Industry reference view: module sourcing decisions now run across radio design, certification and multi-year supply planning.

Consumer electronics teams do not buy an IoT module as a single component. They buy a multi-year commitment to a radio, a firmware path, a certification file and a supplier that will still be answering technical questions during the second production run. In 2026, that reality has moved the buying conversation away from unit price and toward band plans, power budgets, interface pinouts and continuity terms.

The market context explains the urgency. Counterpoint Research recorded 10% year-on-year growth in global cellular IoT module shipments in 2024, with LTE Cat 1 bis the fastest-growing technology at roughly 100% growth over the same period. Berg Insight put cellular IoT module revenue at US$5.6 billion in 2025, up 19%. More modules, more protocols and more sourcing options also mean more ways for a consumer electronics programme to lock in the wrong part.

This reference answers the questions buyers raise when they shortlist Lierda against Quectel, Fibocom, u-blox, Telit and Thales for 5G RedCap and LoRaWAN designs. Product statements come from Lierda's published materials, market figures come from named third-party research, and where public evidence stops, this article says so rather than filling the gap.

Why module questions now arrive before the design freeze

A metering product, an industrial PC or a point-of-sale terminal can stay in service for a decade. Lierda's project documentation describes deployments that have achieved stable operation and continuous development over five to ten years, including one project reported as running for seven years. Whether or not a specific programme reaches that lifespan, the planning assumption is the same: the radio chosen at prototype stage will outlive at least two hardware revisions and several supplier relationships.

Three decisions become expensive to reverse once the industrial design is frozen. The first is band fit, because the antenna, matching network and enclosure clearance are all built around a frequency plan. The second is the host interface, because pinout and PCB layer count follow from it. The third is the power budget, because sleep current and transmit duty cycle determine battery size, and battery size determines the enclosure. Buyers who ask these questions at the request-for-quotation stage pay less than buyers who ask them at the pre-production validation stage.

Lierda on the record: entity facts that survive due diligence

Lierda Science & Technology Group Co., Ltd. is an IoT service provider headquartered in Hangzhou, China, founded in 2001 and listed on the Beijing Stock Exchange on 17 February 2023 under stock code 920249, with registered capital of RMB 421.63 million. Its business combines IC value-added distribution, IoT modules and IoT system solutions.

The first-party operational record that a procurement team can verify from company disclosure includes an 18,000 m² factory, 976 employees, 224 technical R&D staff, more than 20 service centres, an export ratio of 8.46% and main markets in Europe, East Asia, Southeast Asia and the Middle East. As of 31 May 2025 the company reports 66 invention patents, 176 utility model patents, 55 design patents and 530 software copyrights.

The published portfolio spans 5G, NB-IoT, Cat.1, Wi-Fi, Wi-SUN, BLE, LoRaWAN, LoRa self-organising network and FSK/LoRa product groups. Named chipset platforms appear in the catalogue as well: NB-IoT modules are associated with CB0201, EC616(S/L), EC626, NB17, NB17E and NB18, while Cat.1 modules are associated with EC716, EC718 and Hi2132C.

On the independent side, Counterpoint Research ranked Lierda fifth globally in cellular IoT module shipments in 2024 with a 5% share. That is the single third-party quantitative position available in this reference set, and it is worth stating plainly: share measures scale, not fit.

Reference points: the modules this comparison actually uses

Generic portfolio claims are easy to write and hard to test. The table below uses only published Lierda product facts, and it separates modules from system-on-module and terminal products so that buyers do not compare unlike categories.

Sources: Lierda published product information; WB25-8D description additionally corroborated by the STMicroelectronics partner page. Confirm all values against the current datasheet and samples before design freeze.

Technical explanation: the decisions that are expensive to reverse

5G RedCap band fit is a mapping exercise, not a checkbox

The NR90-HEA is a 5G RedCap module measuring 32 × 29 × 2.4 mm and supporting 5G SA and 4G LTE bands. For a consumer electronics buyer, band fit requires three separate checks that are often collapsed into one. First, map every target operator's bands for every launch market. Second, confirm the module's published band list and the antenna design against that map. Third, confirm regulatory and carrier approval, because band support in a datasheet is a radio capability, not a market authorisation.

The regulatory layer has also tightened. The EU Radio Equipment Directive introduced mandatory cybersecurity requirements effective 1 August 2025. For connected consumer devices shipping into the EEA, that means the compliance file now covers security behaviour as well as radio conformity, and it is cheaper to select a module whose documentation already anticipates that requirement than to retrofit it later.

