TR-PSM: Thresholded Reporting with PSM

• TR-PSM is a mechanism that enhances privacy by releasing noisy location updates only when the displacement exceeds a private threshold, reducing cumulative privacy loss.
• It builds on the Planar Staircase Mechanism (PSM) by combining per-location perturbation with a threshold-based reuse policy to mitigate temporal correlation.
• Empirical evaluations demonstrate that TR-PSM improves trace-level privacy and utility in LB-AR applications while managing session-based privacy budgets effectively.

Searching arXiv for the primary TR-PSM paper and closely related works for disambiguation. Thresholded Reporting with PSM (TR-PSM) is the selective release mechanism introduced within PrivAR for real-time privacy protection in location-based augmented reality (LB-AR). In this setting, PSM denotes the Planar Staircase Mechanism. PSM perturbs individual locations to provide per-location privacy, whereas TR-PSM adds a thresholded reporting policy: a noisy location update is released only when the displacement from the last released noisy location exceeds a private threshold. The resulting many-to-one mapping from multiple true locations to a repeated released point is intended to improve trace-level privacy while preserving the low latency and high quality of service required by sub-second GPS streaming in LB-AR applications (Seeam et al., 4 Aug 2025).

1. Context and design objective

PrivAR studies privacy protection for LB-AR applications such as Pokémon-Go-style games and social AR platforms. The motivating setting is continuous, high-frequency, near-real-time location streaming from a trusted client device to an untrusted remote server, with a passive network observer also considered. The design target is threefold: R1: low latency, R2: rigorous privacy, both for each released point and for the whole trajectory, and R3: high QoS. The paper argues that conventional location-based services are less demanding because they issue relatively infrequent queries and can tolerate larger perturbation and higher latency, whereas LB-AR requires sub-second updates and interactive responsiveness (Seeam et al., 4 Aug 2025).

The immediate technical problem is that independently perturbing every location exposes temporal structure. The paper identifies two specific limitations of repeated per-point protection. First, if a mechanism with per-location guarantee $\epsilon$-GeoInd is applied independently at every timestamp over a trajectory $\mathbf{x}=\{x_1,\dots,x_T\}$, the total privacy loss scales linearly as $T\epsilon$. Second, one-to-one release of perturbed points preserves strong spatio-temporal correlations, so an attacker can still exploit clustering around the underlying path. PrivAR addresses these two issues with a two-component architecture: PSM improves per-location utility relative to Planar Laplace, and TR-PSM reduces unnecessary releases so that trajectory privacy depends on threshold crossings rather than on every timestamp (Seeam et al., 4 Aug 2025).

PSM itself is the base perturbation mechanism. It perturbs a true location $x\in\mathbb{R}^2$ by sampling a random radius $r$ and angle $\theta$, then outputting

$$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$

Its radial law is staircase-shaped rather than Laplace-shaped, and the paper states Theorem 1: PSM satisfies $\epsilon$-GeoInd for each location. TR-PSM does not replace this per-location mechanism; it uses it selectively (Seeam et al., 4 Aug 2025).

2. Mechanism and algorithmic structure

TR-PSM is a session-based cached-state mechanism. Its inputs are a trace $\mathbf{x}=(x_1,\dots,x_T)$, a total budget $\epsilon_T$, a per-release budget $\mathbf{x}=\{x_1,\dots,x_T\}$0, and a nominal displacement threshold $\mathbf{x}=\{x_1,\dots,x_T\}$1. It maintains three pieces of state: a private threshold $\mathbf{x}=\{x_1,\dots,x_T\}$2, the most recently released noisy location $\mathbf{x}=\{x_1,\dots,x_T\}$3, and a remaining budget counter $\mathbf{x}=\{x_1,\dots,x_T\}$4 (Seeam et al., 4 Aug 2025).

Initialization consumes two units of privacy budget. First, threshold noise is sampled from PSM’s radial sampler,

$\mathbf{x}=\{x_1,\dots,x_T\}$5

and the private threshold is set to

$\mathbf{x}=\{x_1,\dots,x_T\}$6

Second, the first location is perturbed and released using PSM,

$\mathbf{x}=\{x_1,\dots,x_T\}$7

after which

$\mathbf{x}=\{x_1,\dots,x_T\}$8

If $\mathbf{x}=\{x_1,\dots,x_T\}$9, the algorithm returns an error (Seeam et al., 4 Aug 2025).

