Battery-Free Ambient IoT Tags: Backscatter Radios, Energy Budgets and Domestic Alternatives

A label that costs a few cents, has no battery, sits on a pallet for ten years and still answers when asked is one of the few genuinely disruptive ideas left in wireless. The physics is old — radar engineers were modulating reflections before transistors — but the economics only became interesting when the per-tag silicon cost fell far enough to be printed onto packaging. That is the space now called ambient IoT: tags that harvest their operating energy from the incident carrier, or from a small photovoltaic or thermal source, and reply by changing how much of that carrier they reflect. This article works through the tag side of that link: the rectifier, the detector, the modulator, the microcontroller that wakes up on almost nothing, and the domestic suppliers who can actually build it. For reader-side design, see our UHF reader and energy harvesting article.

Application logic: the tag spends energy it has just borrowed

• The incident carrier is both power and question — the reader radiates a continuous wave; the tag rectifies a fraction of it into a supply rail, and reads the on-off keying or the amplitude modulation that the reader superimposes on the same wave.
• Rectifier first, everything else second — a multi-stage diode charge pump converts a few hundred millivolts of received RF into a usable rail. Its conversion efficiency at low incident power decides the read range more than any digital feature.
• Sensitivity floor — the tag wakes at roughly minus 20 dBm at the antenna in the best current designs. That number, not the reader’s transmit power, is what people mean when they quote a read distance.
• Modulation by reflection — the tag changes its antenna termination between two states; the difference in reflected power is the uplink. Backscatter amplitude or phase modulation, with a subcarrier, is how the reader separates the weak reply from its own strong carrier.
• Collision handling — many tags, one carrier. Slotted random access with a handle assigned by the reader keeps the inventory fast; a tag that replies immediately on power-up paralyses a dense population.
• Non-volatile state — a sensor tag must keep its last reading and its identity with the power off. That is either a small EEPROM embedded in the tag chip or a thin-film element, and it is the part that has to survive ten years of thermal cycling.
• The uplink budget is asymmetric — the downlink is limited by the tag’s rectifier, the uplink by the reader’s sensitivity and by the tag’s modulated reflection coefficient. Improving the reader helps the uplink; improving the rectifier helps both.

The parameters that decide whether the label survives the pallet

• Rectifier efficiency at low input power — the difference between 20 percent and 35 percent efficiency at minus 15 dBm is roughly double the read range, or the same range from half the reader power.
• Wake-up and retention current — a tag in a warehouse is interrogated a few times a day and must hold its state for years. Leakage in the retention path, not active current, sets the shelf life.
• Detector threshold stability — the demodulator compares a tiny envelope against a reference; if that reference drifts with temperature, tags fail intermittently at the edge of coverage, which is the hardest failure to diagnose in the field.
• Modulator depth and speed — a deeper reflection change gives a stronger uplink but disturbs the rectifier; too fast and the reader’s filter cannot separate the subcarrier.
• Antenna impedance under a dielectric load — a tag designed on air detunes when stuck to a carton of liquid. The matching network has to be designed for the material it will sit on.
• Mechanical and environmental — flexing, moisture, and the metal shelf it is applied to. A tag on metal needs a spacer or a specifically designed antenna, and that changes the whole cost structure.
• Security and privacy — a tag that answers anyone reveals inventory to anyone. Kill commands, authentication and short identifiers are now part of the specification rather than an afterthought.

What is changing: sensor tags, printed electronics and standardised ambient IoT

Three shifts matter. First, tags became sensors: temperature, humidity, strain and even a crude accelerometer can sit on the same die or the same printed substrate, so the tag reports a condition rather than an identity. That is what turns a label into a monitoring product, and it changes the energy budget from a fixed cost into a duty-cycled one. Second, printed and hybrid assembly pushed the cost of the antenna and the interconnect down far enough that the silicon is now the dominant term again, which is why the tag chip itself is the part everyone is trying to make cheaper. Third, standardisation: the third-generation partnership work on ambient IoT has been defining device classes and backscatter interfaces, which lets a tag from one vendor answer a reader from another — the single biggest obstacle to the market for the last decade.

