{"id":"e751baea-945e-4cf9-a149-1ceeee4a33ee","arxiv_id":"2412.10774","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":2.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"It reports an IoT parking prototype with IR, DHT22 and MQ-2 sensors plus MQTT app updates, but without experimental data supporting the claimed high accuracy.","lead":"This paper describes a prototype smart parking system: infrared, temperature, humidity, and gas sensors connected to an Arduino and Raspberry Pi, with an MQTT mobile app showing open spots. A generalist reader might look here for a simple IoT parking design, but the paper provides no measured evidence that the system performs as claimed.","discovery_kind":"incremental","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'high accuracy' claim is the load-bearing point; the paper's only quantitative sensor evidence shows IR accuracy degrading with light and dust, and no validation data establish the claimed accuracy.","rationale":"I agree with the reader's weakest assumption: the system's usefulness rests on reliable IR-based vehicle detection, and the paper does not establish that reliability. In fact, the paper's own Section IV.A.1 and Section V point in the opposite direction, reporting accuracy degradation under ambient light and conceding that dust, drift, and deterioration affect the sensors. The claimed 'high accuracy' is therefore not merely unproven; the internal evidence actively undermines it. Table III's unsupported comparison percentages are also damaging, but they are downstream of the sensing claim: even a perfect dashboard cannot deliver real-time availability if the slot state is wrong. There is no machine-checked proof, reproducible code, or parameter-free derivation to serve as independent support. This is not a matter of style or of disagreement with consensus; it is a missing falsifiable measurement that the paper itself indicates is necessary. I would keep the reader's REJECT verdict unchanged. The components and MQTT workflow are plausible as a prototype description, but plausibility does not ground the performance claim, and no experimental campaign with raw counts, ground truth, or error analysis is reported.","tokens_in":10143,"tokens_out":6702,"duration_ms":60071,"concrete_test":"Run a controlled validation trial with the exact detection chain (IR sensor to Arduino/Raspberry Pi to MQTT status change): record at least 200 slot-state transitions and 100 entry/exit events over one week under measured ambient illuminance, with video-based ground truth. Compute per-condition accuracy, precision, recall, and F1 at low light, midday sunlight, and after several days of dust accumulation. If F1 is below 0.95 in any realistic condition, or if the authors cannot supply the raw data behind Fig. 3(a), the claimed high accuracy is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central assertion—'This system demonstrates high accuracy in vehicle detection and environmental monitoring'—cannot hold unless each IR sensor reliably distinguishes occupied from vacant states at the entrance, exit, and every slot under realistic parking conditions. The paper supplies no direct evidence for that reliability. Section IV.A.1 presents Fig. 3(a) as an accuracy-versus-light-intensity curve, but it reports no measurement protocol, sample size, ground-truth labeling, or raw counts; its only stated result is that accuracy falls substantially as ambient light increases. Section V concedes that IR sensors 'can lose precision because of dust and light' and require frequent calibration, and that the DHT22 and MQ-2 sensors also drift or deteriorate. No confusion matrix, precision/recall, or aggregate accuracy figure is given anywhere. Since real-time slot availability, automated gate control, and user satisfaction all depend on correct vehicle detection, this is the single load-bearing evidential gap in the paper's central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper describes an IoT-based smart car parking system comprising IR sensors for vehicle detection, DHT22 sensors for temperature/humidity, MQ-2 gas sensors, Arduino Uno and Raspberry Pi controllers, servo-motor gates, an OLED display, and an MQTT-connected mobile application. It presents a system architecture, pseudocode, workflow, and several performance-related figures and tables. The authors claim that the system demonstrates 'high accuracy in vehicle detection and environmental monitoring' and that it outperforms traditional non-IoT solutions in integration, automation, and responsiveness. However, the quantitative support for these central claims is missing: the reported results lack experimental methodology, sample sizes, ground-truth comparisons, error bars, and measurement descriptions, and the paper's own limitations section states that the IR sensors lose precision due to dust and light and require frequent calibration.","tokens_in":10309,"tokens_out":2812,"duration_ms":28456,"significance":"If the system achieved the claimed high detection accuracy and efficiency gains, it would represent a useful, low-cost contribution to smart parking infrastructure in urban areas. The paper's strength is a coherent architectural integration of widely available components (IR, DHT22, MQ-2, Arduino, Raspberry Pi, MQTT) and the inclusion of a mobile