PaperlayerView exact revision on arXiv

Source-linked AI summary

Spectral and Energy Efficient Wireless Powered IoT Networks: NOMA or TDMA?

Qingqing Wu, Wen Chen, Derrick Wing Kwan Ng, Robert Schober

arXiv:1801.09109v1cs.IT

TL;DR

The paper asks whether NOMA can improve spectral efficiency or reduce total energy consumption relative to TDMA in wireless-powered IoT networks with energy-limited devices. It derives optimal SE-oriented time allocations for both schemes and compares their energy and SE performance. The analysis finds that N-WPCN consumes more energy and is less spectrally efficient than T-WPCN, with a worst-user bottleneck emerging in multiuser NOMA.

  • Problem

    Whether NOMA improves SE or reduces total energy consumption relative to TDMA remains uncertain for energy-constrained uplink WPCNs, where circuit consumption matters.

  • Method

    The paper derives optimal time allocations for SE maximization in T-WPCN and N-WPCN and analyzes their total energy consumption and achieved SE.

  • Results

    N-WPCN generally consumes more energy and achieves lower SE than T-WPCN when circuit power consumption is considered.

  • Takeaways & Limitations

    NOMA may not be practical for spectral- and energy-efficient wireless IoT networks with energy-constrained devices, motivating energy-efficient NOMA research.

Abstract

from arXiv · show

Wireless powered communication networks (WPCNs), where multiple energy-limited devices first harvest energy in the downlink and then transmit information in the uplink, have been envisioned as a promising solution for the future Internet-of-Things (IoT). Meanwhile, non-orthogonal multiple access (NOMA) has been proposed to improve the system spectral efficiency (SE) of the fifth-generation (5G) networks by allowing concurrent transmissions of multiple users in the same spectrum. As such, NOMA has been recently considered for the uplink of WPCNs based IoT networks with a massive number of devices. However, simultaneous transmissions in NOMA may also incur more transmit energy consumption as well as circuit energy consumption in practice which is critical for energy constrained IoT devices. As a result, compared to orthogonal multiple access schemes such as time-division multiple access (TDMA), whether the SE can be improved and/or the total energy consumption can be reduced with NOMA in such a scenario still remains unknown. To answer this question, we first derive the optimal time allocations for maximizing the SE of a TDMA-based WPCN (T-WPCN) and a NOMA-based WPCN (N-WPCN), respectively. Subsequently, we analyze the total energy consumption as well as the maximum SE achieved by these two networks. Surprisingly, it is found that N-WPCN not only consumes more energy, but also is less spectral efficient than T-WPCN. Simulation results verify our theoretical findings and unveil the fundamental performance bottleneck, i.e., "worst user bottleneck problem", in multiuser NOMA systems.

I. INTRODUCTION

Wireless-powered IoT networks address the difficulty of sustaining massive energy-limited device populations, while NOMA's benefits over TDMA remain uncertain when circuit and transmit energy consumption matter. The paper develops and compares TDMA- and NOMA-based WPCNs under these practical constraints.

  • Wireless power transfer is proposed for massive IoT devices because battery replacement is costly and environmentally concerning, while harvested RF energy is scarce.
  • NOMA enables simultaneous same-spectrum access by multiple users, with SIC used at the receiver to mitigate multiuser interference.
  • Prior NOMA superiority results apply to downlink settings, while earlier uplink WPCN studies emphasize throughput without comparing total energy consumption against TDMA.
  • The paper derives optimal SE-maximizing time allocations for T-WPCN and N-WPCN while accounting for circuit energy consumption.
  • N-WPCN generally requires at least as much downlink WET time and achieves lower SE than T-WPCN when circuit power is non-negligible.
  • The model contains one power beacon, K > 1 wireless-powered devices, and one access point using harvest-then-transmit operation within total time Tmax.

B. T-WPCN and Problem Formulation

The T-WPCN formulation expresses device throughput and system throughput under TDMA, then maximizes SE subject to energy-causality, total-time, and non-negativity constraints.

