Submitted:
05 October 2026
Posted:
09 October 2026
You are already at the latest version
Abstract
A detector records an event through a succession of physical changes, from its initial response to the recovery of its working state. We construct a complete finite cycle on the local physical--observation phase space of PODA. An incoming carrier traverses fixed interaction regions, with its arrival, speed, and energy transfer governed by an autonomous Hamiltonian bounded below. Six canonical gates write a smooth event profile, amplify the pointer separation, export the record, reverse the working displacement, and exchange the remaining pointer state with a prepared blank. Their exact reference map identifies the selector and mechanical recoil left by record export. The retained core returns on the event-cell plateaus; narrow transition strips contribute a bounded probability of failure and a recoil that depends on the resolution. With a positive free Hamiltonian present throughout the passage, energy conservation and a finite-time perturbation estimate establish faithful writing, finite retention, and return within prescribed tolerances. An explicit binary realization admits nonzero preparation errors, and independent trajectory integrations check the equations and the endpoint map. These results complete the single cycle for the specified local model and incoming family. The calculation follows the archive, carrier, and waste through the output section, making explicit the resources consumed by reset. Preparation of fresh blanks, replenishment of those resources, and storage beyond the proven holding interval remain physical requirements of continued operation.
Keywords:
PODA
; M-event
; autonomous detection
; canonical backaction
; finite-time retention
; detector reset
1. A Detector Cycle for a Single Event
A recorded event outlives the interaction that first produces it. The pointer moves, energy passes between the detector and its input, and a readable trace remains as the apparatus returns towards its ready state. In a reversible description, the information displaced during this return must be carried by other degrees of freedom. A complete account of the measurement therefore follows both the surviving record and the state left in the apparatus. In PODA, these changes take place within a common variational system for the physical state, the observation state, and the event selector.
The dual-axis formulation distinguishes the physical state from the observation state [1]. Its M-event extension represents individual outcomes by cells in a transported selector circle [2], while canonical readout exposes the conjugate reaction that can remain after an event is recorded [3]. Conditional Haar preparation assigns the required probabilities to the event cells. This probability law must still be joined to a writing interaction. The task is to construct an autonomous evolution that produces a resolved record, preserves it for a specified interval, and returns every retained part of the detector with a controlled error.
Preserving a readout variable and preserving the entire measured system are different requirements. Quantum nondemolition measurement makes this distinction precise for specified quantum observables [4]. Here the selector, its conjugate charge, and the record registers belong to a classical canonical extension of the Born interface. Their disturbance follows from the joint Hamiltonian, and the resulting recoil must be included when the detector returns to its ready state.
The distinction between an outcome law and a change of state is already central to measurement theory. Davies and Lewis introduced instruments to describe joint and conditional probabilities in repeated measurements [5]; Ozawa connected completely positive instruments to interactions between a quantum system and its apparatus [6]. These results specify what an operational description must retain. In the present event model, a selector determines the outcome, and the remaining task is to follow the physical writing interaction and the state from which a later event would begin.
Macroscopic limits address another part of this problem. Hepp’s soluble models obtain distinct pointer sectors and reduction relative to local observables within an infinite-system framework [7]. A finite detector requires quantitative separation at a finite time. We therefore formulate completion in terms of arrival, a readout margin, a holding interval, and a return tolerance, each evaluated along the same trajectory.
The timing of the operations also has a physical cost and a dynamical effect. Hamiltonian models of computation and conservative logic established how a sequence of operations can arise from the motion of an enlarged system [8,9]. More recent work on finite quantum clocks treats the controller’s energy and backaction as part of autonomous operation [10,11]. The construction below follows the same requirement of explicit control degrees of freedom in a local canonical setting: the carrier that orders the gates exchanges energy with the detector and may change its speed during the passage.
We give a local realization of this evolution over a finite interval. A physical carrier, with coordinate c, momentum P, and finite positive kinetic energy, brings an output register and a prepared blank through fixed interaction regions. Arrival initiates writing; continued passage brings the carrier through amplification, record export, and reset. The spatial arrangement determines the order of the operations, while the carrier dynamics determines their timing. Once the initial state is prepared, the passage requires no externally timed pulses.
The reset condition refers to a definite set of retained variables: the physical and observation base coordinates, their conjugate momenta, the selector pair, and the working pointer. Together they form the detector core. The carrier leaves with the archive and the used blank, whose final states retain the changes exported from that core. Following their energy and information through the output section is part of the reset calculation; the resources themselves are consumed by the passage. Reversible transfer and information erasure have different thermodynamic requirements [12,13]. The present energy balance follows one passage and its outgoing resources; a thermodynamic work cost would additionally require a model of blank preparation and the surrounding reservoirs.
The five stages in Table 1 provide the criteria for completion. The theorem below satisfies them on a specified family of incoming states, with explicit margins for readout and return. A small set of phases near the event-cell boundaries falls outside the guarantee. Those trials remain in the outcome probabilities, so that the record law includes the cost of finite resolution.