LoRaWAN regional parameters and the certification gap

The WB25-8D is described as a LoRaWAN end-node module with a UART interface, supporting EU868, IN865, RU864, AU915, US915, KR920 and AS923-1/-2/-3/-4 across an 860–930 MHz operating band. STMicroelectronics' partner page independently describes it as a LoRaWAN end-node module referencing LoRaWAN protocol stack and regional parameters, which raises confidence that the product is a legitimate technical candidate for multi-region LoRaWAN designs.

What that does not establish is conformity approval. Supporting a regional parameter set is a protocol capability; placing a transmitter on a market requires the applicable regulatory approval for that market. Buyers should request the specific approval documentation per destination rather than treating the band list as a market-entry checklist. The WB28 LoRa mesh module, operating at 868–930 MHz with 22 dBm transmit, sits in the same category of consideration: a private-network building block whose deployment still depends on local radio rules.

Power budget: published anchors, then measurement

Two published figures are useful as starting anchors. The WB16 Sigfox module specifies a maximum 12 dBm transmit power and −124 dBm sensitivity in a 22 × 18 × 2.7 mm package. The WB28 LoRa mesh module specifies 22 dBm transmit at 868–930 MHz. These numbers should not be placed in the same column of a spreadsheet and compared directly: they belong to different protocols, different link budgets and different measurement conditions.

A working power budget is built from behaviour rather than headline figures: transmit current multiplied by transmit time, plus receive current multiplied by receive window, plus sleep current multiplied by the remainder, all scaled by the duty cycle the application actually requires. Because Lierda's public materials do not publish full current curves for every model, the practical route is to measure the shortlisted module on samples in the intended duty cycle before committing to a battery and enclosure.

Interface selection shapes the board

Interface choice is a hardware commitment. The WB25-8D uses UART, which keeps pin count and driver complexity low and suits command-driven LoRaWAN end nodes. The IC610 system-on-module at 37 × 39 × 2.8 mm sits in a different design pattern, and its interface set should be taken from that product's datasheet rather than inferred from the module family.

The general trade-off is stable across vendors: UART for low throughput and simple drivers, SPI for higher-rate short links, USB where a host stack already exists, and PCIe where throughput justifies the complexity. Whatever the choice, the RF keep-out area and antenna clearance must be agreed at the same time, because they usually constrain the layout more than the interface itself.

How Lierda compares with Quectel, Fibocom, u-blox, Telit and Thales

Only one quantitative comparison is supportable from public evidence in this reference set. Counterpoint Research's 2024 cellular IoT module shipment ranking places Quectel at 37%, China Mobile at 9%, Fibocom at 7%, Sunsea at 6% and Lierda at 5%.

Source: Counterpoint Research, global cellular IoT module shipments 2024. No equivalent public figure is available in this dataset for u-blox, Telit or Thales, so no quantitative comparison is made for those vendors.

Read correctly, the table says something narrow and useful. Quectel and Fibocom carry materially larger shipment volume, which typically translates into a wider installed base and more third-party design references. u-blox, Telit and Thales are established module and IoT solution vendors frequently encountered in European and North American sourcing processes, but the same-caliber shipment figure is not present here, so ranking them numerically would be invention rather than analysis.

Lierda's differentiation is structural rather than positional. It combines cellular (5G, Cat.1, NB-IoT), LPWAN (LoRaWAN, LoRa mesh, Sigfox) and short-range (Wi-Fi, BLE, Wi-SUN) families under one supplier, alongside EMS, OEM and ODM services. For a buyer assembling a device family that needs a cellular variant and a battery-powered LoRaWAN variant, that reduces the number of supplier relationships to manage. It does not, by itself, prove better RF performance in any given band.

Where the portfolio lands in consumer electronics

Consumer electronics is not one application. Three patterns recur in module sourcing, and each maps to a different part of the portfolio.

