PaperlayerView exact revision on arXiv

Source-linked AI summary

Modulation in the Air: Backscatter Communication over Ambient OFDM Carrier

Gang Yang, Ying-Chang Liang, Rui Zhang, Yiyang Pei

arXiv:1704.02245v2cs.IT

TL;DR

The paper addresses AmBC over ambient OFDM carriers, where direct-link interference complicates reliable backscatter communication. It develops a spread-spectrum model, jointly designs the BD waveform and receiver detector, and adds practical synchronization methods. The proposed transceiver outperforms conventional energy detection in transmission rate, BER performance, and operating range.

  • Problem

    Ambient OFDM-based AmBC must support over-the-air communication despite strong direct-link interference from the ambient RF source.

  • Method

    The paper models the system as spread-spectrum communication and jointly designs BD waveforms, interference-canceling detectors, and timing synchronization algorithms.

  • Results

    The proposed transceiver outperforms the conventional design in transmission rate, BER performance, and operating range.

  • Takeaways & Limitations

    Ambient OFDM carriers can support a higher-performing AmBC transceiver without requiring complete relevant-channel information at the receiver.

Abstract

from arXiv · show

Ambient backscatter communication (AmBC) enables radio-frequency (RF) powered backscatter devices (BDs) (e.g., sensors, tags) to modulate their information bits over ambient RF carriers in an over-the-air manner. This technology also called "modulation in the air", thus has emerged as a promising solution to achieve green communications for future Internet-of-Things. This paper studies an AmBC system by leveraging the ambient orthogonal frequency division multiplexing (OFDM) modulated signals in the air. We first model such AmBC system from a spread-spectrum communication perspective, upon which a novel joint design for BD waveform and receiver detector is proposed. The BD symbol period is designed to be in general an integer multiplication of the OFDM symbol period, and the waveform for BD bit `0' maintains the same state within a BD symbol period, while the waveform for BD bit `1' has a state transition in the middle of each OFDM symbol period within a BD symbol period. In the receiver detector design, we construct the test statistic that cancels out the direct-link interference by exploiting the repeating structure of the ambient OFDM signals due to the use of cyclic prefix. For the system with a single-antenna receiver, the maximum-likelihood detector is proposed to recover the BD bits, for which the optimal threshold is obtained in closed-form expression. For the system with a multi-antenna receiver, we propose a new test statistic, and derive the optimal detector. Moreover, practical timing synchronization algorithms are proposed, and we also analyze the effect of various system parameters on the system performance. Finally, extensive numerical results are provided to verify that the proposed transceiver design can improve the system bit-error-rate (BER) performance and the operating range significantly, and achieve much higher data rate, as compared to the conventional design.

I. INTRODUCTION

The paper studies ambient backscatter over OFDM carriers to address direct-link interference and limited performance in existing AmBC designs. It proposes a spread-spectrum-based transceiver with practical synchronization and reports improved BER, data rate, and operating range over conventional designs.

  • Ambient backscatter uses RF-powered devices to communicate over ambient RF carriers, supporting green communications and future IoT applications.
  • Existing AmBC receivers often treat strong direct-link interference as noise, causing low decoding SNR and data rate.
  • Prior interference-cancellation approaches can require extra antennas, specialized hardware, full-duplex radios, or collaboration among multiple WiFi access points.
  • The paper models AmBC over ambient OFDM carriers from a spread-spectrum perspective and jointly designs the BD waveform and receiver detector.
  • The proposed design cancels direct-link interference without increasing hardware complexity, while supporting optimal single-antenna and multi-antenna detection.
  • Numerical results show lower BER, higher data rate, and significantly greater operating range than the conventional design, with multiple antennas further improving BER.

II. SYSTEM MODEL AND PROTOCOL DESCRIPTION

This section presents the system model and link-layer protocol for ambient backscatter communications over ambient OFDM carriers.

  • The system model describes ambient backscatter communications over OFDM carriers in the air.

A. System Model

The AmBC system coexists with a legacy OFDM system: a powered backscatter device harvests ambient energy and modulates the carrier for transmission to a receiver.

  • The system contains an RF source, legacy users, a backscatter device, and a receiver with M ≥ 1 antennas.The RF source transmits OFDM signals to legacy users while the BD communicates with the receiver over that ambient carrier.
  • The BD includes a backscatter antenna, switched-load transmitter, energy harvester, information receiver, controller, memory, and rechargeable battery.The energy harvester collects energy from ambient OFDM signals for BD operation and transmission.
  • The BD transmits information by switching its load impedance, changing the amplitude and/or phase of its backscattered signal.The resulting signal is received and decoded by the receiver.
  • The model defines source-to-BD, BD-to-receiver, and source-to-receiver channels using their impulse responses, spreads, and propagation delays.The backscatter channel is the concatenation of the source-to-BD and BD-to-receiver channels; relevant channels are assumed independent.
  • The model assumes independent relevant channels and describes low-complexity BD information reception using direct conversion and an ultra-low-power ADC.

