PaperlayerView exact revision on arXiv

Source-linked AI summary

Massive Non-Orthogonal Multiple Access for Cellular IoT: Potentials and Limitations

Mahyar Shirvanimoghaddam, Mischa Dohler, Sarah Johnson

arXiv:1612.00552v1cs.IT

TL;DR

The paper addresses how cellular networks can accommodate the growing number of IoT and M2M devices beyond protocols designed mainly for human communications. It reviews 3GPP and random-access approaches, presents massive uplink NOMA, and discusses its benefits, practical challenges, and research directions.

  • Problem

    Cellular IoT requires access mechanisms that can handle massive numbers of devices, whereas conventional random access is feasible only when interference and per-user resource demands remain limited.

  • Method

    The paper reviews 3GPP cellular-IoT solutions and M2M random-access techniques, then presents uplink massive NOMA and discusses its implementation challenges.

  • Results

    Massive NOMA can support more devices than conventional OMA and ACB strategies while offering high throughput efficiency and a simple system structure.

  • Takeaways & Limitations

    Massive NOMA is presented as a potential multiple-access technology for future cellular systems serving massive IoT applications with low-cost, low-power, and low-complexity devices.

  • Takeaways & Limitations

    Massive NOMA faces practical challenges including device synchronization, because the base station cannot determine every device's timing advance during random data transmission.

Abstract

from arXiv · show

The Internet of Things (IoT) promises ubiquitous connectivity of everything everywhere, which represents the biggest technology trend in the years to come. It is expected that by 2020 over 25 billion devices will be connected to cellular networks; far beyond the number of devices in current wireless networks. Machine-to-Machine (M2M) communications aims at providing the communication infrastructure for enabling IoT by facilitating the billions of multi-role devices to communicate with each other and with the underlying data transport infrastructure without, or with little, human intervention. Providing this infrastructure will require a dramatic shift from the current protocols mostly designed for human-to-human (H2H) applications. This article reviews recent 3GPP solutions for enabling massive cellular IoT and investigates the random access strategies for M2M communications, which shows that cellular networks must evolve to handle the new ways in which devices will connect and communicate with the system. A massive non-orthogonal multiple access (NOMA) technique is then presented as a promising solution to support a massive number of IoT devices in cellular networks, where we also identify its practical challenges and future research directions.

I. INTRODUCTION

Cellular IoT must support billions of heterogeneous devices with low cost, long battery life, broad coverage, and suitable performance. The paper reviews 3GPP solutions and presents massive NOMA as a way for many devices to share radio resources.

  • Massive IoT applications require low-cost, low-energy devices with good coverage for uses such as smart buildings, logistics, tracking, and fleet management.
  • Cellular connectivity is central to IoT, especially for applications requiring large geographical coverage and medium-to-high performance.
  • 3GPP solutions include EC-GSM, NB-IoT, and LTE-M, each adapting cellular systems for low-power wide-area IoT applications.
  • 3GPP improvements reduce device cost and complexity, extend battery life up to 10 years, and improve link budgets by 15 dB for LTE-M and 20 dB for NB-IoT.
  • Massive NOMA is presented as a strategy for allowing millions of devices per square kilometer to share the same radio resources over existing cellular infrastructure.

II. CURRENT ACCESS TECHNIQUES

Conventional wireless networks allocate time and frequency resources orthogonally, while random access lets devices contact the base station to request transmission slots.

  • Most existing wireless networks assign radio resources, including time and frequency, orthogonally among devices.
  • The random access procedure is the process by which devices contact the base station to request a transmission slot.

A. The Random Access Procedure

LTE random access uses a four-step handshake: devices select preambles, receive resource and timing information, request connection, and obtain final allocations before transmitting data.

  • Devices randomly select one of 64 available preambles and transmit it over the physical random access channel.
  • The base station detects preambles, estimates device timing, and sends a random access response containing allocated resources and timing advance information.
  • After receiving the response, each device sends a temporary terminal identity, then receives a resource allocation specifying its identity.
  • Data transmission begins after the connection is established in the allocated time-frequency slots.

