We first consider the download delay for UE j when acquiring file k through GAP i.
Denote the maximum cache capacity of GAP i as (S_{i}^{max }), 1≤i≤M.
Let x ij ∈{0,1} denote the binary association variable between GAP i and UE j, i.e., x ij =1, if UE j is associated with GAP i; otherwise, x ij =0.
where ({sigma _{ij}^{2}}) denotes the noise power of the link between UE j and GAP i.
Let ({W_{i}^{max }}) denote the available bandwidth of GAP i and W i denote the bandwidth of each subchannel of GAP i, the maximum number of users associated to GAP i can be calculated as (A_{i}=left lfloor frac {W_{i}^{max }}{W_{i}}right rfloor ).
Denote the binary content placement variable at the GAPs as δ ik ∈{0,1}, i.e., δ ik =1, if GAP i caches file k; otherwise, δ ik =0.
Denote the maximum cache capacity of GAP i as (S_{i}^{max }), we obtain the cache capacity constraint: {S_{i}} le S_{i}^{max}.
where κ il, a il, and ϖ il are constants, representing the processing capability of the lth hop node in the backhaul link of GAP i.
(2) (3) where = marginal LD for adjacent locus gap i; = observed LD in the tested generation for adjacent locus gap i; = observed LD in the control generation for adjacent locus gap i, and r = number of adjacent locus gaps in the Linkage Group.
All of the locally frequent symbol a are appended to Ψ with a gap i to constitute a longer pattern: Ψ ' = Ψ-x(i)- a.
The state of a CSPBN can be represented as a combination of a context and a GAP, i.e., S = (context j, GAP i).
