Vadim Medyanik spoke to RUBEZH about the use of RFID and video analytics in cultural institutions

Vadim Medyanik: "RFID identifies an object, while video analytics helps understand who is interacting with it and what they are doing."

Radio-frequency identification technologies have long ceased to be a novelty for museums, libraries, and archives. However, there are still many myths and oversimplifications surrounding RFID: some consider it a universal solution for protecting collections, while others view it as unreliable due to the presence of metal, liquids, and historical interiors. Vadim Medyanik, co-founder and CTO of the AI systems developer BPA Tech, spoke to RUBEZH magazine about how to select the right frequency range, ensure discreet labeling, integrate tags with access control and video surveillance, and how much RFID can accelerate inventory processes.

RUBEZH: Which types of RFID tags and frequency ranges do you consider optimal for museum exhibits, library collections, and archival documents, and how does the choice of technology affect read range and interference resistance in historical interiors?

Vadim Medyanik: When choosing RFID technology, it is important to base your decision not only on the required read range but also on the specific use case, the type of collection, the materials of the objects, and the characteristics of the premises.

For libraries and archives, the HF (high-frequency) range is primarily used. For such tasks, a relatively short read range is actually an advantage: it allows for more precise control over exactly which document or object is in the reader's zone and reduces the likelihood of picking up neighboring tags. There is a Russian regulatory framework for this class of solutions, including GOST standards for the use of radio-frequency identification in libraries. These define not only technical parameters but also the structure and rules for encoding data on RFID tags. This is important for equipment compatibility and prevents the system from being tied entirely to one specific manufacturer of readers or tags.

For museum collections, especially if the task involves mass inventory, UHF (Ultra High Frequency) RFID can be used. This technology allows for operation at a greater distance and the simultaneous reading of a significant number of tags, making it well-suited for managing large collections, storage areas, and warehouses. At the same time, it is necessary to consider Russian requirements for the use of radio-frequency equipment: permitted ranges, power levels, and conditions of use. In practice, this means that when purchasing equipment, you must verify the reader's specific Russian radio-frequency profile.

A separate issue is the operation of RFID in historical buildings. Here, the actual read range can differ significantly from the manufacturer's specifications. Signal propagation is affected by metal cabinets, display cases, massive walls, complex room geometry, signal reflections, and the formation of so-called dead zones. Therefore, a mandatory stage of implementation must be a preliminary radio site survey.

It is necessary to check signal propagation on-site, test several types of tags, evaluate different antenna placement options, and only then determine the final system configuration. It is these field tests that allow you to understand which frequency range and equipment will work reliably in a specific museum, library, or archive.

RUBEZH: How can an RFID system be integrated with security alarms and access control systems to automatically block an exit during an unauthorized removal attempt, and what data exchange protocols are used for this?

Vadim Medyanik: From an architectural standpoint, such integration is built quite consistently. An RFID reader or multiple readers are connected to a general information system, into which the access control system is also integrated. The access control system identifies which employee or visitor is passing through a specific checkpoint. At the same time, the RFID portal records which tagged items are moving through that zone. As a result, the information system can correlate several parameters simultaneously: who is passing through the checkpoint, what item they are moving, where the movement is originating from and going to, and whether it is permitted according to established rules.

The typical architecture looks like this: the RFID reader transmits data to an intermediate software layer, then the information is cross-referenced with the management system, and only then is an event generated for the access control or security system. This intermediate layer allows for the decoupling of hardware from business logic, avoiding a rigid dependency of the entire system on a specific reader model or a specific access control manufacturer. At the level of interaction between the RFID reader and the server, specialized protocols such as LLRP can be used, as well as software interfaces and manufacturer SDKs.

If the system detects an attempt to move an item without proper authorization, an alarm event is triggered. From there, it is possible to automatically display the relevant camera feed at the security station, log the incident, send a notification to the responsible staff member, and initiate the protocol-defined scenario in the access control system.

