Machine Learning-Based Resource Allocation Strategy for Network Slicing in Vehicular Networks

Algorithm 1

Resource allocation algorithm based on the primal-dual interior-point method.

Input: Initial , , scale factor , residual error , duality gap error ϵ, times observed load distribution

Output: Optimal solution

Phase 1: Predict service traffic

Training the ConvLSTM networks to obtain parameters , , , , , , , , , , ,

According to times observed load distribution predict sequences in the future

Phase 2: To obtain optimal slice weight

while True do

 Calculate initial value of the surrogate gap

 ifthen

  break

 end if

 Determine

 Compute primal-dual search direction

 Determine initial step length

 whiledo

  Ensure satisfy the constraint condition

 end while

 while

do

  Determine backtracking search step length

 end while

 Update search direction y

end while