Hybrid Analog-Digital Precoder Design for Securing Cognitive Millimeter Wave Networks

Zhengmin Kong,Jing Song,Chao Wang,Hongyang Chen,Lajos Hanzo

DOI: https://doi.org/10.1109/tifs.2020.3039697

IF: 7.231

2021-01-01

IEEE Transactions on Information Forensics and Security

Abstract:Millimeter wave (mmWave) communications and cognitive radio technologies constitute key technologies of improving the spectral efficiency of communications. Hence, we conceive a hybrid secure precoder for enhancing the physical layer security of a cognitive mmWave wiretap channel, where a secondary transmitter broadcasts confidential information signals to multiple secondary users under the interference temperature constraint of the primary user (PU). The optimization problem is formulated as jointly optimizing the analog and digital precoder for maximizing the minimum secrecy rate of all the secondary users under practical constraints. In particular, our design satisfies the constraint on the maximum interference power received by multiple PUs, as well as the secondary users' minimum quality-of-service (Qos), and the unit-modulus constraint on the analog precoder. Due to the non-convexity of the resultant objective function and owing to the coupling between the analog and digital precoder, the optimization problem formulated is nonconvex and nonlinear, hence it is very challenging to solve directly. Hence, we first transform it into a tractable form, and develop a penalty dual decomposition (PDD) based iterative algorithm to locate its Karush-Kuhn-Tucker (KKT) solution. Finally, we generalize the proposed PDD algorithm to a secure hybrid precoder design relying on practical finite-resolution phase shifters and show that the proposed PDD algorithm can be straightforwardly adapted to handle the scenario, where each PU is equipped with multiple antennas and the CSI of multiple eavesdroppers (Eves) is imperfectly known. Our simulation results validate the efficiency of the proposed iterative algorithm.

computer science, theory & methods,engineering, electrical & electronic

What problem does this paper attempt to address?

This paper aims to solve the physical - layer security problems in millimeter - wave cognitive networks, especially to design a hybrid analog - digital precoder to enhance the physical - layer security of cognitive millimeter - wave eavesdropping channels. Specifically, the paper focuses on the secure transmission problem of the secondary transmitter broadcasting confidential information signals to multiple secondary users while satisfying the interference temperature constraints of primary users (PU). Its core objective is to maximize the minimum secrecy rate of all secondary users by jointly optimizing the analog and digital precoders under practical constraints.

### Main Problem Description
1. **Background and Challenges**:
- Millimeter - wave communication and cognitive radio technology are key technologies for improving communication spectrum efficiency.
- There are physical - layer security problems in cognitive millimeter - wave networks, and effective precoders need to be designed to protect communication security.
- The non - convexity of the optimization problem and the coupling effect between analog and digital precoders make direct solution very difficult.

2. **Optimization Objectives**:
- Maximize the minimum secrecy rate of all secondary users.
- Satisfy the maximum interference power constraints of multiple primary users.
- Satisfy the minimum quality - of - service (QoS) requirements of secondary users.
- Satisfy the unit - modulus constraints of the analog precoder.

3. **Methods and Innovations**:
- Introduce auxiliary variables to transform the original problem into a more tractable form.
- Use the Penalty Dual Decomposition (PDD) algorithm to solve the problems caused by coupling variables.
- Use the Block Coordinate Descent (BCD) algorithm and the Successive Convex Approximation (SCA) method to deal with non - convex optimization problems.
- The proposed algorithm can adapt to the practical scenario of finite - resolution phase shifters, and can handle the situation where each primary user is equipped with multiple antennas and the channel state information (CSI) of multiple eavesdroppers is not completely known.

### Mathematical Representation
- **Objective Function**:
\[
\text{maximize} \quad R_s=\min_{k \in K, m \in M}\left[R_k - R_{e,k,m}\right]
\]
where \(R_k\) is the transmission rate of the \(k\) - th secondary user, and \(R_{e,k,m}\) is the leakage information rate of the \(k\) - th secondary user received by the \(m\) - th eavesdropper.

- **Constraint Conditions**:
- Minimum quality - of - service constraint for secondary users:
\[
R_k\geq\gamma_k, \quad \forall k \in K
\]
- Maximum interference power constraint for primary users:
\[
\left\|t_n^H F_{\text{RF}} F_{\text{BB}}\right\|^2_F\leq\Gamma, \quad \forall n \in N
\]
- Total transmit power constraint:
\[
\left\|F_{\text{RF}} F_{\text{BB}}\right\|^2_F\leq P_{\text{tot}}
\]
- Unit - modulus constraint for the analog precoder:
\[
\left|F_{\text{RF}}(i,j)\right| = 1, \quad \forall i, j
\]

### Conclusion
By introducing the PDD algorithm, the paper successfully solves the non - convex optimization problem of hybrid precoder design in millimeter - wave cognitive networks and improves the physical - layer security of the system. The proposed algorithm not only has theoretical convergence guarantees but also shows significant performance advantages in practical applications.