2. Joint Geometry and the Retained State
Let be coordinates on an admissible physical–observation domain, with selector angle , conjugate variable q, and real selector connection . Throughout this paper, q denotes the selector charge. We use the following local first-order action:
Here is the working pointer, the export register, and the incoming blank. The registers and carrier are physical degrees of freedom, while the observation coordinate s remains an independent component of the joint base.
The canonical base momentum is . A change in q therefore changes the mechanical covector even if the canonical momentum remains fixed. For a base Lagrangian and a record Hamiltonian , variation of the same action gives
Here , with both the physical and observation coordinates included in the variation. The charge recoil and the direct derivative of the detector interaction combine into the covariant force in the base equation.
Choose a local origin for the event cells and define . A change of fibre frame acts by , , and , with and for the canonical and mechanical covectors. The relative phase and the full action are invariant when the Hamiltonians transform as scalars. Any identification of this real selector connection with a matrix-valued response connection requires a representation specifying their relation.
For the explicit example we use the inherited binary chart
Here is the physical coordinate and the observation coordinate. At the balanced interface , the two event cells are semicircles. The construction retains this nonzero mixed curvature and treats the observation coordinate as an independent state variable.
3. An Incoming Carrier and Fixed Spatial Interactions
Collect the canonical variables other than in z, and let J be their canonical symplectic matrix. Each interaction occupies a separate spatial region, described by a twice continuously differentiable profile with unit integral. The profiles have disjoint supports. For generators , the carrier coupling and total Hamiltonian are
Here denotes the carrier interaction, distinct from the selector connection a. The profiles and their coefficients are fixed constitutive data, and sets the relative timescale of free evolution in this realization. Its value belongs to the chosen implementation and has no status as a universal PODA constant. Restoring dimensions requires the corresponding masses, register stiffnesses, and interaction strengths.
The Hamiltonian in Eq. (5) is autonomous and bounded below, with free evolution retained throughout the carrier passage. The limit admits an exact solution whose map supplies the reference for the error bounds at nonzero . These bounds account explicitly for the free evolution.
Set . Hamilton’s equations give
As long as , the carrier position can be used as the independent variable:
When , k is constant. A complete passage through gate j then produces the canonical flow of for unit flow parameter, since and . The spatial integral fixes the strength of the operation; the incoming carrier state fixes how long it takes.
In the incoming region, and h is independent of c and . A carrier prepared at with therefore reaches the first interaction region at
For , the free incoming motion never reaches the interaction region, so the trial does not trigger. Arrival with positive momentum is only the beginning of writing: any subsequent failure to produce a resolved pointer record is counted separately.
Before arrival the phase evolution has the skew form
where Z contains all non-phase variables. Conditional Haar preparation is consequently preserved at . Appendix A gives the general proof, a counterexample for phase-dependent stopping, and the sensitivity of a transverse crossing. Uniform bounds on the duration of triggered trials require . Positive incoming momenta arbitrarily close to allow arbitrarily slow arrivals and hence admit no uniform finite-time guarantee.
The energy available to the registers follows directly from conservation of the total Hamiltonian:
At the incoming and outgoing sections, , and the mechanical and canonical carrier momenta coincide. A net increase in the target energy is therefore supplied by a corresponding loss of carrier kinetic energy.
This exchange also changes the time spent in later interaction regions. In quasi-autonomous quantum thermal machines, Frenzel, Jennings, and Rudolph likewise retain the clock degrees of freedom and the backaction associated with energy extraction [14]. Here the energy identity and carrier equation provide the corresponding bookkeeping for a detector passage: both timing and record formation respond to the energy transferred to the registers.
4. Writing, Amplification, Export, and Reset
4.1. A Smooth Profile for the Event Cells
To read the event cells through a smooth canonical interaction, a register with finite resolution must interpolate near their boundaries. On the balanced chart, choose a transition half-width and define
The profile agrees with the event sign on each plateau, with its derivative confined to strips of half-width around the two cell boundaries. Their total conditional Haar measure is . If the partition moves or its cells have unequal weights, the profile must follow that partition; a translation of the origin alone cannot change the cell weights. The quantitative example below uses the balanced binary interface.
4.2. Six Canonical Gates
Let denote the initial pointer displacement and the displacement gain. Table 2 lists the operations in their order along the carrier path. Amplification and export are separated by a positive retention gap.
The dilation increases the separation of the pointer coordinates and reduces their conjugate momentum scale. In the reference system, where the free energy vanishes, its effect is a displacement gain. With present, the accompanying changes in pointer and archive energy enter the balance in Eq. (11). The gain is set by the design of the interaction; it is not a physical constant determined by the event geometry.
Proposition 1
(Complete reference map). For , fixed base coordinates, and an arbitrary initial register state, the six gates give
The base coordinates remain unchanged. The final expression also covers an explicitly base-dependent profile; for the profile in Eq. (12), the derivative at fixed α is zero.
Proof.