However, I would not view RFID as a mechanism that should unconditionally lock every door. Separate safety requirements apply to emergency exits and a number of other zones. Therefore, it is more accurate to speak of a pre-designed response logic: an alarm event, video verification, security notification, and control of actuators only in cases where it is provided for by the project and does not contradict safety requirements.

RUBEZHWhat challenges have you encountered when tagging metal objects that contain liquids or have irregular surfaces, and what technical solutions help prevent signal shielding and false positives?

Vadim MedyanikWorking with metal, liquids, and complex-shaped surfaces is indeed one of the most sensitive aspects of RFID projects, especially when it comes to the UHF range. When a standard RFID tag is placed directly on a metal surface, its performance can change significantly. Metal affects the electromagnetic parameters of the antenna. As a result, a standard RFID sticker attached directly to a metal object may suffer a sharp drop in range or stop being read reliably altogether.

For such cases, specialized on-metal or anti-metal tags are used. Their design includes a calculated dielectric layer between the antenna and the metal surface, which keeps the tag's characteristics predictable and stable. Liquids present a different problem. In the UHF range, water absorbs electromagnetic energy quite well, so items with high liquid content can significantly hinder signal propagation. A similar effect occurs to some extent when a signal passes through the human body. As a result, the tag receives less energy from the reader and operates at a shorter range.

For items with complex or curved surfaces, it is also necessary to consider the geometry of the antenna itself. If a tag is bent or deformed significantly, its characteristics can also change, sometimes quite substantially. Therefore, it is not enough to simply choose a tag from a catalog. You must consider its design, installation location, orientation relative to the reader, distance to metal elements, and antenna configuration. In some cases, it makes sense to change the tag's placement or provide a small technical gap.

It is essential to conduct field testing of several options directly on the items intended for tagging. In practice, the position and orientation of the tag sometimes affect reading stability more than switching to a different RFID chip. Therefore, in such projects, laboratory specifications must always be confirmed by tests under real operating conditions.

RUBEZHHow can you ensure the discreet placement of tags on exhibits without compromising their aesthetics and preservation, while simultaneously protecting the tags from malicious actors who might attempt to shield or damage them?

Vadim MedyanikOf course, in a museum environment, the preservation of the object itself must always come first. An RFID tag should not alter the appearance of an exhibit, damage its surface, or create additional risks during storage. Therefore, the mounting method and installation location are determined individually for each category of object. A tag can be placed on the back of a frame, inside a structural element, or in another hidden spot where it remains functional while being visually unobtrusive and not interfering with the item's use.

In many cases, the optimal solution is to install the tag not directly on the historical surface, but on an element that is clearly associated with the exhibit: a frame, mount, stand, case, transport container, or archival packaging. For particularly valuable objects, such solutions must, of course, be coordinated with a curator or conservator.

One of the advantages of RFID is that it does not require a direct line of sight for reading. This allows tags to be placed quite deeply and discreetly. However, radio physics must also be considered here. For example, if a UHF tag is hidden directly behind a metal element, it may be hidden not only from people but also from the RFID reader.

As for protection against intentional tampering, relying solely on an RFID tag would be a mistake. It can be damaged, removed, or shielded. Therefore, RFID should work as part of a comprehensive security system. For certain scenarios, there are tags with tamper-evident or destruction-detection features that allow for the recording of physical interference attempts. However, a more reliable approach is to integrate RFID with access control systems (ACS), video surveillance, and, for particularly valuable items, additional sensors for removal, vibration, or movement.

For example, AI-based video analytics and computer vision can detect an attempt to pick up or move an object, a person entering a restricted area, prolonged loitering near a valuable exhibit, or other suspicious activities. In this setup, RFID identifies exactly which object is being moved, while video analytics helps understand who is doing it and what they are doing. This way, an attempt to disable one control element does not lead to a loss of monitoring for the object as a whole.

Furthermore, the RFID tag itself should contain only the necessary amount of information. As a rule, this is just the object identifier, while the primary information about the item, its status, authorized movements, and operation history is stored in a secure information system. This approach simultaneously simplifies data management and increases the overall security level of the system.

