def test_001_t (self):
# set up fg
src_data = (0,0,0,0,1,-1j,-1,1j,0,0,0,0,
1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,
0,0,0,0,2,-2j,-2,2j,0,0,0,0,
2,2,2,2,4,4,4,4,6,6,6,6,8,8,8,8)
expected_result = (0,0,0,0,1,-1j,-1,1j,0,0,0,0,
1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,
0,0,0,0,2,-2j,-2,2j,0,0,0,0,
1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4)
src = blocks.vector_source_c(src_data,vlen=4)
scp = ofdm.fbmc_subchannel_processing_vcvc(M=4,syms_per_frame=2,preamble=[0,0,0,0,1,-1j,-1,1j,0,0,0,0],sel_eq=0)
dst = blocks.vector_sink_c(vlen=4)
dst2 = blocks.vector_sink_c(vlen=4)
self.tb.connect((src,0),(scp,0))
self.tb.connect((scp,0),dst)
self.tb.connect((scp,1),dst2)
self.tb.run ()
# check data
result_data = dst.data()
result_data2 = dst2.data()
for i in range(len(result_data2)):
print str(i)+"\t"+str(result_data2[i])
# print result_data
self.assertComplexTuplesAlmostEqual(expected_result,result_data,6)
def test_002_t(self):
# set up fg
src_data = (0, 0, 0, 0, 1, -1j, -1, 1j, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2,
2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 0, 0, 0, 0, 2, -2j, -2, 2j,
0, 0, 0, 0, 2, 2, 2, 2, 4, 4, 4, 4, 6, 6, 6, 6, 8, 8, 8, 8)
expected_result = (0, 0, 0, 0, 1, -1j, -1, 1j, 0, 0, 0, 0, 1, 1, 1, 1,
2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 0, 0, 0, 0, 2,
-2j, -2, 2j, 0, 0, 0, 0, 2, 2, 2, 2, 4, 4, 4, 4, 6,
6, 6, 6, 8, 8, 8, 8)
src = blocks.vector_source_c(src_data, vlen=4)
scp = ofdm.fbmc_subchannel_processing_vcvc(
M=4,
syms_per_frame=2,
preamble=[0, 0, 0, 0, 1, -1j, -1, 1j, 0, 0, 0, 0],
sel_eq=3)
dst = blocks.vector_sink_c(vlen=4)
dst2 = blocks.vector_sink_c(vlen=4)
self.tb.connect((src, 0), (scp, 0))
self.tb.connect((scp, 0), dst)
self.tb.connect((scp, 1), dst2)
self.tb.run()
# check data
result_data = dst.data()
# print result_data
self.assertComplexTuplesAlmostEqual(expected_result, result_data, 6)
def test_009_t(self):
# test estimation with a different size of preamble
M = 2**10
syms_per_frame = 20
num_frames = 2**8
root_center = [1, -1j, -1, 1j]
src_data = list()
expected_result = list()
for i in range(2 * syms_per_frame * M * num_frames):
if i % (2 * syms_per_frame * M) == 0:
mul = [int(10 * random.random()) + 1 for j in range(M)]
center_preamble = [
mul[j] * root_center[j % 4] for j in range(M)
]
preamble = [0] * 5 * M + center_preamble * (
M / len(center_preamble)) + [0] * 5 * M
# add preamble
src_data.extend(preamble)
expected_result.extend(preamble)
val = int(random.random() * 10)
src_data.append(mul[i % M] * val)
expected_result.append(val)
# get original preamble back
preamble = [0] * 5 * M + root_center * (M / 4) + [0] * 5 * M
src = blocks.vector_source_c(src_data, vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(
M=M, syms_per_frame=syms_per_frame, preamble=preamble, sel_eq=0)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src, 0), (scp, 0))
self.tb.connect((scp, 0), dst)
self.tb.connect((scp, 1), dst2)
self.tb.run()
# check data
result_data = dst.data()
f = open('..\..\matlab\sp_python_output.txt', 'w')
s = "i\tfr\tsrc\texp\tres\tdiff\n"
f.write(s)
for i in range(2 * syms_per_frame * M * num_frames):
s = str(i) + "\t" + str(
int(i / (2 * syms_per_frame * M))
) + "\t" + str(src_data[i]) + "\t" + str(
expected_result[i]) + "\t" + str(result_data[i]) + "\t" + str(
abs(expected_result[i] - result_data[i]) > 0) + "\n"
f.write(s)
f.close()
# frame based check
for i in range(num_frames):
self.assertComplexTuplesAlmostEqual(
tuple(expected_result[(i + 1) * len(preamble) + i *
(2 * syms_per_frame * M):(i + 1) *
len(preamble) + (i + 1) *
(2 * syms_per_frame * M)]),
tuple(result_data[(i + 1) * len(preamble) + i *
(2 * syms_per_frame * M):(i + 1) *
len(preamble) + (i + 1) *
(2 * syms_per_frame * M)]), 6)
def test_007_t(self):
