Skip to main content

Table 8 Performance of each algorithm on training data

From: Performance evaluation of deep learning techniques for DoS attacks detection in wireless sensor network

Algorithm

Attack type

Precision, %

Recall, %

F1 Score, %

Accuracy, %

DNN

Normal

99.64

99.19

99.41

–

Blackhole

64.06

83.78

72.61

–

Grayhole

69.56

67.72

68.63

–

TDMA

97.56

86.93

91.94

–

Flooding

88.57

77.50

82.67

–

Overall

83.88

83.02

83.05

97.07

CNN

Normal

99.23

99.78

99.50

–

Blackhole

94.67

95.12

94.89

–

Grayhole

91.77

83.65

87.52

–

TDMA

99.06

87.23

92.77

–

Flooding

90.40

96.47

93.33

–

Overall

95.03

92.45

93.60

98.75

RNN

Normal

98.25

99.91

99.07

–

Blackhole

81.06

47.56

59.94

–

Grayhole

66.02

72.82

69.25

–

TDMA

97.30

76.46

85.63

–

Flooding

89.77

51.97

65.83

–

Overall

86.48

69.74

75.94

96.50

CNN+RNN

Normal

99.57

99.27

99.42

–

Blackhole

53.13

99.77

69.34

–

Grayhole

85.53

43.63

57.79

–

TDMA

100.00

83.66

91.10

–

Flooding

87.62

96.17

91.69

–

Overall

85.17

84.50

81.87

96.84