PaperlayerView exact revision on arXiv

Source-linked AI summary

On the Spectral Efficiency and Security Enhancements of NOMA Assisted Multicast-Unicast Streaming

Zhiguo Ding, Zhongyuan Zhao, Mugen Peng, H. Vincent Poor

arXiv:1611.05413v1cs.IT

TL;DR

The paper addresses mixed multicast-unicast transmission and the risk that multicast receivers intercept broadcast unicast messages. It designs NOMA beamforming and power allocation, finding improved spectral efficiency and secrecy performance relative to OMA.

  • Problem

    The paper studies how to support mixed multicast-unicast traffic while preventing multicast receivers from intercepting unicast messages broadcast to all users.

  • Method

    The proposed scheme jointly designs beamforming and power allocation for NOMA-assisted multicast-unicast transmission.

  • Results

    NOMA improves spectral efficiency over OMA, preserves multicasting performance, and achieves a secrecy unicast rate always larger than or equal to OMA’s.

  • Takeaways & Limitations

    The multicast message can serve as a jamming signal that prevents potential eavesdroppers with weak channels from intercepting the unicast message.

Abstract

from arXiv · show

This paper considers the application of non-orthogonal multiple access (NOMA) to a multi-user network with mixed multicasting and unicasting traffic. The proposed design of beamforming and power allocation ensures that the unicasting performance is improved while maintaining the reception reliability of multicasting. Both analytical and simulation results are provided to demonstrate that the use of the NOMA assisted multicast-unicast scheme yields a significant improvement in spectral efficiency compared to orthogonal multiple access (OMA) schemes which realize multicasting and unicasting services separately. Since unicasting messages are broadcasted to all the users, how the use of NOMA can prevent those multicasting receivers intercepting the unicasting messages is also investigated, where it is shown that the secrecy unicasting rate achieved by NOMA is always larger than or equal to that of OMA. This security gain is mainly due to the fact that the multicasting messages can be used as jamming signals to prevent potential eavesdropping when the multicasting and unicasting messages are superimposed together following the NOMA principle.

I. INTRODUCTION

The paper applies NOMA to mixed multicast-unicast traffic, using excess spatial degrees of freedom to improve unicasting while preserving multicasting reliability. It characterizes spectral-efficiency and secrecy gains over OMA.

  • I. INTRODUCTION: The proposed scheme superimposes unicasting with multicasting so excess spatial degrees of freedom improve unicasting while maintaining multicasting reliability.
  • I. INTRODUCTION: The scheme achieves the same multicasting performance as OMA schemes that provide multicasting and unicasting separately.
  • I. INTRODUCTION: Analytical and simulation results show significant unicasting performance gains over OMA and provide user-scheduling guidelines for further improvement.
  • I. INTRODUCTION: The paper investigates whether NOMA can prevent multicasting receivers from intercepting unicasting messages broadcast to all users.
  • I. INTRODUCTION: NOMA achieves a secrecy unicasting rate always larger than or equal to OMA’s, because the multicasting message can act as a jamming signal.

II. SYSTEM MODEL

The system has one base station transmitting a multicast message to all users and a unicast message to one user. NOMA jointly designs beamforming and power allocation for these streams.

  • II. SYSTEM MODEL: A base station with M antennas serves K single-antenna users, transmitting multicast data to all users and unicast data to a particular user.
  • II. SYSTEM MODEL: OMA uses two orthogonal resource blocks to deliver multicast and unicast messages separately, whereas NOMA delivers them together.
  • II. SYSTEM MODEL: The beamforming vector is designed to increase the difference between users’ effective channel gains, enabling NOMA even with similar channel conditions.
  • II. SYSTEM MODEL: User 1 applies successive interference cancellation, while other users decode the multicast message by treating the unicast message as noise.
  • II. SYSTEM MODEL: Power allocation gives the multicast message higher priority and imposes constraints ensuring all users can receive it at the targeted rate R_M.
  • II. SYSTEM MODEL: The unicast rate can be zero when a user experiences deep fading, causing the base station to allocate all power to multicasting.
  • II. SYSTEM MODEL: The gap between user 1’s unicast rate and other users’ eavesdropping rates is used to study the security of unicast transmission.

B. A sophisticated OMA-based benchmarking scheme

The benchmark uses dynamically allocated OMA resources for multicasting and unicasting, with TDMA as a representative implementation. Its multicasting phase must satisfy all users, while the remaining duration carries unicasting; deep fading can eliminate the achievable unicasting rate.

  • Two OMA benchmarks are considered: fixed orthogonal resources and dynamically adjusted resources based on users’ channel conditions.
  • The dynamic scheme is used as the benchmark because it outperforms the fixed-resource alternative.
  • Dynamic allocation is difficult to implement because arbitrary time durations or bandwidths require high-cost circuits.
  • In the TDMA benchmark, a fraction γ of the slot carries multicasting, while the remaining 1 −γ duration carries unicasting.
  • The multicasting phase uses a beamforming vector, with simulations finding similar performance across choices and p = hH 1 slightly outperforming the other two.
  • The time allocation coefficient is constrained so all users can receive sM, and the achievable unicasting rate can be zero under deep fading.

III. SPECTRAL EFFICIENCY ENHANCEMENTS ACHIEVED BY NOMA

The analysis focuses on unicasting performance because NOMA and OMA provide the same multicasting performance. Spectral efficiency is evaluated through unicasting outage and instantaneous-rate comparisons.

  • NOMA and OMA achieve the same multicasting performance, as stated in Proposition 1.
  • The paper therefore focuses on unicasting performance after establishing the multicasting equivalence.
  • Two criteria evaluate spectral efficiency: unicasting outage probability and comparison of instantaneous unicasting rates between NOMA and OMA.