  • The formulation defines each device's achievable throughput and aggregates these rates into the T-WPCN system throughput.
  • SE maximization is constrained by energy causality, the total available time, and non-negativity of the optimization variables.

C. N-WPCN and Problem Formulation

The N-WPCN formulation models concurrent uplink transmission by all devices, with SIC-based decoding and an SE maximization problem subject to energy and time constraints.

  • The access point uses SIC by decoding users in ordered sequence while treating later-decoded users as noise.
  • The N-WPCN formulation defines individual throughput and system throughput before posing the corresponding SE maximization problem.
  • Its constraints represent energy causality, the total transmission-time limit, and non-negativity of the variables.

III. T-WPCN OR N-WPCN FOR IOT NETWORKS?

The paper derives optimal solutions for both access schemes and then theoretically compares their total energy consumption and achieved SE.

  • The analysis first derives optimal solutions to the T-WPCN and N-WPCN optimization problems.
  • It then compares the total energy consumed and spectral efficiency achieved by the two networks.

A. Optimal Solution for T-WPCN

The T-WPCN optimization is simplified by active energy-causality constraints and solved through a convex optimization framework, yielding optimal time allocations via Lagrangian conditions and bisection.

  • Energy-causality simplification: At the optimum, each device depletes all harvested energy, so the energy-causality constraint holds with equality.Otherwise, transmit power could be increased to improve the objective.
  • Convex optimization: Problem (8) is convex and satisfies Slater’s condition, enabling efficient solution through the Lagrange dual method.The Lagrangian is formed using the multiplier associated with constraint (8b).
  • Optimality conditions: The optimal time variables satisfy stationarity conditions obtained by setting the partial derivatives with respect to τ0 and τk to zero.These conditions determine the optimal xk values through a system of equations.
  • Time allocation: The optimal time allocation for T-WPCN is obtained from the resulting user-specific equations and can be computed efficiently by bisection.The allocation expression follows after solving for the relevant optimization variable.

B. Optimal Solution for N-WPCN

The N-WPCN optimization is reduced to a two-variable problem with the same structure as the single-user T-WPCN case. Consequently, the T-WPCN solution extends directly to N-WPCN.

  • Reduced optimization: Problem (7) is simplified to an optimization over the downlink WET time τ0 and the NOMA uplink duration ¯τ1.The reduced problem includes a total-time constraint and non-negativity constraints.
  • Structural equivalence: Problem (15) has the same structure as problem (8) when K = 1, apart from minor changes in constant terms.This structural equivalence connects the N-WPCN formulation to the single-user T-WPCN formulation.
  • Optimal allocation: The proposed T-WPCN solution can therefore be immediately extended to obtain the optimal time allocation for N-WPCN.The two optimal-solution derivations provide the theoretical foundation for comparing T-WPCN and N-WPCN.

C. TDMA versus NOMA

The paper compares optimal TDMA- and NOMA-based WPCNs in energy consumption and spectral efficiency, accounting for circuit power. NOMA generally requires more energy and achieves lower spectral efficiency than TDMA when circuit consumption is considered.

  • N-WPCN requires at least as much downlink wireless energy-transfer time as T-WPCN at the optimum.The comparison is stated as τ⋆_0 ≥ τ∗_0.
  • N-WPCN consumes at least as much total energy as T-WPCN, with equality when all devices have zero circuit power.The larger consumption reflects both longer downlink energy transfer and circuit-energy effects.
  • T-WPCN achieves maximum SE greater than or equal to N-WPCN.Theorem 2 establishes the maximum-SE comparison between the two schemes.
  • With nonzero circuit power, NOMA achieves strictly lower SE than TDMA because simultaneous access increases circuit energy consumption through longer transmission time.The paper contrasts this energy-limited setting with conventional transmit-power-limited scenarios where NOMA can gain SE.
  • In transmit-power-limited scenarios, NOMA’s SE gain over TDMA comes at the expense of higher energy consumption, whereas energy-limited devices obtain no SE gain from NOMA without circuit power.When circuit power is included, the comparison favors TDMA in both SE and energy consumption.

