Table 2 Commonly used notations and variables.
Notation | Description | Notation | Description |
|---|---|---|---|
\(\mathcalligra{\scriptstyle K} \hspace{000.1cm}, \mathcalligra{\scriptstyle L} \hspace{000.1cm}, \mathcalligra{\scriptstyle W} \hspace{000.1cm},\mathcalligra{\scriptstyle U} \hspace{000.1cm}, \mathcalligra{\scriptstyle G} \hspace{000.1cm}\) | The set of BSs, FBSs, WAPs, UBS, and GDs. | X, Y, H | The UAVs’ X-axis, Y-axis, and altitude. |
K, L, U, N | The number of BSs, LTE-BSs, UBSs, and GDs. | \(P^w,P_{i,k}^u\) | Wi-Fi card, and average uplink power consumption. |
\(N^K, N_k^L,\) \(N_k^W\) | The number of associated GDs with MBS, FBS, and WAP k. | \(R_i^u,\Gamma ^u\) | GD’s uplink average traffic generation rate, and target SNR. |
\(P_k\left( V_k\right) ,E_k\) | UAV’s power consumption and energy consumption. | \(\mathcalligra{\scriptstyle S} \hspace{000.1cm},\mathbb {A},\mathbb {O}\) | Set of the state, joint action, and joint observation spaces. |
s(t), a(t), r(t) | The state, action, and reward at time t. | \(\gamma ,\pi ,\pi ^*\) | Discount factor, UAV’s policy, and optimal policy. |
\(\theta , \theta ^-\) | \(\mathcalligra{\scriptstyle Q} \hspace{000.1cm}\)-network, and target network weights. | \(\mathcalligra{\scriptstyle Q} \hspace{000.1cm}^\pi (s,a)\) | The UAV’s state-action value function. |
\(M^{ep}\) | The number of episodes of the Q-learning algorithm. | \(\hspace{000.2cm} \mathcalligra{\scriptstyle J} \hspace{000.2cm}\) | The regret-matching game. |
\(R_{ik}^d,R_{ik}^{WPHY},S_i\) | The average downlink data rate, downlink WLAN physical data rate, GD’s satisfaction. | \(D_i^t\left( m_i,m_i^\prime \right)\) | The payoff for GD i if it had played action \(m_i\) instead of \(m_i^\prime\). |
\(T_{SC}^{LTE}\) | The duration of an LTE subframe. | \(\psi _i^{t+1}\left( m_i\right)\) | The probability distribution of GD i choosing an action at time t. |
\(C_{i,k}^{LMCS}, C_{i,k}^{WMCS}\) | The coding rate of LTE BSs and WAPs. | \({\bar{z}}_t\) | The empirical distribution of joint actions \(\Sigma\) of all GDs until t. |
A | Association matrix between GDs and base stations. | \(\Sigma ^*\) | Optimal joint strategies. |