Writing translates R by and changes q by . Amplification carries the pointer to and its momentum to . Export then adds the amplified coordinate to E while subtracting from the working momentum. Reversing the dilation leaves , so the inverse write restores and changes the charge by . Adding the two charge increments gives Eq. (14). The last gate exchanges the full working pair with the prepared pair, with the signs shown in Eq. (13). The corresponding calculation for the canonical base momentum, together with , yields Eq. (15). □
On each plateau, and all relevant derivatives vanish. The reference map therefore restores the base and selector pair exactly, replaces the working pointer with the prepared blank, and leaves the event in the outgoing archive. For matched initial resources, the archive centers are separated by . Initial uncertainties satisfying and give deviations from bounded by
Reversing the write alone leaves in the apparatus. Reset also requires the exchange gate, which transfers this residual from the retained pointer to the outgoing blank. All six operations are invertible: the archive and waste carry the information needed to reconstruct the initial state. Export before reversal has the same organization as Bennett’s reversible computation: a copy of the result survives while the working operations are undone [15]. In the present canonical realization, the conjugate momentum transferred by copying also has to be followed through the reversal and exchange.
Returning the working pointer by exchange consumes a prepared blank. Replenishing that blank is a separate physical operation whose cost depends on the preparation and the reservoir. Reeb and Wolf’s finite-reservoir analysis makes this dependence explicit in quantum erasure, where the correction to the Landauer bound depends on reservoir size [16]. Our map identifies the outgoing waste and the state to be replenished; it supplies the mechanical part of that accounting before any reservoir model is chosen.
4.3. The Recoil Carried by the Transition Strips
In the balanced chart, with a fixed cell origin and no explicit base dependence, Eq. (15) reduces to
After export and reversal of the working operations, the residual selector recoil depends on the export momentum. The original working momentum has cancelled from this final reaction.
For an export momentum independent of the Haar phase, the exact reference map gives
Across each strip of length , the magnitude of the change in f is 2. The Cauchy–Schwarz inequality therefore bounds the integral on that strip below by . Writing the total strip mass as gives
For a fixed nonzero second moment of the export momentum, narrowing the transitions reduces the phase fraction they occupy at the cost of greater recoil. This bound depends on the prepared resources: in the classical ideal , the residual vanishes, so the estimate does not imply a universal positive lower bound.
5. A Finite Cycle with the Free Dynamics Retained
5.1. A General Continuation Estimate
Let denote the exact reference trajectory over a carrier displacement ℓ. In fixed, dimensionless canonical coordinates, choose a compact tube of radius r around this trajectory. Measure phase differences by a continuous lift along the two trajectories, or by the corresponding local circular distance. Suppose , , and throughout the tube. If
energy conservation keeps the carrier moving forward for as long as the target trajectory remains in the tube. For identical initial data on the entry section, Eq. (8) and Gronwall’s inequality give
The initial section has been placed at . If the bound at ℓ is strictly less than r, the trajectory cannot reach the tube boundary, and the passage takes at most . For any particular realization, the tube and its derivative bounds must be verified. A smooth finite reference trajectory with a positive tube margin admits a nonempty interval of positive satisfying these conditions.
On the plateaus, the affine structure of the ideal target equations permits a sharper bound than the exponential estimate. Let be the homogeneous propagator. For amplification followed by copying, inverse amplification, and the orthogonal exchange, it satisfies
Writing acts as a translation on a plateau, while amplification and its inverse vary the scale between 1 and . For an interval that crosses the export gate, the nontrivial entries are
The row and column sums are bounded by , as they are for intervals contained within a single gate. The other retained coordinates contribute identity blocks, and the orthogonal exchange preserves the Euclidean operator norm. Applying variation of constants gives
This estimate applies while the actual trajectory remains on the same plateau. For the family constructed below, the energy bound and the margin from the cell boundary establish this condition by a first-exit argument.
5.2. An Explicit Positive-Energy Realization
Use the canonical order
and the free Hamiltonian
The base kinetic term is . In the displayed gauge, and ; a change of frame requires the covariant expression in terms of the mechanical covector. The mixed geometry therefore enters both the free equations and the recoil from readout. Equation (26) specifies a local constitutive model in a fixed chart and scale; no general form for PODA detectors is assumed.
Prepare the balanced base and charge as
On a plateau, the prepared base is an exact invariant equilibrium of the full model. The charge remains constant, while
Substitution of Eq. (27) into the complete Hamiltonian equations makes the physical and observation derivatives vanish. Their equilibrium thus follows from the dynamics. Trajectories in the transition strips generally leave this invariant family; the numerical calculation follows their full equations of motion.
At the arrival section, set , , , , , and prepare the resources within the bounds
The six gate supports are, respectively,
The gap is the working-record retention interval. Each profile on a unit interval is
It is , nonnegative, and has unit integral.
Write . The partial gate maps give for both outcomes and all preparations in Eq. (29). At export, for example, the largest simultaneous coordinate bounds are and , with small remaining momentum and blank contributions. A radius- tube on the invariant balanced family therefore satisfies
The contribution of the phase direction to the latter norm is , even though h is independent of .
Choose the arrival phase farther than from either boundary. The energy and drift estimates hold up to any hypothetical first exit from the plateau or the radius- tube, and rule out both exits. The bound consequently holds throughout the passage, with the positive free Hamiltonian retained.