# test with three-tap equalizer with geometric interpolation
##############################
# this test case should be used along with MATLAB script data_generator_for_qa_scp.m
# you can find that script in gr-fbmc/matlab/
# remember to set eq_select = 2;
# first create src_data and expected_result sequences as well as parameters (M,syms_per_frame,num_frames)
# the parameters will be created in the same folder.
# after generation of required data, you can run this test case.
# with help of scipy libraries, those parameters will be fetched and computed.
##############################
mat_src = scipy.io.loadmat('../../../matlab/src_data_geometric.mat')
mat_exp = scipy.io.loadmat(
'../../../matlab/expected_result_geometric.mat')
mat_par = scipy.io.loadmat('../../../matlab/parameters_geometric.mat')
M = int(mat_par['parameters']['M'][0][0][0][0])
syms_per_frame = int(
mat_par['parameters']['syms_per_frame'][0][0][0][0])
num_frames = int(mat_par['parameters']['num_frames'][0][0][0][0])
# root_center = list()
# for i in range(len(mat_par['parameters']['center'][0][0])):
# root_center.append(mat_par['parameters']['center'][0][0][0][i])
root_center = [1, -1j, -1, 1j] * (M / 4)
src_data = list()
expected_result = list()
for i in range(len(mat_src['src_data'])):
src_data.append(mat_src['src_data'][i][0])
for i in range(len(mat_exp['expected_result'])):
expected_result.append(mat_exp['expected_result'][i][0])
# print src_data
preamble = [0] * M + root_center + [0] * M
src = blocks.vector_source_c(src_data, vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(
M=M, syms_per_frame=syms_per_frame, preamble=preamble, sel_eq=2)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src, 0), (scp, 0))
self.tb.connect((scp, 0), dst)
self.tb.connect((scp, 1), dst2)
self.tb.run()
# check data
result_data = dst.data()
# print len(result_data)
self.assertComplexTuplesAlmostEqual(tuple(expected_result),
tuple(result_data), 4)
def test_009_t(self):
# test estimation with a different size of preamble
M = 2**10
syms_per_frame = 20
num_frames = 2**8
root_center = [1,-1j,-1,1j]
src_data = list()
expected_result =list()
for i in range(2*syms_per_frame*M*num_frames):
if i%(2*syms_per_frame*M)==0:
mul = [int(10*random.random())+1 for j in range(M)]
center_preamble = [mul[j]*root_center[j%4] for j in range(M)]
preamble = [0]*5*M+center_preamble*(M/len(center_preamble))+[0]*5*M
# add preamble
src_data.extend(preamble)
expected_result.extend(preamble)
val = int(random.random()*10)
src_data.append(mul[i%M]*val)
expected_result.append(val)
# get original preamble back
preamble = [0]*5*M+root_center*(M/4)+[0]*5*M
src = blocks.vector_source_c(src_data,vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(M=M,syms_per_frame=syms_per_frame,preamble=preamble,sel_eq=0)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src,0),(scp,0))
self.tb.connect((scp,0),dst)
self.tb.connect((scp,1),dst2)
self.tb.run ()
# check data
result_data = dst.data()
f = open('..\..\matlab\sp_python_output.txt', 'w')
s = "i\tfr\tsrc\texp\tres\tdiff\n"
f.write(s)
for i in range(2*syms_per_frame*M*num_frames):
s = str(i)+"\t"+str(int(i/(2*syms_per_frame*M)))+"\t"+str(src_data[i])+"\t"+str(expected_result[i])+"\t"+str(result_data[i])+"\t"+str(abs(expected_result[i]-result_data[i])>0)+"\n"
f.write(s)
f.close()
# frame based check
for i in range(num_frames):
self.assertComplexTuplesAlmostEqual(tuple(
expected_result[(i+1)*len(preamble)+i*(2*syms_per_frame*M):(i+1)*len(preamble)+(i+1)*(2*syms_per_frame*M)]),
tuple(
result_data[(i+1)*len(preamble)+i*(2*syms_per_frame*M):(i+1)*len(preamble)+(i+1)*(2*syms_per_frame*M)]),6)
def test_007_t(self):
