IoT technologies are mainly in the areas of communication and sensors. This article is about IoT technology NB-IoT vs LoRa technology.
What is the classification of IoT communication technology?
There are many wireless communication technologies for IoT, mainly divided into two categories: one is Zigbee, Wi-Fi, Bluetooth, Z-wave, and other short-range communication technologies; the other is LPWAN (low-power Wide-Area Networks), i.e. wide-area network communication technology.
LPWAN can be further divided into two categories: one is LoRa, SigFox, and other technologies working in the unlicensed spectrum; the other is 2G/3G/4G cellular communication technologies working in licensed spectrum and supported by 3GPP, such as EC-GSM, LTE Cat-m, NB-IoT, etc.
Standards and progress of NB-IoT in NB-IoT vs Lora
On RAN
In May 2014, Huawei acquired Nuel and started to conduct research on narrowband cellular IoT technology with Vodafone, proposing the narrowband technology NB M2M. In May 2015, Huawei and Vodafone jointly developed the relevant upstream and downstream technology standards with Qualcomm, fusing NB OFDMA to form NB-CIoT.
NB-CIoT proposes a new air interface technology, which is relatively more modified on the existing LTE network, but NB-CIoT is the only one among the six Clean Slate technologies proposed to meet the five major objectives (improving indoor coverage performance, supporting large-scale device connectivity, reducing device complexity, and reducing power consumption and latency) proposed in the TSG GERAN #67 conference. Cellular IoT technologies, especially NB-CIoT, have lower communication module costs than GSM modules and NB-LTE modules.
At this time, Ericsson and Nokia jointly launched the narrowband cellular technology NB-LTE, which is more similar to the positioning of NB-CIoT, but NB-LTE is more inclined to be compatible with existing LTE, and its main advantage is easy deployment. In July 2015, Ericsson and Huawei submitted standard proposals to 3GPP respectively.
Finally, after intense discussion and negotiation of unification at the RAN #69 meeting in September 2015, the two technologies were converged by 3GPP in the Rel-13 version to form the NB-IoT standard.
The evolution of NB-IoT from a narrowband technology to a formal 3GPP standard, the active promotion by relevant vendors and operators, and the real existing market demand are two factors that cannot be ignored.
3GPP’s communication technology standards can be mainly divided into Core Parts (main function), performance standards, and RF conformance test standards.
Among them, the main function standards refer to the specific content of the protocol, including signaling protocols, network access, etc., which are mainly related to development; the performance standards are mainly for each sub-technology area.
Performance standards are mainly the performance of each subtechnical area, strongly related to the test.
Consistency test standards, mainly include some process and functional test standards.
SA/CT aspects
Since Rel-12, 3GPP has been gradually studying the core network architecture for MTC communication enhancement, and Rel-13 has focused on NB-IoT and DECOR/eDECOR-related technologies.
Most of the main standards related to NB-IoT on the core network side of 3GPP are in stage2 (service and system architecture), and the work related to stage3 (core network and terminal) is launched from the second half of 2016 to early 2017.
In order to meet the massive fragmentation, low-cost, low-rate, and low-power NB-IoT IoT applications, the core network aspect mainly considers the following aspects.
Efficient support for infrequent small packet transmission
The processing efficiency of infrequent small packet transmission is further improved for NB-IoT. Since the number of NB-IoT terminals may grow exponentially, the data volume and communication period of each terminal is relatively low, and the existing EPS core network (based on the S1 interface) will be very inefficient and risk overloading to handle such services.
Therefore, there is a need to minimize the signaling overhead of the entire EPS system, especially in the null part (e.g., RRC connection establishment and release), and also to strengthen the EPS system security process (this part is out of the SA WG).
There are two optimization directions, one is the control-plane-based optimization scheme, i.e., transmitting small packets through the NAS process; the other is the user-plane-based optimization scheme, i.e., catching the user’s context at both UE and RAN nodes through the RRC suspend state to reduce the signaling interactions.
Both of these optimization schemes have been added in TS23.401 Rel-14, with scheme 1 as a mandatory option and scheme 2 as an optional one. At present, 3GPP prefers to adopt the control-based optimization scheme, and the main work of this part of the standard in CT (core network and terminal) is still in progress.
Efficient support for tracking devices using small packet transmission
3GPP in the Rel-14 version, its service model belongs to the variant of the MAR (Mobile Terminal Periodic Reporting) service model, which needs further enhancement and optimization in terms of positioning, mobility, and transmission efficiency.
Efficient paging area management
For the massive static or restricted mobility terminals, 3GPP SA2 is still discussing paging optimization due to scarce airport resources and limited core network interface resources and will improve this part of the function in Rel-14.