For each later timestamp $T\epsilon$0, TR-PSM computes the Euclidean displacement from the current true location to the last released noisy location,

$T\epsilon$1

The reporting rule is

$T\epsilon$2

When $T\epsilon$3, the previous report is reused rather than suppressed; the mechanism still outputs a stream element, but it is identical to the cached release. When $T\epsilon$4, a fresh PSM-perturbed point is released, $T\epsilon$5 is updated to that point, and $T\epsilon$6 is reduced by $T\epsilon$7. If a threshold crossing occurs when $T\epsilon$8, the algorithm returns BudgetExhausted (Seeam et al., 4 Aug 2025).

This structure is the essential source of the method’s many-to-one mapping. Instead of producing a one-to-one perturbed trajectory, TR-PSM may produce runs

$T\epsilon$9

even though the corresponding true locations differ. In PrivAR, that reuse policy is the defining distinction between PSM and TR-PSM (Seeam et al., 4 Aug 2025).

3. Privacy model and formal guarantees

The privacy notion used is Geo-indistinguishability. PSM provides the per-location guarantee, and TR-PSM extends this to trajectory-level accounting by combining PSM releases with zero-cost reuse steps. The paper states Theorem 2: TR-PSM satisfies $x\in\mathbb{R}^2$0-GeoInd for each location and $x\in\mathbb{R}^2$1-GeoInd for a trace $x\in\mathbb{R}^2$2, where

$x\in\mathbb{R}^2$3

and $x\in\mathbb{R}^2$4 is the number of threshold crossings after the first fix (Seeam et al., 4 Aug 2025).

The accounting is compositional. One unit of $x\in\mathbb{R}^2$5 is spent to privatize the threshold, one unit is spent for the first PSM release, and each subsequent threshold crossing spends one additional $x\in\mathbb{R}^2$6. Timestamps that merely reuse $x\in\mathbb{R}^2$7 incur $x\in\mathbb{R}^2$8-GeoInd because no new information is released. By contrast, independent per-timestamp perturbation would spend

$x\in\mathbb{R}^2$9

over a trajectory of length $r$0. TR-PSM therefore saves

$r$1

whenever $r$2 (Seeam et al., 4 Aug 2025).

The paper also gives a structural privacy argument beyond budget accounting. Independent perturbation yields a one-to-one sequence of noisy points, which still preserves temporal linkage. TR-PSM instead weakens sequential distinguishability because several nearby true locations map to the same released noisy point. In the paper’s phrasing, the mechanism creates many-to-one mappings that improve trace-level privacy (Seeam et al., 4 Aug 2025).

The threat model assumes a trusted client device and an honest-but-curious server, with a passive eavesdropper also considered. The server is not trusted to preserve privacy and may monetize or leak location data. TR-PSM is implemented client-side, so the server receives only the privatized stream; no server-side logic changes are required (Seeam et al., 4 Aug 2025).

4. Empirical behavior and operating trade-offs

PrivAR evaluates TR-PSM on two public datasets, Geolife and T-drive, and on the proprietary GeoTrace dataset, then validates it on a Pokémon-Go-style prototype. Geolife contains 1.05 million points with median step $r$3 meters and $r$4-second intervals. T-drive contains 809K points with median step $r$5 meters and $r$6-second intervals. GeoTrace was collected from five participants over about $r$7 km across walking, running, biking, and driving (Seeam et al., 4 Aug 2025).

The utility metric in the simulation study is Mean Normalized Error,

$r$8

for which lower values are better. Privacy is evaluated with estimated Bayes risk, for which higher values indicate more attacker uncertainty. In the prototype, AR-specific QoS is measured by catchable objects and accumulated loss (Seeam et al., 4 Aug 2025).

The principal empirical pattern is a three-way ranking. PSM has the best QoS, TR-PSM has slightly worse QoS than PSM, and TR-PSM still substantially outperforms Planar Laplace. In the trace privacy experiments at $r$9, Bayes risk decreases for all mechanisms as trace length increases, but TR-PSM degrades slowest. The paper attributes this to many-to-one mappings and reused outputs that weaken sequential correlation. On public datasets, TR-PSM increases Bayes risk by up to $\theta$0; on GeoTrace, it improves Bayes risk by up to $\theta$1 over baseline. The abstract reports that PrivAR improves QoS (Gamescore) by up to $\theta$2, while increasing attacker error by $\theta$3 over baseline with an additional $\theta$4 milliseconds runtime overhead (Seeam et al., 4 Aug 2025).

