PaperlayerView exact revision on arXiv

Source-linked AI summary

Blind and Asynchronous Initial Beam Acquisition and Clock-Assisted Re-Acquisition in Standalone Mobile MIMO-OFDM Systems

Ali Abdi, Firooz B. Saghezchi, Haris Gacanin

arXiv:2610.13436v1eess.SP

TL;DR

Standalone mobile MIMO-OFDM beam acquisition must work without prior synchronization or out-of-band feedback. The paper combines blind BIBA with clock-assisted CARA, reducing re-acquisition time by approximately 55% in the considered hardware configuration.

  • Problem

    Standalone initial acquisition cannot rely on prior synchronization or feedback, and beam misalignment can leave SNR too low for reliable time synchronization.

    Source: §I · p. 1

  • Method

    The framework combines asynchronous BIBA for initial beam-pair selection with CARA, which uses retained clock synchronization for recovery, and derives mobility-feasibility bounds.

    Source: §I · p. 1

  • Results

    Approximately 55% lower theoretical acquisition time for CARA than BIBA (176 ms for BIBA fixed-length and 80 ms for CARA) under the considered hardware; both protocols operated successfully on the testbed.

    Source: C. Beam Acquisition Time · p. 8

  • Takeaways & Limitations

    The framework provides a practical path toward autonomous standalone deployments without an external feedback channel or prior synchronization.

    Source: §VI · p. 10

    Scaling the codebook to massive MIMO and evaluating performance in dense multipath and multi-user environments remain future extensions.

    Source: §VI · p. 10

Abstract

from arXiv · show

Building upon our preliminary study, this paper proposes a standalone beam-management framework for mobile MIMO-OFDM systems that integrates Blind Initial Beam Acquisition (BIBA) with Clock-Assisted Re-Acquisition (CARA). BIBA is a blind and asynchronous protocol that requires neither prior synchronization nor an out-of-band feedback channel, so the Base Station (BS) and User Equipment (UE) can independently identify the beam pair maximizing received signal strength. To handle subsequent link drops, the newly introduced CARA protocol exploits the clock synchronization established after BIBA to reduce the re-acquisition time by approximately 55% compared with BIBA. The operational reliability of both protocols is bounded by a probabilistic feasibility analysis under random UE mobility trajectories. Protocols are validated on a 28 GHz mmWave testbed under practical hardware constraints, including beam-switching and Time Division Duplex (TDD) latencies, and measured Software-Defined Radio (SDR) software processing delays. Finally, we formulate a unified Finite State Machine (FSM) model for both BIBA and CARA to ensure robust integration into practical wireless systems

I. INTRODUCTION

The paper proposes a standalone framework combining blind, asynchronous initial beam acquisition with clock-assisted recovery, and evaluates it alongside probabilistic mobility bounds. It addresses gaps in prior methods that rely on coordination, random scanning, or assumptions that omit hardware latency.

I. INTRODUCTION

  • Standalone initial acquisition must work without prior synchronization or feedback because beam misalignment can leave SNR too low for reliable time synchronization.Reliable OFDM frame synchronization follows only after acquisition establishes sufficient link budget.Source: §I · p. 1
  • BIBA finds the strongest beam pair without prior synchronization, spatial information, CSI, or feedback; CARA reuses retained clock synchronization to cut re-acquisition time by approximately 55%.The framework is validated on a 28 GHz SDR testbed under realistic TDD switching latencies and hardware constraints.Source: §I · p. 1
  • The framework derives closed-form feasibility probabilities and mobility boundaries for reliable operation under random UE trajectories.The bounds specify maximum speeds and distances.Source: §I · p. 1

II. RELATED WORK

  • The framework uses a unified FSM to connect asynchronous initial acquisition, synchronized data transmission, and link recovery.It explicitly integrates physical TDD and beam-switching latencies.Source: §I · p. 2, §II · p. 2
  • Prior approaches commonly depend on synchronization, feedback, randomized scanning, or specialized multi-peak arrays, while SDR beam-management platforms often assume initial acquisition is complete.The related work also notes that practical TDD and beam-switching overhead is not fully accounted for in some implementations.Source: §II · p. 2

III. SYSTEM MODEL

The system model uses half-duplex nodes with analog beamforming and sequential, time-slotted sweeps. Precomputed BS and UE beam codebooks support selecting directional beam pairs over the sparse mmWave link.