The main idea of paging optimization is to consider paging only in the eNB or cell where the user last accessed instead of the whole TA (initially, it is assumed that the TA code of the NB-IoT cell is different from the TA code of the existing eNB cell), in order to save the airports and the related resources of the core network.
In the same coverage area, the devices of NB-IoT are massive, much more than the traditional cellular terminal devices. It is possible that operators operating in the narrowband spectrum do not provide enough resources required for the paging, and identification of UEs (S-TMSI, IMSI).
Compared to conventional cellular, it is extremely limited to include the above identification in a single paging message due to the message volume limitation of small data packets.
On the other hand, coverage enhancement is mandatory in the standard, so paging messages may take a longer time (a longer interval period for sending the same paging message repeatedly).
Most NB-IoT devices are considered to be stationary or rarely mobile, so their paging range can be limited and they do not need to be paged throughout the TA they belong to, which can reduce the consumption of paging resources.
However, when the UE enters IDLE mode, the cell information reported by the eNB to the MME for the last time it served the NB-IoT UE may be inaccurate (this possibility exists even for stationary users).
This is because, in the case of a UE at rest, the change of the user’s primary service cell may be caused by various reasons, such as a change in RF load conditions, change in RF conditions in neighboring cells (similar to the blocking of buildings, which causes the UE to access other base stations).
DECOR/eDECOR
When the existing network is deployed, there may be multiple DCNs (Dedicated Core Networks) of NB-IoT in the core network.
According to the output of TSG RAN side TS23.236, the NB-IoT DCN may be connected to both E-UTRAN and NB-IoT RAN nodes and two different schemes can be adapted to select a suitable DCN for them according to the user type.
One is the redirection scheme, referring to the TR23.707 DECOR function; the other is UE-assisted, referring to eDECOR in TR23.711.
From the current progress of the protocol, DECOR is less likely to be deployed and may be used as a transitional solution since the redirection process leads to additional signaling interactions between the UE and the RAN and the network side.
And eDECOR is still in the early research stage due to its impact on UE, will be gradually improved in the late Rel-14, and is expected to be widely adopted in the future with the deployment of virtualized networks.
Support non-IP data types
Non-I data usage is common in M2M applications, such as 6LowPAN, MQTT-S, etc. When such applications are deployed in NB-IoT networks, non-IP data between the application server AS or service capability server SCS and the user needs to be transmitted over the network, and two options are available.
One is to transmit through the SGi interface via the non-IP exclusive PDN point-to-point tunneling method, and the other is to transmit through the SCEF.
At present, since the T6a interface between CSGN and SCEF is still in the preliminary research stage, and the transmission of non-IP data through the SGi interface can make C-SGN unified data egress, which is convenient for future billing point selection and billing mode design for NB-IoT-type services, therefore, SGi method may be preferred by operators.
Support SMS
Some existing M2M services are supported by SMS. In order to have comprehensive coverage of such services, two issues need to be considered after the deployment of NB-IoT.
Whether to retain joint attachment for SMS delivery capability or only PS attachment.
Whether there will be terminals and their solutions that only use SMS for messaging without establishing any PDN connection.
The solution of SMS support in the NB-IoT core network will be further improved in Rel-14, but operators can consider whether to deploy the SMS function in their existing network according to their actual needs.
For example, only the deployment of IP and non-IP data bearing mode, mainly because the support of SMS function requires the opening of SGd interface between C-SGN and SMS center, and the need to upgrade the existing network SMS center, the CSGN also has relevant functional requirements.
Authorized users support coverage enhancement (CE) technology
For users with a poor propagation environment, such as devices in underground pipes, strong penetration performance is required, and CE technology needs to be used at this time to obtain a better penetration effect. However, the use of CE technology requires additional resources from the network side.
Therefore, users should be authenticated and the users who can use CE technology should be restricted to ensure that only users who have signed up and received CE authorization can enjoy this feature and achieve differentiated services.
OverLoad control
On the topic of reducing the risk of core network overload, 3GPP has initiated several studies and proposed various schemes including access hierarchy, eNB-assisted (denial, delay, queuing on the eNB side), and so on.
And in TS23.401, the congestion control scheme adopted for NB-IoT devices is based on the upgrade of the original backoff timer mechanism of the EPS system, which uses a discrete approach to process concurrent requests from NB-IoT devices to achieve overload control.
Header compression enhancement
Since most of the NB-IoT application scenarios use small packets and low communication frequency, such as periodic MAR (Mobile Autonomous Reporting) and NC (Network Command) use 20~200 byte/30min or longer interval data transmission.