1) Continuous-time Signal Model:

The continuous-time model represents the BD as modulating an ambient passband signal and separates the received signal into backscatter, direct-link interference, and noise components.

  • The cyclic prefix is longer than legacy-channel spreads, and practical AmBC assumptions make it much longer than the maximum AmBC channel spread.This repeating structure supports cancellation of direct-link interference in the receiver design.
  • The BD’s reflection coefficient α controls the power of the backscattered signal, represented as αe_c(t)x(t).α depends on the antenna and load impedances.
  • The paper treats backscatter as modulation of an ambient carrier, calling the technique “modulation in the air.”The BD signal is the baseband modulation signal, while the incident passband signal from the air serves as the carrier.
  • Modulation in the air avoids local RF-carrier generation, reducing BD hardware complexity and power consumption.
  • Direct-link interference is typically much stronger than the BD backscatter, making BD detection challenging.

2) Discrete-time Signal Model:

The discrete-time model samples the propagated channels and received signals, then frames the transceiver objective as recovering the BD signal without knowing the source waveform.

  • Sampling at rate f_s converts the source, received, and BD signals into discrete-time sequences.
  • Discrete-time delays are defined by flooring each propagation delay multiplied by f_s, including the backscatter delay D_b.The minimum delay is D = min{D_f, D_b}, with D_b ≈ D_h under the stated practical assumption.
  • The model defines discrete-time total channel spreads L_h, L_g, L_f, and L_b, with maximum spread L = max{L_f, L_b}.
  • The sampled received signal includes delayed backscatter, direct-link interference, and AWGN, with the relevant discrete-time signals illustrated in Fig. 4.
  • The transceiver objective is to recover x[n] from receiver observations without knowing the source signal transmitted by the RF source.The paper therefore designs both the BD waveform and receiver detector.

B. Link-layer Protocol Design

The link-layer protocol uses frame-based transmission with four phases: wake-up preamble, blind timing synchronization, training-preamble transmission, and device data transmission. Timing allocation affects estimation accuracy and data-phase throughput, but optimal allocation is not derived.

  • Protocol structure: Each frame contains wake-up-preamble, blind-timing-synchronization, training-preamble-transmission, and device-data-transmission phases.The BD normally sleeps to save energy and activates when it has sufficient data to transmit.
  • Protocol structure: The wake-up phase uses a specialized preamble to activate the receiver hardware.A short alternating ‘1’ and ‘0’ sequence can serve as the wake-up preamble, enabling event-driven receiver wake-up.
  • Timing and training: During blind timing synchronization, the BD estimates the source propagation delay by exploiting the cyclic-prefix structure of the unknown ambient OFDM signal.The blind estimation algorithm is presented as a practical synchronization procedure.
  • Timing and training: During training, the BD transmits a receiver-known preamble, after which the receiver estimates propagation delay, channel spread, and average signal power.The training preamble starts at the BD’s estimated propagation delay and is received alongside the direct-link signal.
  • Design scope: Timing allocations affect parameter-estimation accuracy and communication throughput, but the paper does not derive an optimal allocation across phases.Their effect on estimation is investigated numerically, while the main focus remains transceiver design.

III. TRANSCEIVER DESIGN FOR SINGLE-ANTENNA SYSTEM

This section establishes a spread-spectrum signal model and studies joint BD waveform and receiver-detector design for an AmBC system with a single-antenna receiver.

  • Signal model: The section establishes a spread-spectrum model for the general AmBC signal model.The model provides the basis for subsequent transceiver design.
  • Transceiver design: The transceiver design jointly addresses the BD waveform and the optimal receiver detector.The design targets an AmBC system with a single-antenna receiver, M = 1.

A. A Spread-Spectrum Perspective for Signal Model

The AmBC signal is modeled as low-rate BD data spread by an unknown, time-varying ambient signal, creating detection challenges from unknown spreading and strong direct-link interference. The proposed design addresses these challenges through a spread-spectrum-based joint transceiver approach.

  • Signal model: The backscattered signal is modeled as a low-rate BD data signal multiplied by a high-rate ambient OFDM spreading-code signal.The BD signal is scaled by reflection coefficient α, and the spreading-code chip duration equals the OFDM sampling period.
  • Signal model: Choosing a BD symbol duration of N0 sampling periods yields processing gain G = N0 and BD data rate fs/N0.The symbol period must balance processing gain and data rate.
  • Detection challenges: Unknown, time-varying spreading codes prevent traditional correlation detection.The spreading code depends on the unknown ambient channel and signal, so the receiver cannot directly apply conventional correlation detection.
  • Detection challenges: Unknown direct-link signals are typically much stronger than backscattered signals, producing very low SINR when treated as interference.This direct-link component is a central obstacle to detecting the BD signal.
  • Detection challenges: Coherent detection is unavailable because estimating all relevant fading channels is challenging.The paper therefore focuses on a joint transceiver design for the single-antenna case.

