Abstract:Nodes in contemporary radio networks often have multiple interfaces available for communication: WiFi, cellular, LoRa, Zigbee, etc. This motivates understanding both link and network configuration when multiple communication modalities with vastly different capabilities are available to each node. In conjunction, covertness or the hiding of radio communications is often a significant concern in both commercial and military wireless networks. We consider the optimal routing problem in wireless networks when nodes have multiple interfaces available and intend to hide the presence of the transmission from attentive and capable adversaries. We first consider the maximization of the route capacity given an end-to-end covertness constraint against a single adversary and we find a polynomial-time algorithm for optimal route selection and link configuration. We further provide optimal polynomial-time algorithms for two important extensions: (i) statistical uncertainty during optimization about the channel state information for channels from system nodes to the adversary; and, (ii) maintaining covertness against multiple adversaries. Numerical results are included to demonstrate the gains of employing heterogeneous radio resources and to compare the performance of the proposed approach versus alternatives.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
This paper aims to solve the problem of how to optimize routing and link configurations to achieve covert communication in multimodal wireless networks in the presence of multiple communication interfaces (such as WiFi, cellular, LoRa, Zigbee, etc.). Specifically, the paper focuses on maximizing the transmission capacity of wireless networks while ensuring end - to - end covertness.
#### Main problems
1. **Optimal routing selection for multimodal nodes**: When each node can use multiple different wireless communication technologies, how to select the optimal routing path to maximize the end - to - end transmission capacity.
2. **Optimization under covertness constraints**: In the presence of adversaries, how to ensure that the existence of communication is not detected. This involves the concept of covert communication, that is, not only protecting the content of the message, but also hiding the existence of the message.
3. **Handling multiple adversaries**: When there are multiple adversaries, how to optimize routing and link configurations so that the communication still remains covert.
4. **Uncertainty of channel state information**: How to optimize when there is statistical uncertainty in channel state information (CSI).
#### Solutions
- **Case of a single adversary**: A polynomial - time algorithm is proposed to select the optimal routing and link configurations to ensure maximizing the transmission capacity under the given covertness constraints.
- **Case of multiple adversaries**: The model is extended to consider the case of joint observations by multiple adversaries, and the corresponding optimization algorithm is provided.
- **Case of uncertain channel state information**: Statistical uncertainty is introduced, and the reliability of covert communication is guaranteed through the average covertness constraint.
### Presentation of formulas
The key formulas involved in the paper include:
1. **Covertness constraint**:
\[
D(Q_0 \| Q_1) \leq \epsilon
\]
where \(D(Q_0 \| Q_1)\) is the Kullback - Leibler divergence between two probability distributions, and \(\epsilon\) is the covertness level.
2. **Single - link capacity optimization**:
\[
\max_{P_{S_i,1}, P_{S_i,2}, \ldots, P_{S_i,M}} \sum_{m = 1}^M C_m
\]
Subject to the constraint:
\[
D(P^{(i)}_0 \| P^{(i)}_1) \leq \delta_i
\]
3. **Mathematical expression of covertness constraint**:
\[
\sum_{m = 1}^M \left( \frac{g^2(S_i, W_k)_m P_{S_i,m}}{\sigma^2_{W_k,m} d^\alpha_{S_i,W_k}} \right)^2 \leq \delta_i
\]
4. **Solution by Lagrange multiplier method**:
\[
L((P_{S_i,1}, P_{S_i,2}, \ldots, P_{S_i,M}), \lambda) = \sum_{m = 1}^M \frac{g^2(S_i, D_i)_m P_{S_i,m}}{2 \sigma^2_{D_i,m} d^\alpha_{S_i,D_i}} + \lambda \left( \delta_i - \sum_{m = 1}^M \left( \frac{g^2(S_i, W_k)_m P_{S_i,m}}{\sigma^2_{W_k,m} d^\alpha_{S_i,W_k}} \right)^2 \right)
\]
5. **Optimal power allocation**:
\[
P_{S_i,j} = \sqrt{\frac{\delta_i}{\Gamma_i}} \frac{d^{2\alpha}_{S_i,W_k} d^\alpha_{S_i,D_i} g^2(S_i, D_i)_j \