B. Challenges of Conventional Random Access for Massive Cellular IoT

Conventional random access becomes inefficient for massive cellular IoT because device collisions, signaling overhead, diverse QoS requirements, and coexistence with H2H traffic strain limited radio resources.

  • Preamble collision and overload problems: Preamble collisions and overload cause interference, congestion, delays, packet loss, high energy consumption, signaling overhead, and radio-resource wastage.Collisions occur when multiple devices select the same preamble, while overload results from excessive concurrent transmissions violating SINR requirements.
  • Excessive overhead: Connection-oriented access spends substantial signaling resources transmitting very small M2M payloads, limiting scalability when many devices access simultaneously.For 100 bytes of data, approximately 59 uplink and 136 downlink bytes may be required for signaling.
  • Different QoS requirements: Access techniques must accommodate diverse QoS requirements, including messages requiring delivery within 10 msec and delay-tolerant traffic that can tolerate several hours.Treating all MTC devices identically can waste radio resources or interrupt service.
  • Co-existence with H2H devices: MTC traffic sharing radio resources with H2H devices can degrade H2H communication or force many M2M devices to delay transmission when resources are limited.Dedicated or dynamically allocated PRACH resources can protect H2H QoS, but may increase waiting for M2M devices.

III. NON-ORTHOGONAL MULTIPLE ACCESS FOR M2M COMMUNICATIONS

Multiple access schemes differ in whether user signals overlap: OMA separates them, whereas NOMA permits controlled overlap and can improve throughput efficiency. In the illustrated two-user scenario, NOMA generally achieves the highest sum rate except at the symmetric case.

  • III. NON-ORTHOGONAL MULTIPLE ACCESS FOR M2M COMMUNICATIONS: OMA separates users in time or frequency, while NOMA overlaps signals by exploiting power, code, or interleaver domains.OMA includes TDMA, FDMA, and OFDMA; NOMA is described as a non-orthogonal approach.
  • III. NON-ORTHOGONAL MULTIPLE ACCESS FOR M2M COMMUNICATIONS: Fig. 3 compares NOMA and OMA achievable rates for a two-user scenario.The figure caption identifies achievable-rate comparison as the figure's focus.
  • III. NON-ORTHOGONAL MULTIPLE ACCESS FOR M2M COMMUNICATIONS: NOMA schemes are motivated by the need for improved system efficiency, QoS, and throughput in 5G mobile cellular networks and IoT applications.The passage contrasts NOMA's throughput advantages with OMA's suitability for channel-aware packet-domain scheduling.

A. The Basic Concept of Uplink NOMA

Uplink NOMA lets multiple users transmit simultaneously over the same time-frequency resources, while SIC separates their signals at the base station. For IoT, eliminating random access can improve resource efficiency and reduce signaling overhead.

  • Uplink NOMA: Two users transmit over the same frequency band and time slot, creating interference that the base station resolves with successive interference cancellation.The receiver decodes one signal while treating the other as interference, subtracts the decoded signal, then decodes the remaining signal.
  • Uplink NOMA: The receiver orders decoding according to the users’ effective SINR, and NOMA’s gain over OMA grows with channel-gain or path-loss differences.The information-theoretic comparison reports that NOMA achieves the highest sum rate except at the symmetric-throughput point.
  • IoT motivation: NOMA removes the random-access stage and allows devices to transmit in shared channels, improving radio-resource use and reducing conventional signaling overhead.This design targets cellular IoT traffic, where access signaling can be costly relative to short messages.

B. NOMA For Massive Cellular IoT

Random NOMA combines access and data transmission by assigning devices to randomly selected subbands and decoding their overlapping messages with SIC. Simulations report higher device support than conventional OMA and ACB strategies.

  • Random NOMA design: Random NOMA lets devices bypass random access, randomly select subbands, and transmit coded messages that the base station decodes with SIC.The scheme divides total bandwidth into subbands and combines network access with data transmission.
  • Random NOMA design: Raptor codes adapt to unequal random device activity across subbands by generating as many coded symbols as the base station requires.The base station can reproduce the code structure using a shared pseudorandom seed.
  • Random NOMA design: A collision requires devices to select both the same subband and the same code structure, making it less likely than a traditional random-access collision.Devices choosing the same subband but different code structures remain structurally distinguishable.
  • Simulation comparison: NOMA outperforms uncoordinated TDMA and FDMA and supports more devices than ACB under the compared resource conditions.The reported limitations for TDMA and FDMA include high collision probability, while ACB is limited by preamble collisions.