Theorem 1
(One finite detector cycle). For Eqs. (5), (12), and (26) with the parameters and incoming family in Eqs. (27)–(30), let and let the arrival phase satisfy the stated boundary margin. Then:
- (i)
- the carrier traverses the complete detector in less than time units after arrival;
- (ii)
- throughout the retention gap, the amplified working record has the selected sign and magnitude greater than ;
- (iii)
- at export completion the archive has the selected sign, and at final core reset it differs from by less than ;
- (iv)
- after the carrier leaves, the archive retains the same resolved sign for at least five further time units, with magnitude greater than ;
- (v)
- the retained pointer returns within Euclidean distance of its zero ready value, while the base and charge equal their prepared values and the selector advance is bounded by Eq. (33).
A readout threshold of and a retained-pointer ready tolerance of therefore have strictly positive margins throughout their specified intervals.
Proof.
The preceding continuation argument establishes the carrier and state bounds. During the retention gap, the ideal working magnitude is at least ; allowing for leaves a magnitude greater than . The archive’s preparation error is , and its final error relative to is therefore below . After the interaction has ended, and the archive obeys
The pair norm is at most , so the additional five time units increase the record error by at most . The ready blank has norm at most ; with included, the returned pointer lies within of zero. The balanced base and charge remain exact invariants on the successful plateaus, while Eq. (28) bounds the selector advance. Each bound leaves a positive margin relative to the corresponding threshold. □
More generally, the retention guarantee holds for any satisfying
where is the required record magnitude and bounds the archive drift in the specified hold region. Conservation of the harmonic archive’s pair norm supplies this bound. The estimate controls a finite holding interval without assuming an attracting memory state or permanent storage.
The protection of a physical record also depends on its coupling to the surroundings. Zurek showed how an apparatus–environment interaction selects a pointer basis [17]; Brune and collaborators subsequently observed the progressive decoherence of distinct field components in a cavity meter [18]. Those studies resolve the loss of coherence and the selection of record states. The bound here instead follows the archive coordinate inside a prescribed readout region for a finite interval. Connecting these descriptions in a material detector requires its environmental coupling and the resulting fluctuations of the stored coordinate.
5.3. Tolerance to Errors in the Full Prepared State
The balanced equilibrium makes the constants explicit. Its successful trajectories persist in a small neighborhood of the full canonical target state, leaving a positive completion margin for finite preparation errors.
Corollary 1
(An open prepared family). Fix the arrival carrier momentum at and take . Let the target-state preparation differ from the family in Theorem 1 by a Euclidean error of at most . With the cell center independently calibrated, the cycle has total target-state error below , and the returned core lies within of the nominal zero-pointer ready state. The archive retains magnitude greater than throughout the specified five-unit hold. A reference phase guard of half-width
at the two boundaries suffices. Its conditional Haar mass is below .
Proof.
Take the same radius- tube around a plateau reference trajectory, now in the full canonical target space. Within this tube, , , , and . Hence . Combining these bounds with gives
On a plateau, the base directions add identity blocks to the gate propagator, so remains valid in the full target space. Variation of constants now includes the initial discrepancy and gives
The carrier completes its passage in less than time units, and the phase error during the gates is at most . Neither the tube boundary nor a plateau boundary can therefore be reached, which completes the continuation argument. Adding the reference blank norm gives the ready-state bound. The previous retention estimate also applies: after output, the archive evolves as an independent harmonic pair.
If the observation coordinate differs from its balanced value, the inherited binary weight is . Center the positive cell at the calibrated origin and give it width . Its endpoints then differ from the balanced locations by at most , which lies within the transition width already excluded. The reference profile and the actual initial cell consequently assign the same label throughout the successful set. Conditional Haar measure yields the stated exclusion bound. □
For the perturbed family, the probability comparison uses the actual ; equal half-weights apply to the exactly balanced preparation. The tolerance established here belongs to the same detector, with its original state variables and gates, and requires no additional noise law.
6. The Recorded Probability Law and Its Error Budget
Let denote the event selected at carrier arrival. The subsequent archive label is defined for every arriving trial. A trial is unresolved if it fails to reach the output section by the declared deadline, leaves the model’s operating domain, or produces an archive below the resolution threshold. In the remaining trials, is the resolved archive sign. A correct archive label can nevertheless accompany a failed cycle if the core is excessively disturbed. For the successful phase set of Theorem 1, conditional Haar preparation and arrival independent of phase give
for the certified example. This bounds the probability of lying outside the guarantee and hence covers incorrect labels, unresolved records, and excessive error in the return of the core. Some trajectories in the transition strips may still complete the cycle successfully.
On , the recorded and selected events agree: . For any set I of finite arrival times, this gives
Preparation and calibration errors can be included in the same estimate. Suppose the prepared state exceeds the declared input bounds with probability , and an independent calibration bounds the uncertainty in the phase origin by . A conservative extension then replaces the right-hand side of Eq. (39) by . A bound on readout error must also be subtracted from the available record margin. These extensions require calibrated bounds and do not, by themselves, specify a physical noise law.
