1. How does mmWave-MEC with NOMA improve computation efficiency?
The paper explores the optimization of computation efficiency (CE) in mmWave-MEC networks utilizing non-orthogonal multiple access (NOMA) and millimeter-wave (mmWave) communications. By considering both analog beamforming (ABF) and hybrid beamforming (HBF) architectures in partial offloading mode, the study formulates the CE optimization problem based on the max-min fairness criterion. The proposed algorithm, using penalized successive convex approximation, efficiently solves the non-convex problem for ABF. Similarly, the max-min CE optimization problem for HBF is addressed using a penalty function and inexact block coordinate descent method. Simulation results demonstrate the convergence of the proposed algorithms and show that the proposed resource allocation schemes significantly enhance system CE. The study also highlights the superior performance of NOMA over conventional orthogonal multiple access schemes in terms of CE.
read more
2. How can the max-min fairness criterion be used to improve CE in mmWave-MEC-ABF and ensure user fairness?
The max-min fairness criterion can be used to improve the Channel Efficiency (CE) of the mmWave-Multi-Cell-Multi-Element-Based-Fractional-Access (mmWave-MEC-ABF) system and ensure user fairness. By formulating the max-min CE optimization problem, it can be transformed into a non-smooth and non-convex fractional optimization problem with 2(N + K) real optimization variables. However, due to the large number of variables, directly searching for the global optimal solution is computationally complex. To address this, an efficient CE optimization algorithm with polynomial computation complexity is designed to find a suboptimal solution. The max-min fairness criterion ensures that the allocation of resources is fair among users, preventing any user from being unfairly disadvantaged. This criterion is achieved by setting constraints on the modulus of the ABF vector w, ensuring that each element's modulus is equal to 1/N. By solving the reformulated optimization problem, the original problem can be addressed, leading to improved CE and user fairness in the mmWave-MEC-ABF system.
read more
3. What is the convergence and complexity of Algorithm 2 and Algorithm 3?
For Algorithm 2, its convergence comes from the convergence of the MM algorithm [30] . The complexity of Algorithm 2 is given as O ( N 3 RF + I 1 N N 2 RF ) , where I 1 is the iterations of Algorithm 2 [30] . For Algorithm 3, its convergence comes from the convergence of Algorithm 2 and the SCA algorithm. The complexity of Algorithm 3 is given as O(I 2 I 3 (GN 3 RF + N 3 RF + I 1 N N 2 RF + (2N G + 6N U + 1) 3.5 ln(1/d))), where I 2 and I 3 denote the outer and inner iterations of Algorithm 3, respectively, and d is the solution accuracy [24] .
read more
4. How does mmWave-MEC-HBF improve CE performance?
mmWave-MEC-HBF improves CE performance by jointly designing the digital BF and analog BF. The system model includes N antennas, N RF RFCs, N PAs, and N RF N PSs. The HBF consists of the ABF matrix A C N xNRF and the DBF matrix D C NRFxNS. The ABF matrix satisfies the CM constraint, and the DBF matrix is rewritten as D = [d 1 , ..., d G ], where G represents the number of data streams. Users are divided into G groups, each corresponding to an independent data stream. The received signal after user grouping and HBF processing is expressed using equations. The SIC decoding constraint ensures effective channel gains, and the SINR of each user is calculated. The achievable rate and energy consumption are determined based on the SINR. Overall, the joint design of digital and analog BFs in mmWave-MEC-HBF enhances CE performance by optimizing the system model and improving data stream processing.
read more