# test with three-tap equalizer with geometric interpolation
##############################
# this test case should be used along with MATLAB script data_generator_for_qa_scp.m
# you can find that script in gr-fbmc/matlab/
# remember to set eq_select = 2;
# first create src_data and expected_result sequences as well as parameters (M,syms_per_frame,num_frames)
# the parameters will be created in the same folder.
# after generation of required data, you can run this test case.
# with help of scipy libraries, those parameters will be fetched and computed.
##############################
mat_src = scipy.io.loadmat('../../../matlab/src_data_geometric.mat')
mat_exp = scipy.io.loadmat('../../../matlab/expected_result_geometric.mat')
mat_par = scipy.io.loadmat('../../../matlab/parameters_geometric.mat')
M = int(mat_par['parameters']['M'][0][0][0][0])
syms_per_frame = int(mat_par['parameters']['syms_per_frame'][0][0][0][0])
num_frames = int(mat_par['parameters']['num_frames'][0][0][0][0])
# root_center = list()
# for i in range(len(mat_par['parameters']['center'][0][0])):
# root_center.append(mat_par['parameters']['center'][0][0][0][i])
root_center = [1,-1j,-1,1j]*(M/4)
src_data = list()
expected_result =list()
for i in range(len(mat_src['src_data'])):
src_data.append(mat_src['src_data'][i][0])
for i in range(len(mat_exp['expected_result'])):
expected_result.append(mat_exp['expected_result'][i][0])
# print src_data
preamble = [0]*M+root_center+[0]*M
src = blocks.vector_source_c(src_data,vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(M=M,syms_per_frame=syms_per_frame,preamble=preamble,sel_eq=2)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src,0),(scp,0))
self.tb.connect((scp,0),dst)
self.tb.connect((scp,1),dst2)
self.tb.run ()
# check data
result_data = dst.data()
# print len(result_data)
self.assertComplexTuplesAlmostEqual(tuple(expected_result),tuple(result_data),4)
def test_003_t(self):
# test no equalizer
center_preamble = [1, -1j, -1, 1j]
M = 2**5
syms_per_frame = 20
num_frames = 2**5
preamble = [0] * M + center_preamble * (M / 4) + [0] * M
src_data = list()
expected_result = list()
for i in range(2 * syms_per_frame * M * num_frames):
if i % (2 * syms_per_frame * M) == 0:
src_data.extend(preamble)
expected_result.extend(preamble)
val = int(random.random() * 10)
src_data.append(val)
expected_result.append(val)
src = blocks.vector_source_c(src_data, vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(
M=M, syms_per_frame=syms_per_frame, preamble=preamble, sel_eq=3)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src, 0), (scp, 0))
self.tb.connect((scp, 0), dst)
self.tb.connect((scp, 1), dst2)
self.tb.run()
# check data
result_data = dst.data()
f = open('..\..\matlab\sp_python_output.txt', 'w')
s = "i\tsrc\texp\tres\tdiff\n"
f.write(s)
for i in range(2 * syms_per_frame * M * num_frames):
s = str(i) + "\t" + str(src_data[i]) + "\t" + str(
expected_result[i]) + "\t" + str(result_data[i]) + "\t" + str(
abs(expected_result[i] - result_data[i]) > 0) + "\n"
f.write(s)
f.close()
# for i in range(len(result_data)):
# print str(i)+"\t"+str(expected_result[i])+"\t"+str(result_data[i])
# print result_data
self.assertComplexTuplesAlmostEqual(expected_result, result_data, 6)
def test_003_t(self):
# test no equalizer
center_preamble = [1,-1j,-1,1j]
M = 2**5
syms_per_frame = 20
num_frames = 2**5
preamble = [0]*M+center_preamble*(M/4)+[0]*M
src_data = list()
expected_result =list()
for i in range(2*syms_per_frame*M*num_frames):
if i%(2*syms_per_frame*M)==0:
src_data.extend(preamble)
expected_result.extend(preamble)
val = int(random.random()*10)
src_data.append(val)
expected_result.append(val)
src = blocks.vector_source_c(src_data,vlen=M)
scp = ofdm.fbmc_subchannel_processing_vcvc(M=M,syms_per_frame=syms_per_frame,preamble=preamble,sel_eq=3)
dst = blocks.vector_sink_c(vlen=M)
dst2 = blocks.vector_sink_c(vlen=M)
self.tb.connect((src,0),(scp,0))
self.tb.connect((scp,0),dst)
self.tb.connect((scp,1),dst2)
self.tb.run ()
# check data
result_data = dst.data()
f = open('..\..\matlab\sp_python_output.txt', 'w')
s = "i\tsrc\texp\tres\tdiff\n"
f.write(s)
for i in range(2*syms_per_frame*M*num_frames):
s = str(i)+"\t"+str(src_data[i])+"\t"+str(expected_result[i])+"\t"+str(result_data[i])+"\t"+str(abs(expected_result[i]-result_data[i])>0)+"\n"
f.write(s)
f.close()
# for i in range(len(result_data)):
# print str(i)+"\t"+str(expected_result[i])+"\t"+str(result_data[i])
# print result_data
self.assertComplexTuplesAlmostEqual(expected_result,result_data,6)