III. SYSTEM MODEL

  • Half-duplex BS and UE nodes use single-RF-chain UPAs and TDD switching, restricting beam sweeping to sequential time slots.The architecture diagram shows the TDD signal flow and analog beamformers F and W.Source: §III · p. 2, Figure 1 · p. 3

A. Beamformer

  • UPA beam weights align signal phases to maximize the array factor toward a chosen direction.The array has N = N_HN_V elements arranged on a planar grid.Source: A. Beamformer · pp. 2–3
  • The BS and UE store precomputed steering vectors in separate beam codebooks for transmit and receive beam selection over the sparse mmWave channel.The channel is represented by H, with H^T denoting the reciprocal uplink channel.Source: A. Beamformer · p. 3
  • During acquisition, received signal power is measured as RSS and evaluated against a detection threshold.The model represents additive white Gaussian noise in the received signals.Source: A. Beamformer · p. 3

B. Received Signal Strength Measurement

RSS measurements exclude hardware switching transients, and the unified FSM moves the link from asynchronous initial acquisition to synchronized transmission and then recovery. BIBA probes in downlink and uplink phases despite uncoordinated start times.

B. Received Signal Strength Measurement

  • RSS is measured as time-averaged received energy during a beam dwell, excluding a hardware guard interval for switching latency.The guard time is required so the measurement reflects the active beam rather than switching transients.Source: B. Received Signal Strength Measurement · p. 3

B. Blind and Asynchronous Initial Beam Acquisition (BIBA)

  • The BIBA timing diagram illustrates one possible asynchronous start scenario and a threshold-triggered acquisition with a fixed-length variation.In the illustrated case, the BS starts before the UE, with M_BS = M_UE = 4 and C = 5.Source: Figure 3 · p. 4

A. Unified Finite State Machine (FSM) Architecture

  • The FSM uses BIBA from the disconnected state to establish a beam pair and synchronization, then triggers CARA after RSS falls below a threshold.If accumulated clock drift exceeds hardware tolerance before recovery, a timeout resets the system.Source: A. Unified Finite State Machine (FSM)… · p. 4

B. Blind and Asynchronous Initial Beam Acquisition (BIBA)

  • BIBA alternates a downlink phase, where the UE selects its receive beam, and an uplink phase, where the BS selects its receive beam.The nodes start asynchronously without an out-of-band feedback channel, so their relative timing offsets vary.Source: B. Blind and Asynchronous Initial Beam… · p. 4

1) DL Phase:

In the asynchronous DL phase, the UE sweeps its receive beams while the BS independently sweeps its transmit codebook. After detecting RSS above threshold, the UE continues measuring to select its best observed beam.

1) DL Phase:

  • The UE sweeps its receive codebook and measures RSS until the first threshold crossing, then continues sampling to identify its best observed beam.Continuing for MBS full UE sweeps lets it probe all possible BS beams while the BS holds each transmit beam for MUEtd.Source: 1) DL Phase · p. 4
  • The UE selects the receive beam associated with the highest measured RSS across transmit–receive beam pairs and then switches to UL.The RSS value ρi,j is measured for the BS's i-th transmit beam and UE's j-th receive beam.Source: 1) DL Phase · p. 4
  • The BS independently and asynchronously sweeps its transmit codebook, holding each beam for MUEtd so the UE can complete a receive sweep.The BS then transitions to UL without feedback or further synchronization.Source: 1) DL Phase · pp. 4–5

2) UL Phase:

In the asynchronous UL phase, the UE transmits using its selected beam while the BS independently sweeps receive beams. The number of BS sweeps trades acquisition duration against the chance of insufficient transmit–receive overlap.