A parallel shift is the alternative power source. A photovoltaic cell the size of a fingernail, or a thermoelectric element on a warm pipe, can lift a tag from passive to semi-passive, which buys tens of decibels of link margin and makes battery-free sensing realistic in places where the incident carrier is weak. The design question then becomes which source dominates in the actual installation, and that is a site survey, not a datasheet decision.

Domestic alternatives: suppliers and indicative cost

This is one of the few RF categories where the domestic supply chain is genuinely strong, because China is the largest producer of UHF inlay and label products. Tag chip suppliers with published UHF product lines include Shanghai Quanray, whose tag ICs are used in large label volumes; inlay and label manufacturing is deep enough that a project can be sourced end to end domestically. What is thinner is the sensor-tag segment, where a domestic chip with an integrated temperature sensor and a competitive rectifier is not yet a like-for-like substitute for the imported part, and the honest approach is a discrete build or a module.

For a discrete prototype or a small production run: a HP5082-2811 zero-bias Schottky diode as the detector or the rectifier element for characterisation work, a HSMP-3890 PIN diode as the reflection modulator, a COS3502SR comparator as the envelope slicer, an AT32L021C8T7 as a low-power controller on semi-passive designs, and a COS8551TRC zero-drift amplifier for the sensor front end. Our Schottky detector article covers the diode selection in detail.

Price figures below are indicative public-market reference ranges for moderate volumes, compiled from publicly observable listings. They fluctuate with the spot market, package option, grade (industrial, automotive, screened) and order quantity, and must not be read as a quotation. Ask us for a firm price on your exact part numbers and quantities.

Where the cost difference actually comes from

In this category the bill of materials is almost the whole product, which is unusual and makes the arithmetic blunt: a tag chip at six cents instead of nine is a thirty percent reduction in the delivered label price, and at label volumes of millions that is the entire business case. That is why the identity-tag segment was commoditised years ago and why the domestic supply chain, which has the inlay capacity, is competitive on price and on lead time rather than on novelty.

The sensor-tag segment is the opposite situation. A temperature-sensing tag sells for tens of cents because it answers a question worth more than the label — cold-chain integrity, for example. Here the imported part's integrated sensor, calibrated output and retention behaviour are worth their premium, and a domestic entry should be based on a discrete design where you control the sensor and the calibration, not on waiting for a part that is not yet in the catalogue.

Validation flow before you commit

Measure the wake-up power of the tag, not the read distance on a bench: use a calibrated attenuator between reader and tag and find the power at which the tag answers reliably ten times out of ten, then repeat that at minus 20 and 60 degrees Celsius. Place the tag on the actual packaging — a carton, a plastic tote, a metal shelf — and redo the measurement, because detuning will cost more than any silicon change. Run an inventory of two hundred tags in a real doorway and record the read rate and the time to complete; that number, not the single-tag range, is what a logistics customer will judge. Finally, leave a batch powered-off in a warm humid chamber for a month and confirm that the stored data and the wake-up threshold have not moved.

Frequently asked questions

What range can a battery-free tag realistically reach?

A few metres with a compliant reader in a benign environment, and far less on metal or behind liquid. The figure is set by the tag wake-up power, the antenna and the material, not by the reader's transmit power alone.

Is a backscatter tag secure?

Not by default. Identity tags answer anyone with a compatible reader, so privacy requires short identifiers, authentication or a kill command, and the choice is a product decision rather than a radio one.

Do ambient IoT standards make tags interchangeable?

The work in progress defines device classes and interfaces precisely so that they can be, which is the change that would finally let a buyer mix vendors in one deployment.

When is a semi-passive tag worth the battery or the solar cell?

When the incident carrier is weak, when the tag must sense frequently, or when the read point is far from the reader; the added source buys tens of decibels of margin at the cost of a more complex product.

About us & trademarks: JTDZ Tech (Shenzhen Jingteng Electronics Technology Co., Ltd.) is an independent distributor of electronic components. We are not an authorised distributor or representative of the imported manufacturers named in this article. Manufacturer, brand and part names are used for identification purposes only; all trademarks remain the property of their respective owners. Cross-reference information in this article is based on publicly available sources, is subject to change, and is not a substitute for your own engineering validation. Always verify specifications against the manufacturer's datasheet before design-in or purchase.