B. BD Waveform Design

The BD symbol spans K OFDM symbol periods, using a constant waveform for bit 0 and a mid-symbol transition in every OFDM period for bit 1. This design supports interference cancellation and remains implementable with simple, low-cost hardware.

  • Waveform construction: Each BD symbol lasts K OFDM symbol periods, giving spreading gain G = K(N + Nc).Each OFDM symbol contains N + Nc sampling periods, including the cyclic prefix.
  • Waveform construction: For bit 1, the BD waveform transitions state in the middle of every OFDM symbol period.The construction assumes N + Nc is even.
  • Waveform construction: For bit 0, the waveform remains constant throughout the BD symbol period.The waveform is x[n] = 1 over all K(N + Nc) samples.
  • Design rationale: The waveform design enables cancellation of strong direct-link interference at the receiver.It is also similar to the FM0 waveform used in commercial RFID tags, supporting implementation in simple, low-cost BDs.

C. Receiver Detection Design

The receiver detection design is introduced for the system model, with K=1 selected without loss of generality.

  • The detector-design analysis considers K = 1 without loss of generality.

1) Construction of Test Statistic:

The proposed statistic exploits cyclic-prefix-induced repetition to cancel direct-link interference and detect backscatter energy. The resulting detector has analytically characterized performance, while synchronization and parameter choices determine practical rate, reliability, and range.

  • Signal repetition: The cyclic prefix creates repeated received-signal portions across OFDM symbols, enabling construction of a statistic from aligned samples.The repetitions are stated for both the receiver-side direct signal and the BD-side incident signal.
  • Test statistic: The intermediate signal completely cancels direct-link interference when B = 1, leaving the received backscattered signal and increasing detection SNR.
  • Test statistic: The statistic depends only on overall backscatter-channel strength, avoiding estimation of the individual channels g and h[n].
  • Optimal detector: The optimal detector is an energy detector, with R having an exact chi-square distribution and a Gaussian approximation available for large repeating length J.The repeating length can be increased through the spreading-gain parameter K.
  • Optimal detector: The single-antenna ML detector has a closed-form optimal threshold and minimum BER determined by detection SNR γ and repeating length J.For fixed J, the threshold and minimum BER are independent of the number of subcarriers N.
  • Performance and implementation: Timing parameters can be estimated from ambient OFDM signals using autocorrelation or known-preamble methods, including cyclic-prefix repetition when the preamble is unknown.The normalized MSE becomes smaller for longer estimation observations.
  • Performance and implementation: Simulations report substantially improved BER and operating range over the conventional energy detector, including BER values near 0.001, 0.01, and 0.1 at 1.4, 4, and 14 meters.The benchmark scheme reports BER values near 0.15, 0.4, and 0.5 at 0.5, 2, and 6 meters.

C. Performance Comparison for Case of Multi-antenna Receiver

The proposed AmBC transceiver uses OFDM structure to cancel direct-link interference, jointly designs the BD waveform and detector, and supports single- and multi-antenna receivers. Simulations report improved BER, transmission rate, operating range, timing synchronization, and receive-diversity performance over conventional energy detection.

  • Multi-antenna performance: 12 dB, 18 dB, and 20 dB SNR gains are achieved at BER 0.001 for M = 2, 4, and 6, respectively, versus the single-antenna case.BER decreases quickly as the number of receiver antennas increases, although incremental gains become smaller as M increases.
  • System design: The system model adopts a spread-spectrum perspective and jointly designs the BD waveform with the receiver detector.The BD waveform and detector are designed together for AmBC over ambient OFDM carriers.
  • Receiver design: The single-antenna detector cancels direct-link interference using cyclic-prefix-induced repetition and recovers BD bits with a maximum-likelihood rule.Its optimal detection threshold is obtained in closed form.
  • Receiver design: The multi-antenna receiver uses a new test statistic and optimal detector, requiring estimation only of the backscatter-channel strength.The design avoids estimating the complete direct-link and backscatter channels.
  • Overall performance: The proposed design outperforms conventional energy detection in transmission rate, BER performance, and operating range.The conventional detector’s BER decreases slowly with M because of strong direct-link interference.
  • Practical implementation: The timing synchronization methods are practically valid and efficient, while multiple receive antennas significantly enhance BER performance.The EGC scheme does not require SNR information and can reduce computational complexity by avoiding a search for optimal combination weights.
Loading 1704.02245v2…