2) UL Phase:

  • The UE transmits continuously with its selected beam for (MUE + 1)MBS td, accommodating uncoordinated start times and providing overlap with BS receive sweeps.Without feedback, the UE assumes the BS has completed selection by the interval's end; otherwise, a timeout resets the protocol.Source: 2) UL Phase · p. 5
  • After its transmission interval, the UE suspends RSS monitoring while the BS may finish its receive sweeps, preventing false link-drop detection.The UE's non-monitoring interval lasts MUE TULblock.Source: 2) UL Phase · p. 5
  • The BS independently begins receiving after its DL phase and TDD guard, sweeping its receive codebook with dwell time td to select the beam with maximum RSS.The timing is designed to provide sufficient overlap even though the UE may not yet be transmitting when the BS begins its UL phase.Source: 2) UL Phase · p. 5
  • Increasing the number of BS receive sweeps C extends observation time and reduces insufficient-overlap risk, trading reliability against acquisition latency.The fixed-length option uses C = MUE + 1 sweeps; the variable-length option uses C = 1, shortening attempts but allowing acquisition failure and variable total time.Source: 2) UL Phase · p. 6

C. Clock-Assisted Re-Acquisition (CARA)

After an RSS drop, CARA uses the wall-clock synchronization established during data exchange to schedule aligned re-acquisition sweeps. This avoids BIBA's redundant overlapping sweeps, while a timeout bounds how long CARA can rely on retained timing.

C. Clock-Assisted Re-Acquisition (CARA)

  • CARA schedules deterministic DL and UL sweeps at predefined wall-clock times after link loss, using synchronization retained from the preceding data exchange.The timing diagram shows the nodes aligning DL and UL sweep phases rather than starting asynchronously.Source: C. Clock-Assisted Re-Acquisition (CARA) · p. 6, Figure 4 · p. 6
  • By eliminating BIBA's redundant overlapping sweeps, CARA reduces re-acquisition overhead relative to blind asynchronous acquisition.The preceding link drop may follow mobility or blockage, when the previously selected beams are no longer optimal.Source: C. Clock-Assisted Re-Acquisition (CARA) · p. 6
  • CARA's testbed drift bound permits up to 200 seconds of total link loss before a blind asynchronous IA reset is strictly required.This theoretical upper bound uses a combined drift of 5.0 µs/s and an intra-symbol guard time of 1 ms; practical timeouts can be shorter.Source: C. Clock-Assisted Re-Acquisition (CARA) · p. 6

V. EXPERIMENTAL EVALUATION AND DISCUSSION

The evaluation uses a 28 GHz SDR MIMO testbed with hardware components and controlled UE movement to assess protocol operation under practical constraints. The reported experimental scope is bounded by a dominant-LoS setup and a reduced beam codebook.

A. Experimental Setup

  • The 28 GHz mmWave MIMO testbed uses SDR-based 2 GHz IF transceivers, up-converters, and two 4 × 4-element UPAs; a motorized rail automates lateral UE movement to evaluate acquisition under mobility.A mmWave-capable horn antenna connected to a spectrum analyzer captures over-the-air signals to monitor post-acquisition link stability during live video transmission.Source: A. Experimental Setup · p. 7, Figure 5 · p. 7
  • The experiment uses four non-uniformly spaced beam directions per node and linear lateral movement, while broader mobility conditions are evaluated analytically.The reduced codebook and trajectory accommodate hardware and laboratory constraints; the arrays have a limited ±45° electronic steering range.Source: A. Experimental Setup · p. 7
  • Experimental validation is limited to a dominant-LoS environment, with SNR, threshold selection, and multipath effects left for future work.The static detection threshold was configured to maintain a reliable margin above the hardware noise floor.Source: A. Experimental Setup · p. 7

B. SDR Software Latency Measurements and Optimization

The framework accounts for SDR processing delays and hardware switching in acquisition timing, and uses clock-assisted re-acquisition to reduce the time required after a link drop. SDR control-loop consolidation lowers delay and variability; under the considered testbed configuration, CARA is faster than fixed-length BIBA.