The inequality carries the inherited event-cell law through to the archive, with an explicit allowance for errors in writing. In the present example, . Unequal cells admit the same argument once their dependence on the base state and the motion of their boundaries have been specified. Their kernels remain inputs to this construction; the binary prototype does not derive them anew.
To include trials in which the carrier never arrives, suppose that every arriving trial satisfies the certified preparation conditions apart from its phase. Write for the upper bound on the complement of the guarantee set; this quantity need not equal the fraction of unresolved archives. If the prepared ensemble has a fraction of carriers that do not arrive, then
Every trial therefore contributes to one of four categories: a resolved positive record, a resolved negative record, an unresolved record after arrival, or no trigger. All four enter the normalization. Discarding either of the last two would change the operational probability under comparison.
The return theorem concerns a finite cycle with specified preparation. Stability under repeated use requires a further analysis, since the theorem establishes no contraction of arbitrary incoming disturbances. The reference swap, for example, transfers a fresh ready distribution to the working pointer and exports its previous state. Repeated operation consequently depends on the statistics and correlations of the replenished resources. The derivative of the complete return map, including every retained physical and observation variable, is the appropriate object for studying that stability. The selective reset analysed in Ref. [3] concerns a prescribed-drive realization. Its response stability condition does not replace the complete return calculation for the autonomous system considered here.
7. Numerical Verification of the Autonomous Equations
The accompanying code integrates Eq. (7) in physical time with all fourteen canonical coordinates retained. It uses the DOP853 solver through SciPy [19]. An independent integration uses carrier position as its independent variable, allowing the two solutions to be compared at common sections. Both calculations evolve the canonical carrier momentum. Energy conservation and the mechanical carrier momentum are monitored separately, so a prescribed carrier speed is never used to enforce the passage.
The certified representative runs begin at the arrival section . Their resource values equal the positive bounds in Eq. (29), with the exception . A separate set of diagnostics follows the passage from launch to output, beginning at with and the same resource values specified at launch. Free rotation during the incoming passage slightly changes the registers before arrival, so those diagnostics are not used to establish the guarantee for the certified arrival family. Both sets of calculations include the two plateau outcomes and the smooth transition bands. At , every component of the endpoint map is checked against Proposition 1. Positive values of test the effects of free evolution on the finite holding interval and the accuracy of return. The analytic guarantee is established at ; larger values continue the numerical example without extending the uniform theorem to a larger family of inputs. All input scales in these calculations are dimensionless design choices and have not been identified with measured detector parameters.
Table 3 summarizes the runs begun at arrival, taking the largest deviation of the retained core and the smallest archive magnitude over the two plateau phases 0 and . This deviation compares the retained base and selector with their arrival values, and the working pair with the incoming blank pair. The distance of the working pair to the zero ready state is a separate quantity; in these plateau runs it remains approximately .
The four checks of the ideal map include two phases in the transition strips. All components agree with the analytic endpoint map, with a maximum error below . For the nine arrival cases with positive , the largest monitored energy discrepancy is below . The independent integration in carrier position agrees with the integration in physical time at the endpoints to within . These comparisons establish the numerical consistency of the calculations; the finite-error guarantee follows from the theorem.
A broader scan uses and 256 equally spaced midpoint phases, all prepared exactly at arrival. Applying the archive threshold gives 126 positive records, 126 negative records, and 4 unresolved records. Every resolved record has the correct sign on this grid. The phase guard covers 244 samples; the remaining 12 fall outside the guarantee and fail the stated test for return of the core. Eight of those twelve nevertheless leave a resolved archive with the correct sign. The scan thus separates successful recording from successful completion of the entire cycle.
For an independently specified mixture in which of the carriers arrive and do not, the calculated probability masses on this grid are
These masses sum to one. Before writing, the ideal event masses are , , and . Failure of the complete cycle has mass in the mixed grid, exceeding the mass of unresolved archives because a correctly recorded event can still disturb the core beyond tolerance. No trials are selected or discarded by phase. The allowance conservatively accounts for deterministic quadrature near the cell boundaries; it has no interpretation as experimental uncertainty. The maximum energy discrepancy in this less costly scan is below . The refined representative trajectories supply the tighter numerical checks reported above.
Arrival and pointer resolution remain separate events throughout the scan. Every positive carrier in these examples reaches the writing region, including carriers with phases in the transition bands for which the pointer never crosses the readout threshold. The population of negative carriers is counted separately. Since the frequencies are obtained by deterministic quadrature over phase, the associated grid errors do not constitute statistical confidence intervals.
Figure 1.
Autonomous passage through the detector for the two balanced outcomes, starting at arrival with . Fixed spatial interactions make the incoming carrier activate writing, amplification, a finite hold, archive export, and reset of the core in succession. The curves solve the complete Hamiltonian equations, including the positive free term. The archive persists after the working pointer is reset, while the outgoing resource variables carry the corresponding changes.
Figure 1.
