Otherwise, when the data is offloaded for edge execution (xti = 1), we denote Pit as the transmit power constrained by the maximum power Pt ≤ Pmax and τtT as the amount of time iii allocated to the ith WD for computation offloading. Here, τit ∈ [0, 1] and 􏰏Ni=1 τit ≤ 1. The energy consumed on data offloading is Eit,O = PitτitT. Similar to [4] and [8], we neglect the delay on edge computing and result downloading such that the amount of data processed at the edge within the time frame is Dt =WτitTlog 􏰉1+Pithti􏰊=WτitTlog 􏰉1+Eit,Ohti􏰊, ∀xt =1, (2) i,O vu 2 N0 vu 2 τitTN0 i where vu ≥ 1 denotes the communication overhead and N0 denotes the noise power. Let Dit 􏰗 (1 − xti )Dit,L + xti Dit,O and Eit 􏰗 (1 − xti )Eit,L + xti Eit,O denote the bits computed and energy consumed in time frame t. We define computation rate rit and power consumption eti in 9 the tth time frame as ttttttt rt=Di=(1−xi)fi+xtWτilog􏰉1+ei,Ohi􏰊,et=Ei=(1−xt)κ􏰁ft􏰂3+xtet ,(3) i T φ ivu 2 τitN0 i T i i ii,O where eti,O 􏰗 Eit,O/T. For simplicity of exposition, we assume T = 1 without loss of generality in the following derivations.，解释cit

这段文字描述了一种边缘计算下的功耗和数据处理速率模型。其中，Pit表示限制最大功率Pt≤Pmax的传输功率，τtT表示分配给第i个终端设备的计算离线化时间。τit∈[0,1]且∑iτit≤1。数据离线化时的能量消耗为Eit,O=PitτitT。忽略边缘计算和结果下载的延迟，数据在时间框架内在边缘处理的数量为Dt=WτitTlog(1+Pithti)=WτitTlog(1+Eit,Ohti)，当xt=1时。Dit和Eit分别表示在时间框架t内处理的比特数和所消耗的能量。rit和eti分别表示在第t个时间框架中的计算速率和功耗。其中，et i,O = Eit,O/T。整个模型的目的是为了简化边缘计算下的计算速率和能量消耗的计算。