def test_error(test_time):
# file_name = "D145_1"
test_file = "D51_4"
data = csv2data("positionData/{}.csv".format(test_file))
ideal = processDistance(test_file)
row, col = data.shape
print(data.shape)
times = int(test_time / row)
times = test_time
result_bucket = []
# for i in tqdm(range(times),ncols=100, desc="Progress"):
item = data[0,:]
for i in tqdm(range(times)):
on_signal = item[0:3]
off_signal = item[3:6]
noise_on_signal = add_noise(on_signal)
noise_off_signal = add_noise(off_signal)
noise_RSS = np.array([[RSS_cal(noise_on_signal, noise_off_signal)]])
noise_result = estimate_distacne(noise_RSS)
# print(noise_result)
# print("The result: {}".format(noise_result))
result_bucket.append(noise_result)
result_bucket = np.asarray(result_bucket)
print(result_bucket.size)
print(get_median(result_bucket))
draw_histogram(result_bucket, xlabel="Estimated Distance")
def positionDataProcess():
outdata = {}
data = csv2data("result/RSSdata.csv")
data = numpy_sort(data, 0)
row, col = data.shape
with open("observe.csv", "w") as ob:
for i in tqdm(range(row),ncols=100, desc="Progress"):
# for i in range(row):
# for i in range(row):
# print(data[i,:])
on_signal = data[i, 1:4]
off_signal = data[i, 4:7]
origin = np.array([[RSS_cal(on_signal, off_signal)]])
origin_result = estimate_distacne(origin)
# print("origin data: {}, data: {}".format(origin_result, data[i,0]))
# print("abs: {}".format(abs(origin_result - data[i,0])))
flag = abs(origin_result - data[i, 0])
if flag < 1:
# print("origin data: {}, data: {}".format(origin_result, data[i,0]))
# print("abs: {}".format(abs(origin_result - data[i,0])))
index = str(round(data[i, 0], 1))
ob.write("{},{},{},{}\n".format(index, origin_result, data[i,0], flag))
ob.write("{}\n".format(data[i, 1::]))
if index not in outdata:
outdata[index] = []
outdata[index].append(data[i, 1::].tolist())
else:
outdata[index].append(data[i, 1::].tolist())
for key, value in outdata.items():
key = key.replace(".", "_")
with open("positionData/D{}.csv".format(key), "w") as f:
f.write("Bxon,Byon,Bzon,Bxoff,Byoff,Bzoff\n")
for item in value:
f.write("{},{},{},{},{},{}\n".format(item[0], item[1], item[2], item[3], item[4], item[5]))
def cal_distribution(test_times, cal_function, write_file="DistributionAnalysis.csv"):
path_files = path_file_list()
with open(write_file, "w") as fi:
fi.write("Distance, Means, std, median\n")
for file_index in tqdm(range(len(path_files)), ncols=100, desc="Progress"):
# for file_index in tqdm(range(20), ncols=100, desc="Progress"):
# file_name = "D145_1"
data = csv2data("positionData/{}".format(path_files[file_index]))
# print("file: {}".format(path_files[file_index]))
# print("file: {}".format(path_files[file_index]))
ideal = processDistance(path_files[file_index])
row, col = data.shape
# times = int(test_time / row)
result_bucket = []
# for i in tqdm(range(times),ncols=100, desc="Progress"):
item = data[0,:]
for i in range(test_times):
on_signal = item[0:3]
off_signal = item[3:6]
noise_on_signal = add_noise(on_signal)
noise_off_signal = add_noise(off_signal)
noise_RSS = np.array([[RSS_cal(noise_on_signal, noise_off_signal)]])
noise_result = cal_function(noise_RSS)
# print("The result: {}".format(noise_result))
result_bucket.append(noise_result)
result_bucket = np.asarray(result_bucket)
fi.write("{},{},{},{}\n".format(ideal, result_bucket.mean(), result_bucket.std(), get_median(result_bucket)))
def test_distribution():
path_files = path_file_list()
for path_file in path_files:
# test_file = "D49_9.csv"
data = csv2data("positionData/{}".format(path_file))
distance = processDistance(path_file)
print("IN distance: {}".format(distance))
for item in data:
on_signal = item[0:3]
off_signal = item[3:6]
origin = np.array([[RSS_cal(on_signal, off_signal)]])
noise_on_signal = add_noise(on_signal)
noise_off_signal = add_noise(off_signal)
noise_RSS = np.array([[RSS_cal(noise_on_signal, noise_off_signal)]])
origin_result = estimate_distacne(origin)
noise_result = estimate_distacne(noise_RSS)
print("Origin : {}, Result: {}".format(origin_result,noise_result) )
def see_origin_distance(filename):
data = csv2data("positionData/{}.csv".format(filename))
ideal = processDistance(filename)
item = data[0,:]
on_signal = item[0:3]
off_signal = item[3:6]
RSS = np.array([[RSS_cal(on_signal, off_signal)]])
result = estimate_distacne(RSS)
print(result)
def test_actual(filename):
data = csv2data("positionData/{}.csv".format(filename))
data = numpy_sort(data, 0)
for item in data:
print(item)
on_signal = item[0:3]
off_signal = item[3:6]
RSS = np.array([[RSS_cal(on_signal, off_signal)]])
print(RSS)
distance_data = estimate_distacne(RSS)
print(distance_data)
def test_data():
test_file = "D50_5"
data = csv2data("positionData/{}.csv".format(test_file))
ideal = processDistance(test_file)
result_bucket = []
row, col = data.shape
# for i in tqdm(range(row)):
for item in data:
on_signal = item[0:3]
off_signal = item[3:6]
RSS = np.array([[RSS_cal(on_signal, off_signal)]])
distance_data = estimate_distacne(RSS)
result_bucket.append(distance_data)
result_bucket = np.asarray(result_bucket)
x_data = np.zeros(row)
x_data = x_data + ideal
plt.plot(x_data, result_bucket, "o")
plt.show()
def nonRandomDataProcess():
outdata = {}
data = csv2data("result/RSSdata.csv")
data = numpy_sort(data, 0)
row, col = data.shape
for i in tqdm(range(row),ncols=100, desc="Progress"):
on_signal = data[i, 1:4]
off_signal = data[i, 4:7]
origin = np.array([[RSS_cal(on_signal, off_signal)]])
origin_result = cal_distance_simple(origin)
index = str(round(data[i, 0], 1))
if index not in outdata:
outdata[index] = []
outdata[index].append(data[i, 1::].tolist())
else:
outdata[index].append(data[i, 1::].tolist())
for key, value in outdata.items():
key = key.replace(".", "_")
with open("positionData/D{}.csv".format(key), "w") as f:
f.write("Bxon,Byon,Bzon,Bxoff,Byoff,Bzoff\n")
for item in value:
f.write("{},{},{},{},{},{}\n".format(item[0], item[1], item[2], item[3], item[4], item[5]))