C. Potentials of NOMA for Massive Cellular IoT

NOMA’s reported potentials for massive cellular IoT include higher spectral efficiency, robustness in high mobility, compatibility with existing access technologies, and reduced access overhead. It can also integrate with resource management, clustering, beamforming, and multiantenna methods.

  • System efficiency: NOMA can improve spectrum use and system throughput by exploiting the power domain and non-orthogonal multiplexing.
  • System efficiency: NOMA can retain performance gains in high-mobility scenarios because it relies on receiver-side channel-state information rather than frequency-domain scheduling.
  • System integration: NOMA is compatible with OFDMA and its variants, supporting downlink and SC-FDMA uplink deployments.
  • System integration: NOMA can be combined with beamforming and multiantenna technologies to improve system performance.
  • Massive IoT access: NOMA can combine with radio-resource management and random-access techniques to address collision and overload problems in M2M communications.
  • Massive IoT access: Using NOMA, random access can be eliminated, significantly reducing access delay and signaling overhead.

IV. PRACTICAL CONSIDERATIONS OF MASSIVE NOMA FOR MASSIVE CELLULAR IOT AND FUTURE DIRECTIONS

Massive NOMA offers throughput and scalability benefits for massive M2M, but practical deployment must address traffic estimation, channel acquisition, synchronization, coding, decoding complexity, and fairness.

  • Practical challenges: Massive NOMA can improve spectrum efficiency and system capacity, but practical deployment faces several unresolved challenges.The article identifies traffic estimation, channel estimation and power allocation, synchronization, channel coding, SIC complexity, and user fairness as key considerations.
  • Traffic and load estimation: The base station must estimate the number of devices transmitting on each randomly selected subband.Power control can make received power proportional to the number of devices transmitting over each subband.
  • Channel estimation and power allocation: Estimating channels for all simultaneously transmitting devices is almost impossible, especially with multiple antennas.A reciprocal channel assumption allows devices to estimate their channels from broadcast pilots and adjust power to equalize received signal power.
  • Channel estimation and power allocation: NOMA improves throughput by eliminating the random-access procedure and enabling multiuser detection at the base station.For small M2M packets, these throughput gains are emphasized over NOMA's capacity gains.
  • Synchronization among devices: Random NOMA cannot provide timing advance information for every device because devices are identified during data transmission.Location information or timing learned from previous transmissions may be practical for fixed-location M2M devices, but large-scale synchronization remains challenging.
  • Proper channel code design: Random subband occupancy makes each device's effective rate and required code rate variable rather than fixed.Rateless codes are proposed so devices stop transmitting after receiving a base-station acknowledgment.
  • Complexity of SIC: A common sequential decoder can serve all devices, while separate decoders treating other signals as noise may slightly reduce throughput.The stated degradation can be neglected because achievable rate is mainly determined by the strongest received signal.
  • User fairness: Allocating more bandwidth to weak-link devices can lower their required transmit power and equalize throughput or energy efficiency.Strong-link devices can use smaller bandwidth or higher power under this fairness strategy.

V. CONCLUSIONS

The article reviews random-access techniques for M2M and presents massive NOMA as a scalable option for the growth of cellular IoT, while identifying deployment challenges and research directions.

  • V. CONCLUSIONS: The article reviews random-access techniques for M2M communications and summarizes their benefits and challenges.It also presents massive NOMA as a potential multiple-access technology for future cellular systems.
  • V. CONCLUSIONS: Massive NOMA offers high throughput efficiency and a simple system structure for low-cost, low-power, low-complexity IoT devices.These properties are described as particularly beneficial for massive IoT applications.
  • V. CONCLUSIONS: Massive NOMA can provide system scalability for the massive number of devices involved in M2M communications.The article identifies practical challenges and highlights future research directions for this approach.
Loading 1612.00552v1…