Autonomous passage through the detector for the two balanced outcomes, starting at arrival with . Fixed spatial interactions make the incoming carrier activate writing, amplification, a finite hold, archive export, and reset of the core in succession. The curves solve the complete Hamiltonian equations, including the positive free term. The archive persists after the working pointer is reset, while the outgoing resource variables carry the corresponding changes.

Figure 2.
Checks of the reference map and its realization with free dynamics retained. Panel (a) varies ; panels (b)–(d) use . The phase scan includes the transition strips, showing both unresolved records and recoil of the selector. The reference recoil is compared with the analytic export formula, and errors at finite are evaluated at the same output section of the carrier. All trajectories are calculated from the specified model.
Figure 2.
Checks of the reference map and its realization with free dynamics retained. Panel (a) varies ; panels (b)–(d) use . The phase scan includes the transition strips, showing both unresolved records and recoil of the selector. The reference recoil is compared with the analytic export formula, and errors at finite are evaluated at the same output section of the carrier. All trajectories are calculated from the specified model.

8. Experimental Tests and the Completed Cycle
The construction gives a set of couplings and quantitative consequences to test in a physical realization. Translating them into an experiment requires independent calibration of the carrier coordinate, the working and output registers, the blank preparation, and the relative event phase. Response comparisons must establish these identifications over the range in which the proposed interactions can be realized. The present calculation provides the model for such a test; its implementation in a material detector remains to be established.
Carrier arrival supplies a direct timing test. At fixed incoming distance, the pre-interaction travel time varies as , with the two preparation derivatives in Eq. (9). Arrival marks the onset of writing. A resolved pointer appears later, after a delay determined by its trajectory. Recording both times, as in the simulation, would allow an experiment to distinguish transport to the detector from the formation of a readable response.
The archive displacement tests the writing and amplification stages. On a successful plateau, the reference displacement is together with the prepared offsets ; the retained free dynamics adds the bounded correction derived above. Varying g or at fixed preparation and readout threshold tests both the gain and the resulting change in the unresolved fraction. The offsets must be measured independently of the output records used for this comparison, otherwise an apparent agreement could be absorbed into their fitted values.
Trials near a cell boundary reveal the reaction left by record export. In the reference map, the selector-charge recoil after reset is , accompanied by the mechanical recoil in Eq. (15). At the balanced interface the physical reaction is half the charge change; the full dynamics determines the correction due to the finite duration. Reversing the prepared export momentum reverses the reference recoil. A scan of the transition width would also test the second-moment bound in Eq. (18), provided the export-momentum distribution is measured for each preparation. The resolution and the recoil can then be compared under the same resource conditions.
The energy balance links these tests to the carrier. Its loss of kinetic energy equals the increase in , including the archive and waste as well as the working pointer. Measurements confined to the visible pointer displacement would leave part of this transfer unaccounted for. Delayed archive readout then tests the holding-time bound, while tomography of the retained core tests return to the ready state. That reconstruction must include the selector-conjugate channel and the physical and observation momenta, since the pointer position alone cannot reveal the residual reaction.
The detector cycle thus connects the joint action to a persistent record and a returned core through an explicit dynamical construction. The action fixes the covariant measurement reaction, and the six gates show how writing and reset can coexist with that reaction. The finite-error theorem carries the reference construction into a regime with positive free dynamics and nonzero preparation tolerances. Its readout, retention, and return margins can be compared with experiment once the material variables and incoming resources have been calibrated.
For this local Hamiltonian and incoming family, the calculation completes the single cycle: the initial resources evolve into a readable archive, an explicitly determined outgoing state, and a detector core within its ready-state tolerance. Phases near the cell boundaries remain in the failure bound. The result is a realization for the specified detector model, with a finite holding interval and a supplied blank. Extending operation to further cycles would require fresh resources; indefinite storage would require a separate retention mechanism.
Data Availability Statement
The accompanying source package contains the manuscript, vector figures, numerical results, build instructions, and the programs for independent verification of the autonomous cycle and the auxiliary first-passage examples. The trajectories are computed from the specified equations, without experimental data or a fitted noise law. The appendices include the first-trigger proof and its diagnostic examples, completing the mathematical account from the selected event to the recorded outcome.
Appendix A. General First-Trigger Probabilities and Timing Response
Appendix A.1. Conditional Haar Preparation
Let collect the complete non-phase initial data, including and the retained environment. Under deterministic dynamics, fixes both the non-phase trajectory and the accumulated selector increment. An effective random drive can be included by conditioning on its entire realization, provided the conditional preparation assumption continues to hold.
Suppose the initial selector is conditionally Haar distributed:
The theorem takes conditional Haar preparation as an input, consistent with the cell-volume construction of Ref. [2]. Dynamical formation of this preparation has been established for a specified affine reservoir model [20]; that reservoir is not part of the present cycle. Integrating Eq. (10) gives
Let partition the selector fibre into finitely or countably many cells, jointly measurable in and with boundaries of Haar measure zero. Set . At a finite trigger, the selected label is defined by
For continuous pre-trigger evolution, the left limits equal the trigger values. Equation (A3) defines the selected event label; the construction above establishes its agreement with the exported record on the successful set.