RUBEZHHow much does RFID speed up scheduled inventory compared to manual tracking, and what economic impact—such as labor savings or loss prevention—have you observed in real-world projects in museums or archives?

Vadim MedyanikThe main effect of RFID on inventory is that the technology eliminates the need for manual visual searches for inventory numbers and individual scanning of each object. If we are talking about UHF RFID, a single reader can simultaneously receive responses from a large number of tags and process hundreds of reads per second. Depending on the specific equipment, tags, and operating conditions, the working distance can be several meters.

This fundamentally changes the inventory process itself. A staff member can walk along the shelves with a handheld reader, and the system automatically compares the detected tags with the inventory database. In doing so, you can not only verify the presence of an item but also identify objects that are in the wrong storage area.

Passive RFID tags are the most interesting option for mass inventory. They do not require their own power source, are small in size, and are well-suited for labeling large quantities of storage items. Active tags are also used, but typically in more specific scenarios, such as for the constant tracking of high-value or mobile assets.

Beyond inventory itself, RFID simplifies access to information about an object. After scanning a tag, staff can immediately retrieve the item's record from the information system, including its status, current storage location, movement history, and other data. RFID can also be used to monitor an object's presence in a specific zone and for anti-theft scenarios.

That said, I wouldn't cite a universal acceleration figure, such as ten or twenty times faster. Such a metric depends heavily on the density of the collection, the materials of the items, storage organization, labeling quality, and the degree of RFID integration with the accounting system. It is much more accurate to define measurable KPIs during the pilot phase: the number of items checked per hour, the number of man-hours per inventory, the percentage of tags successfully read on the first pass, the number of objects requiring manual verification, and the average time to locate an item missing from its designated spot.

The economic impact is not just about direct labor savings. The system helps reduce the duration of inventory, decrease the number of manual operations and associated errors, locate misplaced items faster, and more quickly detect the absence or unauthorized movement of an object.

The maximum impact is achieved when RFID does not exist separately from the accounting system but becomes part of a unified digital process. If only a small portion of the collection is labeled, while a significant volume of information remains in separate spreadsheets or paper documents, the technology's potential will only be partially realized.

RUBEZH: How can data on RFID tags and in the information system be protected against cloning, unauthorized reading, or cyberattacks, especially if the institution uses cloud services for information processing and storage?

Vadim Medyanik: Here, it is important to distinguish between protecting the RFID tag itself and the security of the entire information system. It is important to note that a standard tag identifier cannot be considered full protection against cloning on its own. For objects with higher security requirements, more secure RFID tags can be used that support authentication mechanisms, restricted access to specific memory areas, and additional security features.

However, the system architecture still plays the primary role. It is advisable to place RFID readers in a separate network segment and not provide them with direct internet access. Communication with the server side should take place over secure channels, administrative access should be role-based, and critical accounts should be further protected by multi-factor authentication. User activity logging, change control, backups, and monitoring for unauthorized access attempts are also necessary. In other words, you need to protect not just the RFID radio channel, but also the servers, databases, APIs, accounts, and integrations with external systems.

When using cloud infrastructure, it is additionally important to understand where the data is physically located, who has access to it, and how key management, backups, and disaster recovery are organized. If the system processes personal data of employees or visitors, relevant local requirements for their storage and processing must also be taken into account.

RUBEZH: When implementing RFID systems in cultural institutions, what organizational and technical barriers arise most often—from staff resistance to incompatibility with existing IT infrastructure—and how can they be overcome without stopping the museum or library's operations?

Vadim Medyanik: In practice, the main difficulties during RFID implementation are most often related not to the technology itself, but to the institution's existing processes and information infrastructure. A museum, library, or archive may simultaneously use several databases, proprietary or custom-built accounting systems, and different formats for inventory numbers. Some information may be in a specialized program, some in Excel, and certain details may still exist only on paper.