IV. NUMERICAL RESULTS

The numerical evaluation uses a 10-device wireless-powered IoT network with specified propagation, hardware, noise, power, and time-budget parameters.

  • Simulation setup: 10 IoT devices are uniformly distributed within a 5-meter disc centered on the power beacon.The carrier frequency is 750 MHz and bandwidth is 180 kHz.
  • Channel model: Rayleigh fading uses unit-mean exponential channel power gains with path-loss exponent α = 2.2.Both downlink and uplink gains follow the stated distance-dependent model.
  • Device parameters: Each device has harvesting efficiency η_k = 0.9 and circuit power p_c,k = 0.1 mW.The devices are assumed to have identical parameters.
  • Operating parameters: The simulation sets noise power to −117 dBm, PB transmit power to 40 dBm, and total transmission time to 0.1 s.These are among the stated system parameters.

A. SE versus PB Transmit Power

As PB transmit power increases, both schemes achieve higher throughput, while optimized T-WPCN substantially outperforms N-WPCN and N-WPCN consumes more energy.

  • Throughput: Throughput increases with PB transmit power for both T-WPCN and N-WPCN.Higher PB power enables devices to harvest more energy for uplink transmission.
  • Throughput: Optimized T-WPCN significantly outperforms optimized N-WPCN, with the performance gap widening as PB transmit power increases.Higher PB power shortens downlink wireless energy transfer, leaving more time for uplink information transmission; simultaneous NOMA transmissions increase circuit consumption.
  • Time allocation: Fixed-time baseline schemes lose throughput relative to their corresponding optimized schemes.This indicates that optimizing the downlink energy-transfer duration matters for spectral-efficiency maximization.
  • Energy consumption: Optimized N-WPCN is generally more energy demanding than optimized T-WPCN.The result verifies the theoretical energy-consumption comparison.
  • Energy consumption: At P_E = 28 dBm, optimal N-WPCN and T-WPCN have nearly equal energy consumption, yet N-WPCN still suffers substantial spectral-efficiency loss.The passage attributes this loss to harvested energy being consumed substantially by circuitry during simultaneous transmission.

B. SE versus Device Circuit Power

Increasing device circuit power affects N-WPCN much more severely than T-WPCN, especially as the number of devices grows.

  • Zero circuit power: For p_c,k = 0, T-WPCN and N-WPCN achieve identical throughput and energy consumption for K = 10 and K = 50.This agrees with the stated theoretical findings.
  • Circuit-power effects: As p_c,k increases, T-WPCN throughput moderately decreases and energy consumption increases.The corresponding N-WPCN changes are much sharper.
  • Circuit-power effects: For N-WPCN, increasing p_c,k sharply decreases throughput and sharply increases energy consumption.Concurrent uplink transmissions raise circuit energy use while energy causality is constrained by the worst downlink channel.
  • Worst-user bottleneck: The N-WPCN uplink duration is limited by the worst downlink channel gain when p_c,k > 0.The paper names this phenomenon the “worst user bottleneck problem.”
  • Number of devices: When K increases from 10 to 50, N-WPCN performance decreases, whereas T-WPCN exploits multiuser diversity to improve performance.TDMA can allocate each user's uplink duration individually according to its channel gains.

V. CONCLUSIONS

Accounting for circuit energy, the paper concludes that N-WPCN is neither more spectrally efficient nor more energy efficient than T-WPCN in energy-constrained wireless IoT networks.

  • Main conclusion: N-WPCN generally requires longer downlink wireless energy transfer and consumes more energy than T-WPCN.The comparison includes circuit energy consumption.
  • Main conclusion: N-WPCN generally achieves lower spectral efficiency than T-WPCN.The conclusion applies to the wireless-powered IoT setting studied in the paper.
  • Implications: The results suggest that NOMA may not be practical for spectral- and energy-efficient wireless IoT networks with energy-constrained devices.The paper identifies energy-efficient NOMA transmission as a direction for future study.
Loading 1801.09109v1…