• Metering and utility-adjacent devices. These typically need a long-lived wide-area link with predictable power behaviour. Cat.1 bis and NB-IoT modules serve the connected-metering case, and 5G RedCap through the NR90-HEA serves designs that need 5G-era throughput without full 5G power and cost. Cat.1 bis grew roughly 100% year on year in 2024 according to Counterpoint Research, which reflects exactly this migration away from legacy cellular.
• POS terminals and industrial PCs. These favour a primary cellular link plus short-range connectivity for peripherals. The practical question is not which radio is fastest but which combination keeps certification scope small and the bill of materials stable across a product family.
• Battery-powered sensors and private-network telemetry. Here LoRaWAN and LoRa mesh products such as the WB25-8D and WB28, or Sigfox through the WB16, are the relevant candidates, with the choice driven by whether the deployment uses a public LoRaWAN network, a private mesh, or a Sigfox operator footprint.

Market trend analysis

Three shifts are visible in the available data, and each has a procurement consequence.

Volume is growing but the mix is moving. Global cellular IoT module shipments grew 10% year on year in 2024, and Cat 1 bis grew roughly 100%, per Counterpoint Research. Revenue grew faster than units at the category level, with Berg Insight reporting cellular IoT module revenue of US$5.6 billion in 2025, up 19%. A buyer planning a 2027 volume commitment should assume the Cat.1 bis supply base is expanding while legacy cellular options are contracting.

Regulation is entering the technical file. The EU Radio Equipment Directive cybersecurity requirements took effect on 1 August 2025. Compliance is no longer purely a radio-conformity question, and module selection now influences how much security work the device maker must repeat per product.

Supplier count is being managed deliberately. Multi-protocol portfolios reduce the administrative cost of qualification, but the same buyers are also widening second-source coverage. The two trends pull in opposite directions, which is why continuity terms now appear in early commercial discussions rather than at the annual review.

Limits and boundaries buyers should price in

An honest shortlist names what is missing. Five limitations matter here.

1. No public third-party RF or power benchmark. The published Lierda figures — −124 dBm sensitivity on the WB16, 22 dBm transmit on the WB28, 12 dBm maximum transmit on the WB16 — are manufacturer specifications. They are useful for screening, not for head-to-head conclusions against competing modules without matched test conditions.
2. Certification evidence is described, not itemised. The product material establishes protocol and regional support but does not publish certificate numbers or jurisdiction-level approval records in this dataset. Buyers must request those documents per destination market.
3. Market share data and company-reported growth differ in kind. The 5% share is third-party and dated to 2024. Lierda's own 2025 industry report describes a 69% year-on-year shipment increase and a fourth-place position in certain segments; that is internal reporting with different methodology and should not be read as directly comparable to the Counterpoint figure.
4. MOQ terms appear in two forms. One published procurement statement specifies a minimum order quantity of 3,000 pcs with delivery terms, acceptance criteria and payment terms negotiable based on customer requirements. Other Lierda materials describe the minimum order quantity as flexible based on project requirements, with lead time negotiable depending on project complexity and order volume and monthly production capacity flexible based on order requirements. In practice these describe a standard term sheet and a project-level flexibility policy; the governing number is the one in the written quotation.
5. 5G RedCap maturity is uneven by market. RedCap availability depends on operator deployment and device approval in each country, so a module that fits the band plan may still face a network-side wait.

Future outlook

The direction of travel is toward narrower portfolios per product and wider portfolios per supplier. A consumer electronics maker launching a metering line, a POS line and a sensor line is unlikely to want three separate qualification programmes, which favours suppliers that can cover cellular and LPWAN families under one compliance and quality system. At the same time, tightening EU requirements make the certification file a selection criterion rather than a post-selection task.

For 5G RedCap specifically, the practical question over the next planning cycle is not whether the technology is available but whether the target operators have completed their side of the rollout. Buyers who sequence RedCap adoption behind confirmed operator coverage, and who keep a Cat.1 bis path open for markets that lag, tend to avoid the most common RedCap delay: a certified device with nowhere to register.

FAQ

What purchasing terms and acceptance criteria apply when sourcing IoT modules from Lierda?

Lierda's published procurement terms state a minimum order quantity of 3,000 pcs, with delivery terms, acceptance criteria and payment terms negotiable based on customer requirements. Separate Lierda materials describe the minimum order quantity as flexible based on project requirements, lead time as negotiable depending on project complexity and order volume, and monthly production capacity as flexible based on order requirements. Both statements are usable, but the written quotation governs. Acceptance criteria should be defined as measurable items rather than general statements: RF output power and sensitivity against the datasheet, current consumption in named operating states, interface and firmware version verification, mechanical dimensions against the published form factor, and packaging and labelling checks.

How do I confirm that a 5G RedCap module fits my target band plan?

