step3_rules checking_allData.py

# -*- coding: utf-8 -*-
"""
Created on Tue Oct  8 17:12:12 2019

@author: jiayi
"""

#set environment to use GPU
import os
os.environ["THEANO_FLAGS"] = "device=gpu1"

from sklearn.model_selection import train_test_split
import copy
import matplotlib.pyplot as plt
from pandas import Series
from sklearn.metrics import f1_score, precision_score, recall_score
from sklearn.metrics import mean_squared_error
import pandas as pd
from sklearn import preprocessing
import numpy as np


def plot(t1,t2):
    
    x1 = np.arange(len(t1))
    x2 = np.arange(len(t2))
    
    plt.figure(1)
    
    plt.subplot(211) 
    plt.plot(x1, t1)
    
    plt.subplot(212)
    plt.plot(x2, t2)
    
    plt.show()




###############PARTS TO IMPORT DATA###############
#import the rules
df_rules = pd.read_excel("./checking rules/New_Rules.xlsx")
#import the attack data (only sensor and actuator value)
df_data = pd.read_csv("attack_x.csv")
#import the attack data label for later comparision
df_y_true = pd.read_csv("attack_p1_y.csv")



#################RULES CHECKING###################33
y = [0]*len(df_data)

#rule 0
RULE = 0
V1 = 'LIT101'
V1_judge = 500
V2 = "MV101"
V2_judge = 2

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))

#rule 1
RULE = 1
V1 = 'LIT101'
V1_judge = 800
V2 = "MV101"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 2
RULE = 2
V1 = 'LIT101'
V1_judge = 250
V2 = "P101"
V2_judge = 1
V3 = "P102"
V3_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 3
RULE = 3
V1 = 'LIT301'
V1_judge = 800
V2 = "P101"
V2_judge = 2
V3 = "P102"
V3_judge = 2

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge and df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 4
RULE = 4
V1 = 'LIT301'
V1_judge = 1000
V2 = "P101"
V2_judge = 1
V3 = "P102"
V3_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 5
RULE = 5
V1 = 'LIT301'
V1_judge = 800
V2 = "MV201"
V2_judge = 2

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))

#rule 6
RULE = 6
V1 = 'LIT301'
V1_judge = 1000
V2 = "MV201"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 7
RULE = 7
V1 = 'FIT201'
V1_judge = 0.5
V2 = "P201"
V2_judge = 1
V3 = "P202"
V3_judge = 1
V4 = "P204"
V4_judge = 1
V5 = "P206"
V5_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] < V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge or df_data[V4].iloc[i] != V4_judge or df_data[V5].iloc[i] != V5_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 8
RULE = 8
V1 = 'AIT201'
V1_judge = 260
V2 = "FIT201"
V2_judge = 0.5
V3 = "P201"
V3_judge = 1
V4 = "P202"
V4_judge = 1


flag = 0
r1_idx = np.where(df_data[V1] > V1_judge)[0]
r1_idx_and = np.where(df_data[V2].iloc[r1_idx] > V2_judge)[0]
r1_idx_t = r1_idx_and + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V3].iloc[i] != V3_judge or df_data[V4].iloc[i] != V4_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 9
RULE = 9
V1 = 'AIT503'
V1_judge = 260
V2 = "FIT201"
V2_judge = 0.5
V3 = "P201"
V3_judge = 1
V4 = "P202"
V4_judge = 1


flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_and = np.where(df_data[V2].iloc[r1_idx] > V2_judge)[0]
r1_idx_t = r1_idx_and + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V3].iloc[i] != V3_judge or df_data[V4].iloc[i] != V4_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 10
RULE = 10
V1 = 'AIT202'
V1_judge = 6.95
V2 = "P203"
V2_judge = 1
V3 = "P204"
V3_judge = 1



flag = 0
r1_idx = np.where(df_data[V1] < V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 11
RULE = 11
V1 = 'AIT203'
V1_judge = 500
V2 = "P205"
V2_judge = 1
V3 = "P206"
V3_judge = 1



flag = 0
r1_idx = np.where(df_data[V1] > V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 12
RULE = 12
V1 = 'AIT203'
V1_judge = 420
V2 = "FIT201"
V2_judge = 0.5
V3 = "AIT402"
V3_judge = 250
V4 = "P205"
V4_judge = 2
V5 = "P206"
V5_judge = 2

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_and = np.where(df_data[V2].iloc[r1_idx] > V2_judge)[0]
r1_idx_and2 = np.where(df_data[V3].iloc[r1_idx] > V3_judge)[0]
r1_idx_t = r1_idx_and2 + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V4].iloc[i] != V4_judge and df_data[V5].iloc[i] != V5_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))




#rule 13
RULE = 13
V1 = 'AIT402'
V1_judge = 250
V2 = "P205"
V2_judge = 1
V3 = "P206"
V3_judge = 1



flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))





#rule 14
RULE = 14
V1 = 'LIT301'
V1_judge = 250
V2 = "P301"
V2_judge = 1
V3 = "P302"
V3_judge = 1



flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))




#rule 15
RULE = 15
V1 = 'LIT401'
V1_judge = 1000
V2 = "P301"
V2_judge = 1
V3 = "P302"
V3_judge = 1



flag = 0
r1_idx = np.where(df_data[V1] >= V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 16
RULE = 16
V1 = 'LIT401'
V1_judge = 250
V2 = "P401"
V2_judge = 1
V3 = "P402"
V3_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 17
RULE = 17
V1 = 'LIT401'
V1_judge = 250
V2 = "UV401"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] <= V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 18
RULE = 18
V1 = 'FIT401'
V1_judge = 0.5
V2 = "UV401"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] < V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if i<len(df_data):
        if df_data[V2].iloc[i] != V2_judge:
            y[i] = 1
            flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 19
RULE = 19
V1 = 'AIT402'
V1_judge = 240
V2 = "P403"
V2_judge = 1
V3 = "P404"
V3_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] < V1_judge)[0]
r1_idx_t = r1_idx + df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if i<len(df_data):
        if df_data[V2].iloc[i] != V2_judge or df_data[V3].iloc[i] != V3_judge:
            y[i] = 1
            flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))


#rule 20
RULE = 20
V1 = 'UV401'
V1_judge = 1
V2 = "P501"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] == V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 21
RULE = 21
V1 = 'FIT401'
V1_judge = 0.5
V2 = "P501"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] < V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



#rule 22
RULE = 22
V1 = 'LIT101'
V1_judge = 1100
V2 = "P601"
V2_judge = 1

flag = 0
r1_idx = np.where(df_data[V1] > V1_judge)[0]
r1_idx_t = r1_idx+df_rules["Time"].iloc[RULE]
for i in r1_idx_t:
    if df_data[V2].iloc[i] != V2_judge:
        y[i] = 1
        flag = 1

if flag == 1:
    print("Find attacks by RULE"+str(RULE))



df_y = pd.DataFrame(y[:4000])
df_y.to_csv('invariants check_Y.csv',index = False)

plot(df_y,df_y_true)
print('checking error...')
f1 = f1_score(df_y_true,df_y,average='binary')
precision = precision_score(df_y_true,df_y,average='binary')
recall = recall_score(df_y_true,df_y,average='binary')
print('Precision:',precision)
print('recall:', recall)
print('f1:',f1)