Table 14 Computational cost across different network scales.

From: Design of a dynamic trust management and defense decision system for shared vehicle data based on blockchain and deep reinforcement learning

Network scale (Nodes)

Block time (s)

CPU utilization

Memory usage (GB)

Inference latency (ms)

10,000

0.8

15%

2.1

42

100,000

1.2

28%

4.8

45

500,000

1.9

32%

9.6

48

  1. To mitigate data poisoning and adversarial sample attacks targeting the DRL model, a three-layer defense framework is implemented:
Back to article page