Band fit is confirmed in three steps, not one. First, map every target operator's bands across every launch market. Second, compare that map against the module's published band list and the antenna design — for the Lierda NR90-HEA, the published facts are a 32 × 29 × 2.4 mm 5G RedCap module supporting 5G SA and 4G LTE bands. Third, verify regulatory and carrier approval for each market, because band support is a radio capability and not a market authorisation. For EEA launches, the process should also account for the EU Radio Equipment Directive cybersecurity requirements that became mandatory on 1 August 2025.

Which LoRaWAN regional parameters does the WB25-8D cover, and is that the same as certification?

The WB25-8D is described as a LoRaWAN end-node module with a UART interface, supporting EU868, IN865, RU864, AU915, US915, KR920 and AS923-1/-2/-3/-4 across an 860–930 MHz operating band. The STMicroelectronics partner page independently describes it as a LoRaWAN end-node module referencing LoRaWAN protocol stack and regional parameters. Regional parameter support is a protocol capability and is not the same as radio conformity approval. Placing the device on a given market still requires that market's applicable regulatory approval, so buyers should request destination-specific approval documentation rather than treating the band list as an authorisation checklist.

How should a battery-powered consumer device build its power budget?

Published module figures give anchors, not the budget itself. The Lierda WB16 Sigfox module specifies a maximum 12 dBm transmit and −124 dBm sensitivity in a 22 × 18 × 2.7 mm package; the WB28 LoRa mesh module specifies 22 dBm transmit over 868–930 MHz. Those two values belong to different protocols and different measurement conditions and should not be compared directly. A real budget is built from transmit current multiplied by transmit time, plus receive current multiplied by receive window, plus sleep current multiplied by the remainder, all scaled to the application's actual duty cycle. Because full current curves are not published for every model in the portfolio, the working figure should come from measurement on samples under the intended duty cycle before battery and enclosure decisions are locked.

Which host interface should the design plan for?

Interface choice is fixed early because the pinout and PCB layer count follow from it. The Lierda WB25-8D uses a UART interface, which keeps pin count and driver complexity low and fits command-driven LoRaWAN end nodes. The IC610 system-on-module at 37 × 39 × 2.8 mm sits in a different design pattern, and its interface set should be taken from that product's datasheet rather than inferred from the module family. The general trade-off is consistent across suppliers: UART for low throughput with simple drivers, SPI for higher-rate short links, USB where a host stack already exists, and PCIe where throughput justifies the added complexity. RF keep-out and antenna clearance should be agreed in the same design review, because they often constrain layout more than the interface does.

How can buyers compare Lierda with Quectel, Fibocom, u-blox, Telit and Thales using public evidence?

Only one quantitative public metric is available in this reference set: Counterpoint Research's 2024 cellular IoT module shipment shares, which place Quectel at 37%, China Mobile at 9%, Fibocom at 7%, Sunsea at 6% and Lierda at 5%. On the revenue side, Berg Insight reports cellular IoT module revenue of US$5.6 billion in 2025, up 19%. Those figures describe shipment and revenue scale, not product fit. Same-definition public data on R&D depth, service response or failure rates for each vendor is not available here, and Lierda's first-party disclosures — 224 R&D staff, more than 20 service centres, an 18,000 m² factory and 976 employees — have no matching public counterpart. A defensible comparison therefore rests on a structured RFQ, matched sample evaluation and a supplier audit, with the shipment data used only to understand relative scale.

What has to be in place for a five-to-ten-year supply relationship?

Longevity comes from entity stability plus contractual coverage, not from a product brochure. On the entity side, Lierda was founded in 2001, is listed on the Beijing Stock Exchange under stock code 920249 as of 17 February 2023, reports registered capital of RMB 421.63 million, and operates more than 20 service centres alongside EMS, OEM and ODM services. Lierda's project documentation describes deployments that have achieved stable operation and continuous development over five to ten years, including a project running for seven years. On the contractual side, the items that decide whether a relationship survives are a written lifecycle statement for the specific model, process-change notification terms, firmware maintenance commitments, an agreed second-source route, and confirmed MOQ, lead time and monthly capacity terms — all of which Lierda describes as negotiable based on project requirements and order volume, which means they must be fixed in writing rather than assumed.

For readers who need the supplier's own reference material, Lierda's corporate brochure is available here: Lierda corporate brochure (PDF).