Considering the header overhead of IP and transport layer, such as 20 bytes for IPv4, 40 bytes for IPv6, 8 bytes for UDP, 20 bytes for TCP, and 12 bytes for RTP, header compression enhancement is imperative to support massive NB-IoT/eMTC type terminals more efficiently.
Due to infrequent data transmission and mobility, the header compression context retained in eNB and UE may be reset (e.g., when UE enters IDLE mode or switches eNB), which will result in full header overhead or additional overhead for data packets if data is sent or moved frequently.
At this time, header compression will be an important guarantee to efficiently support IP-like small packet services. Therefore, when adopting a scheme based on control-plane optimized packet transmission, the header compression function needs to support the transition from connected to IDLE state and mobility management of NB-IoT end users.
It should also be noted that the IP header compression function must be turned off when non-IP class service scenarios occur, so the network side also needs to decide whether to enable the header compression function according to different situations.
LTE-M, EC-GSM, and NB-IoT Evolution
The Internet of Everything is a major trend and an inevitable development, and various IoT technologies are also shuttles.
Facing various emerging IoT technologies, 3GPP has three main standards: LTE-M, EC-GSM, and NB-IoT, which are based on LTE evolution, GSM evolution, and Clean Slate technology respectively.
What is LTE-M?
LTE-M, or LTE-Machine-to-Machine, is an IoT technology based on LTE evolution, called Low-Cost MTC in R12 and called LTE enhanced MTC (eMTC) in R13, designed to meet the needs of IoT devices based on the existing LTE carrier.
Those who know LTE UE categories will not be unfamiliar. In order to adapt to IoT application scenarios, 3GPP defined the lowest rate UE device was UE Cat-1 in R11, which has an uplink rate of 5Mbps and downlink rate of 10Mbps.
To further adapt to the low-power and low-rate needs of IoT sensors, in R12, 3GPP defined a lower-cost, lower-power Cat-0 with an uplink and downlink rate of 1Mbps.
What is EC-GSM?
EC-GSM, or Extended Coverage-GSM technology. With the rise of various LPWA technologies, the disadvantages of traditional GPRS applications for IoT are highlighted.
In March 2014, the 3GPP GERAN#62 conference “Cellular System Support for Ultra LowComplexityandLow Throughput Internet of Things” proposed a research project that would narrowband (200kHz) IoT technology to GSM, seeking a wider coverage than traditional GPRS by 20dB, with 5 major objectives.
Improve indoor coverage performance, support large-scale device connectivity, reduce device complexity, and reduce power consumption and latency.
In 2015, the TSG GERAN #67 conference report stated that EC-GSM has met the 5 major objectives.
GERAN (GSM EDGE Radio Access Network) is an acronym for GSM/EDGE Radio Access Network.
GERAN is led by 3GPP and focuses on developing GSM standards. Since the early cellular IoT technology is based on GSM, some IoT projects are conducted by GERAN.
As the technology evolved, cellular IoT communications needed to be redefined, which led to NB-IoT. since NB-IoT technology is not based on GSM, it is a clean-slate solution, so the work on cellular IoT was transferred to the RAN group. geran will continue to work on EC-GSM until the R13 NB-IoT standard is frozen.
What is NB-IoT?
In August 2015, 3GPP RAN started a project to study a completely new air interface technology for narrowband radio access, called CleanSlateCIoT, and this Clean Slate solution covers NB-CIoT.
NB-CIoT was jointly proposed by Huawei, Qualcomm, and Neul, and NB-LTE was proposed by Ericsson, Nokia, and other manufacturers.
NB-LTE prefers to be compatible with existing LTE, and its main advantage is easy deployment. Ultimately, negotiated and unified at the RAN #69 meeting in September 2015, NB-IoT can be considered as a convergence of NB-CIoT and NB-LTE.
In contrast to the MTC technology optimization approach, the Cellular Internet of Things (CIoT) technology project proposes a completely new design for IoT characteristics, not necessarily compatible with the established LTE technology framework.
The fierce competition that NB-IoT will face after commercialization.
The emergence of innovative technology will inevitably lead to a battle with the traditional technology, which may be a fish-out-of-water, a compromise and a peaceful coexistence, or a surrender of multiple parties.
With the emergence of NB-IoT innovative technology, the visible competition will be reflected in the following three aspects and illustrated with examples.
Competing NB-IoT vs LoRa technology solutions
NB-IoT vs LoRa using spectrum
In the field of low-power wide area network in China, NB-IoT and LoRa are undoubtedly the two most popular low-power wide area network (LPWAN) technologies.