application for real-time slot status. However, the significance is substantially undermined because the central accuracy and efficiency claims are asserted rather than demonstrated. The empirical sections provide no reproducible protocol, no statistical measures, and no validation against ground truth; the only concrete sensor-performance statement is a degradation curve. The paper does not currently meet the evidentiary standard for a systems paper claiming measurable performance improvements.","major_comments":[{"comment":"The central claim of high accuracy in vehicle detection is unsupported. Fig. 3(a) reports an IR-sensor accuracy curve versus ambient light intensity, but no measurement protocol, sample size, ground-truth labeling, or raw counts are given, and no absolute accuracy values are stated. The text says accuracy dropped by 52% to 98% as light increased, which is a degradation, not evidence of high accuracy. Because every downstream feature—real-time slot availability, gate control, user satisfaction—depends on reliable detection, this missing validation is a load-bearing gap.","section":"§IV.A.1, Fig. 3(a)"},{"comment":"The efficiency comparison between IoT and non-IoT systems is presented as a table of percentages (e.g., Automation 40% vs 100%), but the paper provides no definitions of these metrics, no measurement procedure, no sample size, and no baseline description. Without such information, the numbers cannot substantiate the claim that the proposed system outperforms traditional systems. The comparison must either be backed by controlled experiments or clearly labeled as qualitative/subjective.","section":"§IV.B, Table III"},{"comment":"The limitations section explicitly states that IR sensors 'can lose precision because of dust and light' and require frequent calibration, and that DHT22 and MQ-2 sensors drift or deteriorate. This admission directly contradicts the abstract and contribution list's assertion of 'high accuracy in vehicle detection and environmental monitoring.' No data are provided to show that the system maintains its claimed accuracy under realistic operating conditions, so the central claim is not merely unproven but in tension with the paper's own statements.","section":"§V"},{"comment":"The Poisson-occupancy model, Little's Law, and the ventilation response-time formula are standard textbook results, but none of the parameters (λ, T_avg, r, ΔG) are measured or estimated from the system. The equations are decorative rather than analytical tools: they neither feed into the results section nor constrain the design. Either tie them to empirical data or remove them to avoid implying quantitative modeling support that is not present.","section":"§III, Eqs. (8)–(10)"},{"comment":"Fig. 6 and the associated narrative compare IoT versus non-IoT systems with qualitative descriptors ('advanced connectivity', 'quick response'), but no experimental setup, scenario, or controlled trial is described. The figure appears to be an illustration rather than a measurement, and as such it cannot support efficiency or user-satisfaction claims. The comparison should be repositioned as a feature listing unless actual comparative measurements are provided.","section":"§IV.B, Fig. 6"}],"minor_comments":[{"comment":"There are typographical errors, including 'Bangladeshover' (missing space), 'parking loT' (should be 'parking IoT'), and 'Fig. Fig.7(c)' (duplicate 'Fig.'). The paper also inconsistently uses 'MQ2' and 'MQ-2'.","section":"Throughout"},{"comment":"Phrases such as 'setting a new standard for smart parking systems' and 'seamless integration' are promotional rather than technical. They should be replaced with concrete, testable claims.","section":"Abstract and §I"},{"comment":"The related-work table lists strengths and weaknesses qualitatively, but several entries lack citations or page numbers in the reference list (e.g., [17] gives pages 'xx–xx'). This should be corrected for completeness.","section":"§II, Table I"},{"comment":"The DHT22 and MQ-2 results (Figs. 4 and 5) report observations such as temperature rising 0.4°C and gas levels ranging 1–25 ppm, but no indication is given of how many days of data were collected, where sensors were placed, or what environmental conditions prevailed. Adding this context would improve reproducibility.","section":"§IV.A.2 and §IV.A.3"}],"recommendation":"reject","confidential_remarks":"The paper's core claim is an empirical one about accuracy and efficiency, yet the manuscript provides no valid experimental evidence for either. The limitations section itself concedes the sensors are unreliable under realistic conditions. This is not a case of a sound result needing minor polish; the evidence base for the central claim is absent. I would not encourage a major revision unless the authors are able to supply a proper measurement study with ground-truth comparisons, error statistics, and a defined baseline, at which point the paper would be substantially different."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The one thing to know: this is a build-and-demo paper, not a validated research result. The genuinely new piece is bolting DHT22 and MQ-2 environmental sensors onto the standard IR-sensor-plus-microcontroller parking system, then tying gas levels to an exhaust fan. That is a routine extension, and the paper's own Table I shows at least ten near-identical prior systems.