.5)
rx_abs = np.abs(rxframe_aligned)
rx_A = np.vstack([rx_abs,
rx_abs ** 3,
rx_abs ** 5,
rx_abs ** 7,
rx_abs ** 9,
]).T
rx_dpd = np.sum(rx_A * self.coefs_am, axis=1)
rx_dpd = rx_dpd * (
np.median(np.abs(tx_choice)) / np.median(np.abs(rx_dpd)))
ax.plot(np.abs(rx_dpd[:128]),
label="RX DPD Frame",
linestyle="-.",
linewidth=0.5)
tx_abs = np.abs(np.abs(txframe_aligned[:128]))
tx_A = np.vstack([tx_abs,
tx_abs ** 3,
tx_abs ** 5,
tx_abs ** 7,
tx_abs ** 9,
]).T
tx_dpd = np.sum(tx_A * new_coefs, axis=1)
tx_dpd_norm = tx_dpd * (
np.median(np.abs(tx_choice)) / np.median(np.abs(tx_dpd)))
ax.plot(np.abs(tx_dpd_norm[:128]),
label="TX DPD Frame Norm",
linestyle="-.",
linewidth=0.5)
ax.legend(loc=4)
ax.set_title("RX DPD")
ax.set_xlabel("Samples")
ax.set_ylabel("Amplitude")
ax = plt.subplot(3,3,4)
ax.scatter(
np.abs(tx_choice[:1024]),
np.abs(rx_choice[:1024]),
s=0.1)
ax.plot(rx_range_dpd / self.coefs_am[0], rx_range, linewidth=0.25)
ax.set_title("Amplifier Characteristic")
ax.set_xlabel("TX Amplitude")
ax.set_ylabel("RX Amplitude")
ax = plt.subplot(3,3,5)
ax.scatter(
np.abs(tx_choice[:1024]),
phase_diff_rad[:1024] * 180 / np.pi,
s=0.1
)
ax.plot(tx_range, phase_range_dpd * 180 / np.pi, linewidth=0.25)
ax.set_title("Amplifier Characteristic")
ax.set_xlabel("TX Amplitude")
ax.set_ylabel("Phase Difference [deg]")
ax = plt.subplot(3,3,6)
ccdf_min, ccdf_max = 0, 1
tx_hist, ccdf_edges = np.histogram(np.abs(txframe_aligned),
bins=60,
range=(ccdf_min, ccdf_max))
tx_hist_normalized = tx_hist.astype(float)/np.sum(tx_hist)
ccdf = 1.0 - np.cumsum(tx_hist_normalized)
ax.semilogy(ccdf_edges[:-1], ccdf, label="CCDF")
ax.semilogy(ccdf_edges[:-1],
tx_hist_normalized,
label="Histogram",
drawstyle='steps')
ax.legend(loc=4)
ax.set_ylim(1e-5,2)
ax.set_title("Complementary Cumulative Distribution Function")
ax.set_xlabel("TX Amplitude")
ax.set_ylabel("Ratio of Samples larger than x")
ax = plt.subplot(3,3,7)
coefs_history = np.array(self.coefs_history)
for idx, coef_hist in enumerate(coefs_history.T):
ax.plot(coef_hist,
label="Coef {}".format(idx),
linewidth=0.5)
ax.legend(loc=4)
ax.set_title("AM/AM Coefficient History")
ax.set_xlabel("Iterations")
ax.set_ylabel("Coefficient Value")
ax = plt.subplot(3,3,8)
coefs_history = np.array(self.coefs_pm_history)
for idx, coef_hist in enumerate(coefs_history.T):
ax.plot(coef_hist,
label="Coef {}".format(idx),
linewidth=0.5)
ax.legend(loc=4)
ax.set_title("AM/PM Coefficient History")
ax.set_xlabel("Iterations")
ax.set_ylabel("Coefficient Value")
ax = plt.subplot(3,3,9)
coefs_history = np.array(self.coefs_history)
ax.plot(self.mses, label="MSE")
ax.plot(self.errs, label="ERR")
ax.legend(loc=4)
ax.set_title("MSE History")
ax.set_xlabel("Iterations")
ax.set_ylabel("MSE")
fig.tight_layout()
fig.savefig(fig_path)
fig.clf()
self.coefs_am = new_coefs
self.coefs_history.append(self.coefs_am)
self.coefs_pm = new_coefs_pm
self.coefs_pm_history.append(self.coefs_pm)
return self.coefs_am, self.coefs_pm
# The MIT License (MIT)
#
# Copyright (c) 2017 Andreas Steger
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.