Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Power Minimization of Downlink Spectrum Slicing for eMBB and URLLC Users

Fabio Saggese, Marco Moretti, Petar Popovski

TL;DR

This work analyzes a downlink communication serving two types of traffic: enhanced mobile broadband and ultra-reliable low-latency communication, and proposes a lookup table-based approach and a block coordinated descent (BCD) algorithm that is optimal for the URLLC power allocation.

Abstract

5G technology allows heterogeneous services to share the wireless spectrum within the same radio access network. In this context, spectrum slicing of the shared radio resources is a critical task to guarantee the performance of each service. We analyze a downlink communication serving two types of traffic: enhanced mobile broadband (eMBB) and ultra-reliable low-latency communication (URLLC). Due to the nature of low-latency traffic, the base station knows the channel state information (CSI) of the eMBB users while having statistical CSI for the URLLC users. We study the power minimization problem employing orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) schemes. Based on this analysis, we propose a lookup table-based approach and a block coordinated descent (BCD) algorithm. We show that the BCD is optimal for the URLLC power allocation. The numerical results show that NOMA leads to lower power consumption than OMA, except when the average channel gain of the URLLC user is very high. For the latter case, the optimal approach depends on the channel condition of the eMBB user. Even when OMA attains the best performance, the gap with NOMA is negligible, showing the capability of NOMA to reduce power consumption in practically every condition.

Power Minimization of Downlink Spectrum Slicing for eMBB and URLLC Users

TL;DR

This work analyzes a downlink communication serving two types of traffic: enhanced mobile broadband and ultra-reliable low-latency communication, and proposes a lookup table-based approach and a block coordinated descent (BCD) algorithm that is optimal for the URLLC power allocation.

Abstract

5G technology allows heterogeneous services to share the wireless spectrum within the same radio access network. In this context, spectrum slicing of the shared radio resources is a critical task to guarantee the performance of each service. We analyze a downlink communication serving two types of traffic: enhanced mobile broadband (eMBB) and ultra-reliable low-latency communication (URLLC). Due to the nature of low-latency traffic, the base station knows the channel state information (CSI) of the eMBB users while having statistical CSI for the URLLC users. We study the power minimization problem employing orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) schemes. Based on this analysis, we propose a lookup table-based approach and a block coordinated descent (BCD) algorithm. We show that the BCD is optimal for the URLLC power allocation. The numerical results show that NOMA leads to lower power consumption than OMA, except when the average channel gain of the URLLC user is very high. For the latter case, the optimal approach depends on the channel condition of the eMBB user. Even when OMA attains the best performance, the gap with NOMA is negligible, showing the capability of NOMA to reduce power consumption in practically every condition.

Paper Structure

This paper contains 21 sections, 4 theorems, 39 equations, 10 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Contributions
  3. System Model
  4. Signal model
  5. Mutual information
  6. Outage events
  7. Resource allocation
  8. Time-frequency selection
  9. eMBB allocation
  10. URLLC allocation
  11. Single frequency resource
  12. Interference-limited scenario
  13. Simplified solutions for power allocation
  14. A lookup table estimation of URLLC outage probability
  15. A lookup table-based feasible algorithm for power allocation
  16. ...and 6 more sections

Key Result

Proposition 1

Assuming fixed values of $r_u$, and $\mathbf{P}_e$, the outage probability eq:outage:ufull is a non-increasing monotone function of $\mathbf{P}_u$.

Figures (10)

  • Figure 1: A toy example with a resource grid of $F = 6$ frequency channels and $M = 4$ mini-slots. In the first case, $F_u = 3$ channels, $\mathcal{F}_u = \{1, 3, 5\}$, are reserved for URLLC traffic in an OMA paradigm. In the second case, all the mRBs are reserved for $u$ and $e$, i.e., $F_u = 5$ following a NOMA approach. The URLLC packet is transmitted using the first $M_u = 2$ mini-slots for both OMA and NOMA.
  • Figure 2: $P_u^\text{single}(f)$ as a function of $\Gamma_u$ according to \ref{['eq:power:u_single']}, when $\epsilon_u = 10^{-5}$ and $P_e(f) = 0$ dBm.
  • Figure 3: Experimental results of $p_u$\ref{['eq:outage:ufull']} versus $P_u(f)$ with different number of $F_u$, $F_u r_u = 1$, $\Gamma_u = 30$ dB and $P_u(f) = P_u$, $\forall f \in \mathcal{F}_u$.
  • Figure 4: Average power consumption required by SIC as a function of $d_e$, $F_u = 12$.
  • Figure 5: Outage probability $p_u$ versus $P_u(f)$, for $\Gamma_u = 30$ dB, $F_u = 12$ and different $M_u$. $P_u(f) = P_u$, $P_e(f) = P_e$, $\forall f \in\mathcal{F}_u$. The solid lines are for $P_e(f) = 0$ dBm while dashed lines represent $P_e(f) = -\infty$ dBm, i.e., no interference.
  • ...and 5 more figures

Theorems & Definitions (8)

  • Proposition 1
  • proof
  • Proposition 2
  • proof
  • Proposition 3
  • proof
  • Proposition 4
  • proof