\n\nWhat the paper does well: the architecture is clearly described, the pseudocode is sensible, and the comparison table is honest about what earlier systems did and where they fell short. The equations in Section III are standard textbook material (Poisson, Little's Law, data rate, delay), so there is no circular derivation and no invented math. The figures show a real deployment area, and Section V openly concedes that IR sensors degrade with dust and light, DHT22 drifts, MQ-2 deteriorates, and the whole system depends on network stability. That limitations section is the most truthful part of the paper.\n\nWhere it falls apart: the central claim of 'high accuracy' and 'significantly improving parking management efficiency' is asserted, not shown. Section IV gives plots without sample sizes, error bars, ground-truth methodology, or raw counts. Figure 3(a) actually demonstrates the opposite of the claim: it shows IR detection accuracy dropping as ambient light rises. Table III lists efficiency percentages (65%, 90%, 100%) with no measurement protocol anywhere. There is no confusion matrix, no precision/recall, no comparison against a baseline, and no statistical summary. Given that the entire system's real-time slot availability and gate automation depend on reliable IR detection, this is a load-bearing evidential gap. The writing also has rough patches ('parking loT', 'backward and nerdy redundant MQ-2') and the abstract's 'setting a new standard' overreach matches nothing in the results.\n\nOn citations: the paper leans heavily on Rahman et al. self-citations, but they are mostly relevant IoT/blockchain background, not a hidden source of the claimed result. I would not call the citation pattern deceptive.\n\nThe bottom line: this is an undergraduate-capstone-level engineering demonstration. A reader interested in a quick survey of smart parking architectures or a parts list for a prototype might skim it, but it does not answer a research question and the performance claims are unsupported as written. It deserves a desk rejection, not referee time. If the authors return with a real measurement campaign, raw data, and a calibration-aware accuracy analysis, it could be worth a look then.","headline":"A competent but routine IoT parking build whose core 'high accuracy' claim has no experimental support; the honest limitations section is the most credible part of the paper.","tokens_in":10845,"tokens_out":1479,"would_cite":false,"duration_ms":16075,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that an IoT parking system built from IR sensors, DHT22 and MQ-2 sensors, an Arduino Uno, a Raspberry Pi, and an MQTT mobile app achieves accurate real-time slot detection and outperforms traditional non-IoT parking.","keywords":["Internet of Things","Smart Parking System","Arduino UNO","Raspberry Pi","MQTT Protocol","Real-time Monitoring","Environmental Monitoring","Mobile App Interface"],"falsifier":"Run the system for one week in an open parking lot with direct sunlight and dusty conditions, and compare each slot's IR-reported state against a video ground truth of whether a car is present; if false occupancy errors occur often enough to mislead users, the central accuracy claim fails.","tokens_in":9956,"feed_emoji":"🅿️","tokens_out":5369,"duration_ms":47088,"temperature":0.7,"pith_summary":"An IoT parking prototype combines IR sensors at each slot and at the gates, DHT22 temperature and humidity sensors, MQ-2 gas sensors, an Arduino Uno for gate and fan control, a Raspberry Pi for communication, and an MQTT link to an Android dashboard. The paper argues this integrated setup provides accurate, real-time slot availability and environmental safety, and that it outperforms traditional non-IoT parking systems in integration, automation, and responsiveness. A sympathetic reader takes away a concrete blueprint for a low-cost smart parking system that could reduce searching and emissions, provided the sensing layer stays reliable.","feed_headline":"Parking sensors stream open slots to a phone app","feed_subtitle":"IR, DHT22, and MQ-2 sensors feed an MQTT app with slot and air-quality updates as they happen.","key_machinery":"The load-bearing mechanism is the MQTT publish-subscribe loop. In this design MQTT, a lightweight publish-subscribe messaging protocol, treats each parking slot as a topic; the Raspberry Pi collects IR occupancy readings, publishes them, and the mobile app subscribes so the display updates when a car arrives or leaves. The Arduino Uno closes the control loop for gates, buzzer, and exhaust fan, while the DHT22 and MQ-2 sensors feed environmental readings into the same display and app. Equations in the paper formalize publish and subscribe behavior and model lot utilization with a Poisson occupancy formula and Little's Law, though those formulas are not fitted to data.","core_discovery":"The authors claim that a network of commodity sensors and microcontrollers is enough to turn a parking lot into a real-time information system. Each slot's IR sensor publishes occupancy to the Raspberry Pi, which forwards it over MQTT to the mobile app; the app then shows free slots in green and occupied slots in red. The same loop drives automated entry and exit