The two form two major technology camps, NB-IoT represented by Huawei on one side and LoRa represented by ZTE on the other.
There is no doubt that the radio spectrum is a national resource, a limited resource that is not renewable and can only be used wisely. Here’s a look at the frequency bands used by the two technologies.
NB-IoT frequency band in NB-IoT vs LoRa
NB-IoT uses the authorized frequency band and has three deployment methods: standalone deployment, protected band deployment, and in-band deployment. The global mainstream frequency bands are 800MHz and 900MHz.
China Telecom will deploy NB-IoT on the 800MHz band, while China Unicom will choose 900MHz to deploy NB-IoT, and China Mobile will probably re-farm the existing 900MHz band.
NB-IoT belongs to the authorized frequency band, just like 2G/3G/4G, which is a specially planned band with relatively less interference in the band. NB-IoT network has the standard of the carrier-grade network, which can provide better network standards of signal quality of service, security, and authentication.
It can be integrated with existing cellular network base stations and is more conducive to rapid mass deployment. Operators have a mature telecom network industry ecosystem and experience to better operate NB-IoT networks.
From the current point of view, NB-IoT network technology will only be deployed by the network operators above, other companies or organizations cannot deploy the network by themselves.
To use the NB-IoT network one must wait for the operator to lay the NB-IoT network, and its progress and development depend on the construction of the operator’s infrastructure network.
LoRa frequency band in NB-IoT vs LoRa
LoRa uses license-free ISM bands, but the use of ISM bands varies from country to country or region to region. The following table shows some of the bands used as mentioned in the LoRa Alliance specification.
In the Chinese market, the China LoRa Application Alliance (CLAA), led by ZTE, recommends the use of 470-518 MHz. 470-510 MHz is the frequency band used for radio metering meters.
The 470-510MHz band can be used as a frequency band for civilian radio metering meters under the condition that the transmitter’s operating time does not exceed 5 seconds when transmitting data.
The frequencies used are 470-510MHz and 630-787MHz. transmit power limit: 50mW(e.r.p).
Since LoRa works in the license-free band, no application is required for network construction, the network architecture is simple and the operating cost is low.
The LoRa Alliance is vigorously promoting the standardized LoRaWAN protocol globally, which enables devices conforming to the LoRaWAN specification to interoperate. China LoRa Application Alliance has made improvements and optimization on the basis of LoRa to form a new network access specification.
Communication distance of NB-IoT vs LoRa
Communication distance and communication capability of NB-IoT vs LoRa are the most important performance indicators of wireless communication under the premise of equal power consumption.
NB-IoT communication distance in NB-IoT vs LoRa
The signal coverage of the mobile network depends on the base station density and link budget. 164dB link budget for NB-IoT, 144dB (TR 45.820) for GPRS and 142.7dB (TR 36.888) for LTE.
Compared with GPRS and LTE, NB-IoT has a 20dB improvement in link budget, and the signal coverage in an open environment can be increased by seven times.
20dB corresponds to the loss of signal penetration through the exterior wall of the building, and the signal coverage in an indoor environment is relatively better with NB-IoT. Generally, the communication distance of NB-IoT is 15km.
LoRa communication distance in NB-IoT vs LoRa
Lora provides a maximum link budget of 168dB and power output of +20dBm with its unique patented technology. Generally, the wireless distance range is 1~2km in the city and up to 20km in the suburbs.
Relay for NB-IoT vs LoRa
In the actual network deployment, both NB-IoT and LoRa wireless network signals will not cover places, which can be called signal “blind areas”, and if the signal coverage is achieved by setting up more base stations for “blind areas”, it will inevitably result in the higher network The construction cost is high.
This requires a low-cost “relay” product to expand and extend the network to complete the signal coverage in the “blind” areas.
Chip source of NB-IoT vs LoRa
Both NB-IoT and LoRa networks need radio chips to connect and deploy.
NB-IoT works in the licensed band, which is basically a market for operators, and base station equipment is generally provided by communication equipment service providers. Lora works in the license-free band, and any company can design and develop its own gateway and set up its own network.
NB-IoT vs LoRa ends wireless RF chip companies.
LoRa chips companies
LoRa technology is patented by Semtch, and Semtech provides SX127x series LoRa products. The domestic market mainly focuses on SX1278 in the low-frequency band (137-525MHz). In order to meet the development and demand of the market, Semtech has granted more companies to manufacture the chips of LoRa technology by IP licensing, similar to the IP licensing of ARM.
