Skip to main content

Table 3 Parameter setting

From: A reinforcement learning-based sleep scheduling algorithm for compressive data gathering in wireless sensor networks

Parameters

Value

Monitoring area Z

2000 m × 1000 m

The number of nodes

N = 512

Node’s initial energy

E0 = 0.5 J

Nodes’ communication radius

R = 140 m

Length of a data packet

500 bits

Reference distance in free space

d0 = 88 m

Circuit energy consumption

Eelec = 50 nJ/bit

Amplifier energy consumption factor in free space (d < d0)

\(\varepsilon_{{{\text{fs}}}} = 10\;{\text{pJ/bit/m}}^{{2}}\)

Amplifier energy consumption factor in multi-path channel (d ≥ d0)

\(\varepsilon_{{{\text{mp}}}} = 0.0013\;{\text{pJ/bit/m}}^{4}\)

Discount rate

\(\gamma = 0.8\)

Learning rate

\(\alpha = 0.5\)

Explore probability

\(\varepsilon = 0.2\)

The number of CS measurements in each round

M = 130

The number of sampled nodes in a CS measurement

t = 20