The threshold parameter $\theta$5 governs the central utility–privacy trade-off. If $\theta$6 is too small, minor GPS jitter can trigger unnecessary releases. If $\theta$7 is too large, updates are over-suppressed and stale-location error rises. The paper gives explicit MNE sensitivity values on Geolife: at $\theta$8, MNE rises from $\theta$9 at $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$0 to $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$1 at $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$2; at $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$3, it rises from $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$4 at $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$5 to $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$6 at $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$7 (Seeam et al., 4 Aug 2025).

Prototype measurements place TR-PSM in the sub-millisecond regime. Mean runtime per location update is reported as $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$8 ms on Galaxy A04, $$z=x+r(\cos\theta,\sin\theta), \qquad \theta\sim \mathrm{Unif}(0,2\pi).$$9 ms on Pixel 6a, and $\epsilon$0 ms on Galaxy S22. In the end-to-end prototype, applying PrivAR costs only $\epsilon$1–$\epsilon$2 ms, less than $\epsilon$3 of total end-to-end latency, which is dominated by network round-trip (Seeam et al., 4 Aug 2025).

5. Terminological scope and adjacent literatures

The abbreviation PSM is overloaded across multiple research areas, and TR-PSM in PrivAR is specific to the Planar Staircase Mechanism (Seeam et al., 4 Aug 2025). This differs from several unrelated uses.

In causal inference and epidemiology, PSM commonly denotes propensity score matching. Methodological reviews of COVID-19-related studies identify omissions in covariate-selection justification, propensity-score model specification, balance diagnostics, and estimand reporting, while later methodological work argues that the so-called “PSM paradox” should not be treated as a reason to abandon matching (Gu et al., 2024, Wan, 2024). That literature concerns design and reporting in observational studies, not real-time location privacy.

In cryptography, PSM denotes Private Simultaneous Messages. The quadratic-residue line of work shows that threshold predicates are symmetric Boolean functions that can be realized in the PSM model with information-theoretic privacy, but this is a different primitive stack from PrivAR’s client-side noisy release mechanism (Shinagawa et al., 2022). Functionally adjacent telemetry systems such as STAR address threshold aggregation reporting through VOPRF, secret sharing, and symmetric encryption rather than through Planar Staircase noise and displacement-triggered reuse (Davidson et al., 2021).

In optimization, PSM can denote the parametric simplex method for sparse learning (Pang et al., 2017). In sleep staging, PSM can denote a pressure-sensitive mat, used as a sensing modality in multimodal fusion with EOG; that work is relevant to bed-based sensing but explicitly contains no thresholded reporting stage for PSM-derived signals (Papillon et al., 7 Jun 2025). The PrivAR meaning is therefore narrow and should be interpreted within the LB-AR privacy setting (Seeam et al., 4 Aug 2025).

6. Limitations and open directions

TR-PSM’s main limitation is threshold sensitivity. Optimal $\epsilon$4 values vary with mobility pattern and update frequency and must be empirically tuned. The paper states that TR-PSM is most beneficial on denser traces such as Geolife, where there are many closely spaced updates, while the advantage is smaller on sparser trajectories such as T-drive (Seeam et al., 4 Aug 2025).

A second limitation is explicit session-level budget management. If a threshold crossing occurs when

$\epsilon$5

the mechanism returns BudgetExhausted. The paper therefore requires either a sufficiently large session budget or session restart logic at the application layer (Seeam et al., 4 Aug 2025).

A third limitation is that TR-PSM trades some freshness for trace privacy. When the user moves quickly or when $\epsilon$6 is large, reuse of $\epsilon$7 can produce stale reports that degrade AR quality. The results show that this degradation is typically modest relative to Planar Laplace, but it is intrinsic to the mechanism (Seeam et al., 4 Aug 2025).

The paper does not present a learned or theoretically optimized thresholding policy. It explicitly suggests tuning $\epsilon$8 based on application update rate and movement characteristics. This suggests that adaptive threshold selection remains open. A plausible implication is that future work could preserve the exact release rule of TR-PSM while modifying how $\epsilon$9 or $\mathbf{x}=(x_1,\dots,x_T)$0 is chosen across sessions or mobility regimes, but that extension is not part of the reported method (Seeam et al., 4 Aug 2025).