Theorem A1
(First-trigger probability law). Assume Eqs. (A1) and (A2). Let τ depend only on the non-phase trajectory, and suppose and the event partition are defined whenever . For every Borel set ,
On , assign a null label and define the integrand in Eq. (A4) to be zero; no value of is required there. In particular, the selector at a finite trigger remains Haar distributed conditionally on .
Proof.
Conditioning on fixes the trajectory, , and . Translation invariance of Haar measure gives, on ,
For a fixed trajectory, replace E by , multiply by , and take the expectation to obtain Eq. (A4). Conditional Haar invariance therefore supplies the result without an optional-stopping theorem. □
When the inherited kernel has the form , the same kernel governs selection at the autonomous trigger. The theorem extends the prescribed-readout law to this trigger under the stated conditions, taking both the kernel and conditional Haar preparation as inputs.
Appendix A.2. Trials Without a Trigger and Conditional Samples
Summing Eq. (A4) over the outcome partition gives
Trials without a trigger remain in the experimental denominator. When every finite trigger is subsequently written without loss, Eq. (A6) also normalizes the recorded outcomes. Readout failures must be represented by an additional outcome or by a separately derived writing kernel.
Conditioning on the occurrence of a trigger may nevertheless change the distribution of non-phase states:
This average need not equal at a fixed initial state. The selector can remain unbiased while the trigger selects a particular sample of base states.
Appendix A.3. A Phase-Dependent Trigger
A uniform phase distribution at every fixed time need not remain uniform at a time chosen from the trajectory. Consider
and take the trigger to be the first positive arrival at phase zero. Outside a null set of initial phases, and . For the half-circle ,
The crossing is transverse and can be written locally as , . Here the same phase determines both the trigger and the event, producing the change in probability. The independence assumption of Theorem A1 is therefore violated.
Phase zero is specified relative to the cell reference. Under a change of fibre frame,
the trigger reference transforms together with the cells, preserving event membership and cell measures. The construction is thus intrinsic to the circle fibre and assigns no preferred angular coordinate.
Appendix A.4. Sensitivity of the Trigger Time to Preparation
Consider a smooth autonomous flow whose finite first crossing remains on the same local branch as the preparation varies. Suppose the initial state lies strictly before the surface and the crossing is transverse:
The implicit-function theorem applied to gives
The numerator measures the trajectory variation at the unperturbed time . Evaluating the state at the perturbed event also requires the displacement accumulated over the change in trigger time,
The timing contribution must be retained when the outcome kernel varies along the trajectory.
A change in the calibrated surface adds to the numerator of Eq. (A12). For small centered preparation errors with covariance , write . To linear order,
Correlations between surface calibration and preparation errors enter through their joint covariance. As a crossing approaches grazing, becomes small and the linear timing approximation may fail. These expressions apply to smooth trajectories; Brownian first passage requires a separate analysis.
Appendix B. Independent First-Passage Diagnostics
The first-crossing calculation can be checked exactly with an auxiliary oscillator, independently of the detector construction. Let
where , . A positive crossing exists for and obeys
No crossing occurs for , and gives a grazing contact beyond the transverse formula. As , the crossing time remains finite while its sensitivity diverges, as shown in Figure A1.
For the probability check, take , , an amplitude uniform on , an independent Haar selector, and . The cell weight is prescribed as a diagnostic input, with no detector-specific derivation. Direct integration gives
Each triggered outcome has the unconditional probability obtained by integrating its cell weight over , with the trials without a trigger retained in the full sample.
The calculation uses 3000 midpoint amplitudes and 2048 equally spaced midpoint phases for each amplitude that triggers. The largest probability error at a fixed amplitude is , below the deterministic cell quadrature bound . The unconditional positive and negative masses are and , respectively; the no-trigger mass is exactly on this grid. These errors measure deterministic quadrature accuracy and have no statistical confidence-level interpretation.
Independent integration of the ordinary differential equation reproduces the analytic first-crossing time with maximum absolute error . Away from grazing, central differences of agree with Eq. (A16) to relative error . The phase-selected counterexample gives probability one at its trigger and one half at a fixed time. All results are dimensionless computational diagnostics; experimental records are not used.
Figure A1.
First-passage diagnostics for the auxiliary oscillator and circle examples. Independent trajectory integration verifies the analytic trigger time and resolves the increase in its sensitivity near grazing. With a phase-independent trigger, conditional event frequencies follow the assigned cell weights. A phase-selected trigger places the selector at the same phase on every trial. These examples test the probability and timing statements independently of the construction of a PODA detector action.
Figure A1.
First-passage diagnostics for the auxiliary oscillator and circle examples. Independent trajectory integration verifies the analytic trigger time and resolves the increase in its sensitivity near grazing. With a phase-independent trigger, conditional event frequencies follow the assigned cell weights. A phase-selected trigger places the selector at the same phase on every trial. These examples test the probability and timing statements independently of the construction of a PODA detector action.