B. SDR Software Latency Measurements and Optimization

  • 3.2 ms to 0.8 ms average control-loop delay and 4.6 ms to 0.1 ms standard deviation after consolidating RSS calculation, thresholding, and beam control.The mean delay falls below the 1 ms guard time.Source: B. SDR Software Latency Measurements… · p. 8

C. Beam Acquisition Time

  • Acquisition-time accounting includes DL and UL sweeps plus TDD switching latency; UE timing includes fixed slots through RSS-drop monitoring and matches the fixed-length BS duration.The total acquisition time is the longer of the BS and UE durations.Source: C. Beam Acquisition Time · p. 8
  • 80 ms for CARA versus 176 ms for BIBA (fixed-length), a ~55% reduction under the considered hardware configuration; both protocols operated successfully on the testbed.The authors note these testbed-limited values would scale down linearly in commercial deployments.Source: C. Beam Acquisition Time · p. 8
  • BIBA's exhaustive search has O(MBSMUE) time complexity, making larger codebooks such as MBS = 64 and MUE = 16 a significant latency concern.Hierarchical wide-to-narrow codebooks are proposed as a future way to mitigate this scalability bottleneck.Source: C. Beam Acquisition Time · p. 8

D. Mobility Boundaries and Probabilistic Feasibility

The feasibility analysis models acquisition under random UE trajectories, with feasibility constrained by how quickly the UE traverses a BS beam’s decision region. Its probability bounds compare BIBA and CARA across mobility conditions and show how acquisition time and overhead shape supported speeds.

D. Mobility Boundaries and Probabilistic Feasibility

  • Random trajectory feasibility depends on the UE’s perpendicular velocity component, while perpendicular movement toward the worst-case boundary is the most restrictive assumption.The decision-region duration is the time for the UE’s angular position to traverse one BS beam spacing.Source: D. Mobility Boundaries and… · p. 8
  • The analysis derives closed-form probability bounds for feasible acquisition under random UE mobility trajectories.The efficiency factor α limits acquisition or re-acquisition to a small fraction of the decision region.Source: D. Mobility Boundaries and… · pp. 8–9
  • For random trajectories at D = 10 m and D = 50 m, Fig. 7 shows feasibility probability declining as UE speed rises, with CARA retaining near-certain reliability at higher speeds.The figure uses α = 10.Source: D. Mobility Boundaries and… · p. 9, Figure 7 · p. 9
  • With α = 10, acquisition overhead is bounded to 10% of the decision interval, balancing mobility tolerance against time for synchronization and OFDM data transmission.Decreasing α to 5 doubles maximum supported speeds but reduces the data window; increasing it to 20 halves mobility tolerance.Source: D. Mobility Boundaries and… · p. 9
  • CARA’s 55% lower acquisition time yields uniformly steeper maximum-speed boundaries than BIBA across target feasibility probabilities.A lower target probability, such as P = 0.75, expands the allowable mobility region.Source: D. Mobility Boundaries and… · p. 9, Figure 8 · p. 9

VI. CONCLUSIONS

The paper presents a hardware-robust standalone beam-acquisition framework validated on a 28 GHz platform, combining BIBA and CARA with probabilistic mobility bounds and FSM logic. CARA reduces re-acquisition overhead by approximately 55%, while the framework’s stated next steps include broader array and channel evaluations.

VI. CONCLUSIONS

  • The integrated BIBA and CARA framework provides asynchronous, feedback-free beam acquisition and reduces re-acquisition overhead by approximately 55%.The framework was validated on a 28 GHz mmWave platform.Source: §VI · p. 9
  • Closed-form probabilistic feasibility bounds establish operational limits for both protocols under random UE mobility trajectories.Source: §VI · p. 9
  • FSM logic accommodates TDD latencies and SDR software processing delays to support continuous operation.Source: §VI · pp. 9–10
  • The authors identify massive-MIMO codebook scaling and evaluation in dense multipath and multi-user environments as future extensions.Source: §VI · p. 10

APPENDIX A DERIVATION OF ACQUISITION PROBABILITY

The derivation converts the acquisition-feasibility condition into a bound on the random trajectory angle. It then evaluates the probability by considering whether that bound covers all angles or only a restricted angular range.

APPENDIX A DERIVATION OF ACQUISITION PROBABILITY

  • Substituting the stochastic decision-region duration into the feasibility condition yields an angular constraint on |sin ψ|.Source: Appendix A · p. 10
  • Defining β = (∆ϕ D)/(αvUE Tacq) reduces the condition to |sin ψ| < β.Source: Appendix A · p. 10
  • When β ≥ 1, feasibility holds for all trajectory angles; when β < 1, the valid angular range determines the probability.The derivation uses symmetry of |sin ψ| and the uniform distribution of ψ to obtain the closed-form expression in (17).Source: Appendix A · p. 10
Loading 2610.13436v1…