At present, Semtech’s IP licensing companies include Hoperf, Microchip, Gemtek, ST, etc. HopeRF’s LoRa products are data transmission modules, Microchips are LoRaWAN modules, and Gemtek has made LoRaWAN products with SiP. In the future, there may be more companies to manufacture LoRa technology products through IP licensing.
NB-IoT chips companies
NB-IoT is supported by telecom operators and telecom equipment service providers, with a mature and complete telecom network ecosystem.
Huawei:
Huawei’s NB-IoT chip is Boudica, an ultra-low-power SoC chip based on an ARM Cortex-M0 core, which will be equipped with HuaweiLiteOS embedded IoT operating system. It is expected to be available in early 2017.
ZTE Microelectronics:
ZTE Microelectronics’ chip for NB-IoT is Wisefone7100. isefone7100 is said to have an internally integrated CK802 chip from ZTE Microsystems. It is expected to be available in Q2 2017.
Intel:
XMM7115 supports NB-IoT standards. samples available in H2 2016. xMM 7315, supports both LTE Category M and NB-IoT standards, the single chip integrates an LTE modem and IA application processor. Commercialization is expected in 2017.
Qualcomm:
MDM9206 with Cat-M (eMTC) and NB-IoT support.
Nordic:
NordicSemiconductornRF91 family is Nordic’s NB-IoT cellular technology offering. Samples are expected to be available in the second half of 2017, with availability starting in 2018.
Other NB-IoT chip vendors may include Sequans, Altair, Janjona Electronics Ltd, MARVELL, MTK, RDA, etc.
Cost of NB-IoT vs LoRa modules
Recently, SNS Telecom predicted the cost of a typical LPWA module to be $4-18, with module prices varying by technology. As LPWA network deployments mature, the cost per module is expected to drop to $1-$2 in volume.
What is the cost of NB-IoT vs LoRa?
Let’s do a simple collation of NB-IoT vs LoRa from some public information.
Cost of NB-IoT modules of NB-IoT vs LoRa
Huawei mentioned in “NarrowBandIoT Wide Range of Opportunities WMC2016” that the price of the NB-IoT chipset is $1-2 and the price of the module is $5-10.
Cost of LoRa module of NB-IoT vs LoRa
Since LoRa is commercially available earlier, there are many companies selling LoRa modules in the market. Here, we will not discuss the module of data transmission, but only the module based on the LoRaWAN protocol.
Comparison of NB-IoT vs LoRa solutions
NB-IoT works in the authorized band, the devices need to enter the network license, and the interference will be relatively less, LoRa works in the license-free band, and the license-free band relatively more kinds of devices inevitably will be interfered with by other wireless devices.
The advantage of LoRa is its patented technology, which can maintain high acceptance sensitivity and strong anti-interference ability even in complex environments.
The data rates of NB-IoT vs LoRa are different, with LoRa data rates up to 50kbps and NB-IoT up to 200kbps.
The different data rates of the two technologies actually form different market segment applications, and the suitable technology can be selected according to the actual project requirements.
From the NB-IoT vs LoRa chip products, many of them have integrated MCU or processors, which makes it easier for signal and data processing and communication protocol management.
NB-IoT vs LoRa wireless networks are deployed in different environments, and the NB-IoT vs LoRa communication distance will be different.
The “blind spot” issue needs to be taken into account during actual deployment. It can also be combined with other wireless technologies (such as FSK, etc.) to solve the “blind spot” problem (of course, the PCB design and antenna matching will also affect the communication distance).
Lora can be made into SoC or SiP products through licensing and can be integrated with some product technologies to meet different market demands. For example, Semtech’s EV8600 is an SoC product that combines PLC and LoRa.
NB-IoT vs LoRa has a common drawback in battery product applications, because of power consumption, it cannot do real-time communication and is only suitable for conditional or timed upload.
However, LoRa can start the low-power FSK short-range networking method to achieve two-way real-time communication; NB-IoT can only obtain downlink commands by constantly communicating with the platform, but too often, it will increase power consumption, and if the communication frequency decreases, there will be the problem of decreasing user experience.
NB-IoT vs LoRa Conclusion
NB-IoT vs LoRa, each has its own advantages. We need to choose the appropriate technology according to the actual project requirements and our own situation.
Both NB-IoT and LoRa are still in the initial stage of development and require input from all parties and joint development. When large-scale deployment becomes a realistic possibility, the cost of NB-IoT vs LoRa modules will naturally be further reduced.
Besides this NB-IoT vs Lora article, you may also be interested in the below articles.
What is the difference between WIFI and WLAN?
Summary of 41 Basic Knowledge of LTE
What Is The 5G Network Slicing?