References
- X. Meng, The Natural Dual-Axis Structure of Observation: Categorical Foundations and Consequences for Quantum Theory, Preprints (2026), version 3. [CrossRef]
- X. Meng, Deterministic Quantum Events on a Dual-Axis State Bundle: A Falsifiable Recurrence Test of Single-Event Continuity, Preprints (2026), version 1. [CrossRef]
- X. Meng, Measurement Backaction, Selective Reset, and Stability of M-Event Memory, Preprints (2026), version 1. [CrossRef]
- V. B. Braginsky and F. Ya. Khalili, Quantum nondemolition measurements: the route from toys to tools, Rev. Mod. Phys. 68, 1–11 (1996). [CrossRef]
- E. B. Davies and J. T. Lewis, An operational approach to quantum probability, Commun. Math. Phys. 17, 239–260 (1970). [CrossRef]
- M. Ozawa, Quantum measuring processes of continuous observables, J. Math. Phys. 25, 79–87 (1984). [CrossRef]
- K. Hepp, Quantum Theory of Measurement and Macroscopic Observables, Helv. Phys. Acta 45, 237–248 (1972). [CrossRef]
- P. Benioff, The computer as a physical system: A microscopic quantum mechanical Hamiltonian model of computers as represented by Turing machines, J. Stat. Phys. 22, 563–591 (1980). [CrossRef]
- E. Fredkin and T. Toffoli, Conservative logic, Int. J. Theor. Phys. 21, 219–253 (1982). [CrossRef]
- M. P. Woods, R. Silva, and J. Oppenheim, Autonomous Quantum Machines and Finite-Sized Clocks, Ann. Henri Poincaré 20, 125–218 (2019). [CrossRef]
- P. Erker, M. T. Mitchison, R. Silva, M. P. Woods, N. Brunner, and M. Huber, Autonomous Quantum Clocks: Does Thermodynamics Limit Our Ability to Measure Time?, Phys. Rev. X 7, 031022 (2017). [CrossRef]
- R. Landauer, Irreversibility and Heat Generation in the Computing Process, IBM J. Res. Dev. 5, 183–191 (1961). [CrossRef]
- T. Sagawa and M. Ueda, Minimal Energy Cost for Thermodynamic Information Processing: Measurement and Information Erasure, Phys. Rev. Lett. 102, 250602 (2009); erratum 106, 189901 (2011). [CrossRef]
- M. F. Frenzel, D. Jennings, and T. Rudolph, Quasi-autonomous quantum thermal machines and quantum to classical energy flow, New J. Phys. 18, 023037 (2016). [CrossRef]
- C. H. Bennett, Logical Reversibility of Computation, IBM J. Res. Dev. 17, 525–532 (1973). [CrossRef]
- D. Reeb and M. M. Wolf, An improved Landauer principle with finite-size corrections, New J. Phys. 16, 103011 (2014). [CrossRef]
- W. H. Zurek, Pointer basis of quantum apparatus: Into what mixture does the wave packet collapse?, Phys. Rev. D 24, 1516–1525 (1981). [CrossRef]
- M. Brune, E. Hagley, J. Dreyer, X. Maître, A. Maali, C. Wunderlich, J. M. Raimond, and S. Haroche, Observing the Progressive Decoherence of the “Meter” in a Quantum Measurement, Phys. Rev. Lett. 77, 4887–4890 (1996). [CrossRef]
- P. Virtanen, R. Gommers, T. E. Oliphant, et al., SciPy 1.0: fundamental algorithms for scientific computing in Python, Nat. Methods 17, 261–272 (2020). [CrossRef]
- X. Meng, Dynamical Formation of Born Equilibrium and the M-Event Fingerprint, Preprints (2026), version 1. [CrossRef]
Table 1.
Conditions for completing one detection cycle.
| Stage | Calculated quantity |
|---|---|
| Trigger | Carrier arrival time and the fraction of incoming states that never reach the interaction. |
| Writing | Agreement between the selected event cell and a resolved pointer region, outside a bounded error set. |
| Amplification | Pointer separation and its energy exchange with the carrier. |
| Retention | Persistence of the working and exported records for specified intervals. |
| Reset | Return of the retained core within a stated tolerance, with archive, carrier, and waste states accounted for. |
Table 2.
The autonomous gate sequence. Each row is generated by a spatial profile of unit area.
| Operation | Generator | Principal canonical map |
|---|---|---|
| Write | , with joint recoil. | |
| Amplify | , . | |
| Export | , . | |
| Undo gain | Inverse amplification. | |
| Undo write | Inverse working displacement with the current . | |
| Exchange | ; . |
Table 3.
Autonomous evolution from arrival to output. The archive minima also cover five time units after output. All quantities are expressed in the declared dimensionless scales.
Table 3.
Autonomous evolution from arrival to output. The archive minima also cover five time units after output. All quantities are expressed in the declared dimensionless scales.
| Max. core deviation |
Min. archive magnitude |
Min. carrier momentum |
Max. cycle duration |
|
|---|---|---|---|---|
| 1.514998 | 1.999986 | 6.500021 | ||
| 1.514981 | 1.999858 | 6.500215 | ||
| 1.514761 | 1.998584 | 6.502147 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.