A. Characterizing the Unicasting Outage Probability

The paper derives a closed-form characterization of the proposed NOMA unicasting outage probability and analyzes its diversity gain. The derivation uses order-statistics decompositions, bounds, high-SNR approximations, and Chebyshev-Gauss quadrature.

  • The proposed NOMA transmission has a closed-form unicasting outage-probability expression.The result follows by combining separated terms and substituting the derived expressions into the outage formula.
  • The derivation models z1 and the minimum u of z2 through zK, then separates the outage probability into three terms.
  • Chebyshev-Gauss quadrature approximates Q3 because its underlying integral lacks a convenient exact expression.
  • The proposed NOMA scheme achieves diversity gain 1 for unicasting transmission.Upper and lower outage-probability bounds both yield one.
  • The diversity gain of 1 is attributed to the weakest channel gain u becoming the system bottleneck under the cognitive-radio power-allocation policy.When the weakest channel cannot detect the multicasting message, all power is spent on multicasting, creating the dominant unicasting-outage event.
  • The diversity-gain result is consistent with a result previously reported in.

B. Performance Gain of NOMA over OMA

The analysis identifies when NOMA outperforms OMA and motivates scheduling the strongest-channel user for unicasting. This scheduling avoids a damaging channel-ordering event and increases the NOMA–OMA performance gap.

  • When all power or time is allocated to multicasting, NOMA and OMA achieve the same performance.
  • NOMA may have a lower unicasting rate than OMA, even at high SNR.
  • The event z1 < u eliminates NOMA’s performance gain over OMA.
  • The base station schedules the user with the largest channel norm for unicasting.
  • The proposed scheduling scheme effectively increases the performance gap between NOMA and OMA.

IV. SECURITY ENHANCEMENTS ACHIEVED BY NOMA

The paper studies whether NOMA can improve unicasting security relative to OMA and how scheduling can further enhance that security. It establishes a non-inferiority result for NOMA’s secrecy unicasting rate.

  • NOMA unicasting can always improve unicasting security compared with OMA.
  • The analysis also studies secrecy outage probability to identify ways of improving NOMA’s security enhancement.

A. The reduction of the eavesdropping capability by using NOMA

The secrecy-rate analysis proves that NOMA’s secrecy unicasting rate is never below OMA’s, with equality in specific weak-legitimate-channel or zero-unicasting cases. Scheduling can avoid the weak-channel equality case.

  • Theorem 2 establishes RS ≥ ¯RS, so NOMA’s secrecy unicasting rate is always at least OMA’s in the high-SNR regime.
  • If all power or time is allocated to multicasting, both unicasting rates are zero and the secrecy-rate difference is zero.
  • The proof uses monotonicity of FzK(x) to show that the secrecy-rate difference is nonnegative when z1 ≥ z2.
  • When user 1 has a weak channel, NOMA and OMA can have the same secrecy rates.
  • The proposed user scheduling scheme avoids this undesirable weak-channel situation and improves NOMA’s secrecy-performance gain over OMA.
  • NOMA increases every user’s capability to detect the unicasting message, but improves user 1 more than the other users.

B. Characterizing the Secrecy Outage Probability

The secrecy outage probability is decomposed into three terms and approximated using channel-order statistics and Chebyshev–Gauss integration. Q6 is the dominant term in demanding high-SNR, large-user settings.

  • The outage expression is derived by decomposing cases involving z1, u, and the multicast threshold ǫM, then applying channel-order statistics and Chebyshev-Gauss approximation.
  • The secrecy outage probability consists of three parts, Q4, Q5, and Q6.
  • Q6 is dominant over Q4 and Q5, particularly at high SNR and when M and K are large.
  • Q5 becomes small when K and M are large.
  • Increasing SNR reduces Q4 for fixed u and v by reducing the difference between its associated functions.
  • For large K, the relevant function difference can make Q6 large, and increasing SNR does not reduce it.
  • The proposed user scheduling reduces outage by making the event z1 < 2 ˜RSv less likely, thereby reducing Q6.

V. NUMERICAL RESULTS

Numerical results show that NOMA substantially improves unicasting and secrecy performance over OMA while preserving multicasting performance. User scheduling further increases NOMA's advantage, and analytical results closely match simulations.

  • Unicasting performance: At SNR 16dB with M = 10, NOMA supports 4 BPCU for unicasting versus 0.8 BPCU for OMA.
  • Unicasting performance: NOMA achieves nearly five times the OMA unicasting rate, with corresponding improvements in outage probability.
  • Multicasting performance: NOMA maintains the same multicasting performance as OMA despite its unicasting gain.
  • Validation: Simulation curves match analytical results, and the diversity gain remains 1 regardless of the number of base-station antennas.
  • Beamforming: The beamforming choice in (2) provides a slight gain, while differences among OMA beamforming schemes are insignificant.
  • Secrecy performance: NOMA secrecy unicasting rate is always at least OMA's; at 10dB SNR, NOMA achieves 1.2 BPCU versus more than 0.3 BPCU for OMA.
  • Secrecy scheduling: At 20dB SNR, scheduling increases the NOMA–OMA secrecy-rate gap from 0.6 BPCU to 1.1 BPCU.

VI. CONCLUSIONS

The paper applies NOMA to mixed multicast–unicast traffic using jointly designed beamforming and power allocation. The scheme improves spectral efficiency and secrecy unicasting relative to OMA while preserving multicasting reliability.

  • The proposed design jointly selects beamforming and power-allocation coefficients to improve unicasting while maintaining multicasting reception reliability.
  • The NOMA-assisted scheme significantly improves spectral efficiency over OMA, where multicast and unicast services are realized separately.
  • NOMA secrecy unicasting rate is always larger than or equal to OMA's while multicasting reliability is maintained.
Loading 1611.05413v1…