Anti-Attack Scheme for Edge Devices Based on Deep Reinforcement Learning

Table 1

Sample dataset.

Number	Data 1	Data 2	Data 3	Data 4	Weight
1	22.44	21.95	21.96	15.82	0.099066
2	23.01	21.94	22.67	16.69	0.08585
3	23.1	22.4	22.4	16.43	0.082904
4	23.2	21.7	21.88	15.84	0.07981
5	21.51	20.85	21.44	15.66	0.078681
6	22.08	21.3	21.45	16.2	0.073956
7	21.86	21.15	21.79	15.97	0.07361
8	21.39	20.77	21.11	16.65	0.07244
9	21.43	20.85	21.23	16.45	0.067382
10	21.48	20.96	21.25	15.89	0.051061
11	21.19	20.8	21.1	16.56	0.046455
12	22	21.24	22	16.54	0.045742
13	22.47	21.5	21.64	16.28	0.03978
14	21.6	20.74	20.81	16.44	0.03306
15	20.48	19.51	19.88	16.11	0.028871
16	20.19	19.75	19.8	15.9	0.019483
17	20.06	18.16	18.59	15.1	0.010127
18	18.77	18.06	18.31	14.51	0.004814
19	18.35	17.6	18.2	14.06	0.003591
20	18.41	17.73	17.99	13.88	0.003319