gates and triggers an exhaust fan when the MQ-2 gas sensor crosses a threshold. The paper's central assertion is that this system demonstrates high accuracy in vehicle detection and environmental monitoring, significantly improving parking management efficiency and user satisfaction with automated, real-time updates.","pith_inferences":["The paper does not report a ground-truth trial of app-reported availability against actual lot occupancy; a direct field test is the natural next step.","The Poisson and Little's Law equations are presented as models but no data is fitted to them, so they are conceptual scaffolding rather than validated predictions.","The weak point is the sensor layer itself because IR accuracy drops with light and dust, so the advertised real-time accuracy would degrade in outdoor lots unless the calibration problem is solved.","Network dependence is built into the MQTT design; adding offline buffering would be required before the system could be trusted during outages."],"forward_implications":["Drivers can check open slots from the mobile app before entering and be routed to the nearest vacant spot.","Gate and buzzer automation removes the need for a human attendant at entry and exit.","On detecting gas above threshold, the system automatically turns on ventilation, improving safety in enclosed lots.","Peak-hour detection data could support dynamic pricing and staffing schedules, since the paper records up to 100 detections per hour at midday.","If networked across a city, the same MQTT architecture could aggregate many lots into one availability service."],"supporting_citations":[{"why":"Provides the baseline IoT system with IR sensors, cloud, and mobile app that the proposed work extends.","marker":"[13]"},{"why":"ESP32 light-sensor online parking system used as a comparison for real-time web and app updates.","marker":"[14]"},{"why":"ESP32 ultrasonic reservation system whose update delays and wrongful parking are cited as problems the proposed system avoids.","marker":"[15]"},{"why":"IR, ESP32, and servo motor car park monitoring system used as a reliability and scalability comparison.","marker":"[18]"},{"why":"Raspberry Pi camera and ultrasonic agent whose calibration and Wi-Fi dependence are cited as weaknesses.","marker":"[21]"},{"why":"MQTT Dashboard app documentation, the mobile client through which slot status is displayed to drivers.","marker":"[28]"},{"why":"Source for the IoT versus non-IoT performance metric table covering integration, data acquisition, safety, automation, and response.","marker":"[29]"}],"fun_headline_variants":["IR sensors and MQTT turn parking lots into live maps","Smart parking: sensors + app = real-time slot tracking","IoT parking system streams live slots to your phone","Parking sensors feed app with live slot and air data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The IR sensors used at gates and slots must reliably detect cars despite dust, changing sunlight, and varying angles, because the paper's own results show accuracy falling as ambient light increases.","fun_headline_variants_meta":{"raw":{"variants":["IR sensors and MQTT turn parking lots into live maps","Smart parking: sensors + app = real-time slot tracking","IoT parking system streams live slots to your phone","Parking sensors feed app with live slot and air data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1422,"prompt_tokens":817,"completion_tokens":605,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":433,"completion_tokens_details":{"reasoning_tokens":539}},"tokens_in":433,"tokens_out":605,"duration_ms":5336,"temperature":1.0,"reasoning_tokens":539,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:36:08.348103+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the system for one week in an open parking lot with direct sunlight and dusty conditions, and compare each slot's IR-reported state against a video ground truth of whether a car is present; if false occupancy errors occur often enough to mislead users, the central accuracy claim fails.","supporting_citations":[{"cited_title":"Iot based vehicle parking model with mobile apps software,","cited_arxiv_id":null,"evidence_quote":"Provides the baseline IoT system with IR sensors, cloud, and mobile app that the proposed work extends."},{"cited_title":"Online parking system using esp32,","cited_arxiv_id":null,"evidence_quote":"ESP32 light-sensor online parking system used as a comparison for real-time web and app updates."},{"cited_title":"Iot-based smart parking man- agement system using esp32 microcontroller,","cited_arxiv_id":null,"evidence_quote":"ESP32 ultrasonic reservation system whose update delays and wrongful parking are cited as problems the proposed system avoids."},{"cited_title":"Car park monitoring system (cpmos),","cited_arxiv_id":null,"evidence_quote":"IR, ESP32, and servo motor car park monitoring system used as a reliability and scalability comparison."},{"cited_title":"Shaikh, H","cited_arxiv_id":null,"evidence_quote":"Raspberry Pi camera and ultrasonic agent whose calibration and Wi-Fi dependence are cited as weaknesses."},{"cited_title":"MQTT Dashboard Documentation,","cited_arxiv_id":null,"evidence_quote":"MQTT Dashboard app documentation, the mobile client through which slot status is displayed to drivers."}],"review_version":1}