Therefore, the first serious stage of such a project involves an audit of existing records. It is necessary to understand where the current information is located, eliminate duplication, standardize identifiers, and determine which system will be the primary source of data. It is by no means always necessary to completely replace the existing museum or library information system. In many cases, it is more appropriate to add an intermediate software layer that connects the RFID equipment with the active databases and systems. This approach allows for a reduction in the scope of changes and significantly lowers implementation risks.

The second major stage is the physical labeling of the collection. If we are talking about tens or hundreds of thousands of storage units, this is a significant organizational process in its own right. Each item must be labeled, the tag registered in the information system, linked to a specific record, and the accuracy of that link verified. This is precisely why implementation is best carried out in stages: by zone, collection, or individual holdings. During the transition phase, RFID can work in parallel with existing barcodes and inventory numbers. This allows for a transition to the new technology without halting the institution's operations.

Separately, it is necessary to survey the physical infrastructure of the facility: the layout of metal cabinets and display cases, room geometry, routes for people and items, existing control points, and technical integration capabilities with access control systems, video surveillance, and accounting software. I would definitely start with a pilot area that sufficiently represents the conditions of the entire facility. During the pilot, you can test read quality, select tags, determine antenna placement, refine integration, and simultaneously establish clear operating procedures for staff.

Overall, the most sustainable approach is not to try to replace the entire infrastructure at once, but to gradually integrate RFID into existing processes. For a time, the old and new systems can operate in parallel, with scaling conducted only after technical parameters and the organizational model have been confirmed.

RUBEZHWhat new opportunities does the combination of RFID with other technologies—such as BLE beacons, UWB positioning, and temperature and humidity sensors—open up for comprehensive security and real-time monitoring of cultural assets?

Vadim MedyanikRFID, BLE, and UWB solve different problems, so the maximum effect is achieved precisely when they are used together. To simplify their roles: RFID primarily answers the question, "What is this object, and is it in a specific zone?" Positioning technologies then allow us to answer the next question: "Where exactly is this object, and how is it moving?"

Passive RFID is well-suited for identifying large numbers of items, monitoring their presence in specific zones, and recording their passage through checkpoints. Such tags are compact, do not require their own power source, and can therefore be used in very large quantities.

BLE is used in cases where it is necessary to determine the presence and location of an object in a room in more detail. BLE beacons are active devices and can transmit information more stably and over longer distances. In more complex systems, signal direction-finding technologies can be used, which improves accuracy compared to simple distance estimation based on signal strength.

UWB is advisable for the most valuable or mobile objects when it is necessary to track their location and movement trajectory indoors with high precision. Such infrastructure is more expensive, so there is usually no need to equip the entire museum collection with UWB tags. It is much more logical to use it selectively for specific categories of cultural assets.

Another important layer is environmental sensors. For museums and archives, this primarily means temperature and relative humidity, but depending on the type of collection, lighting, vibrations, and other parameters can also be monitored. All this data can be integrated into a single information system. As a result, a digital passport for the exhibit is effectively created, which includes its identifier, current location, movement history, storage parameters, access control events, and associated video recordings.

For example, the system can determine that a specific item has left an authorized zone, correlate this event with an employee passing through an access control point, automatically pull up the corresponding camera, save a video recording of the event, and simultaneously check whether the temperature and humidity parameters meet the established requirements.

Therefore, I would view RFID not as a standalone security system, but as one of the sensor layers of a comprehensive digital infrastructure. RFID identifies the item, BLE or UWB allow for more precise location tracking, sensors monitor storage conditions, access control determines who has the right to move the object, and video surveillance and analytics allow for physical confirmation of what exactly occurred.

At the same time, a preliminary site survey still plays a key role. RFID technology itself is quite mature, and passive tags are compact, require no power, and can be used in very large quantities. However, the stability of the entire system depends on how correctly the frequency range, tag types, antenna placement, and infrastructure configuration are selected. And, of course, the completeness of digital records is fundamentally important. If only part of the collection is tagged, and the rest of the data is scattered across different databases, spreadsheets, and paper documents, the system's capabilities will be limited.