Communication
WLAN in the factory - the selection criteria
WLAN is suitable for various applications in the factory. However, due to the many influencing factors, it is not always obvious how a network based on the IEEE 802.11 standard can be optimized to meet the requirements of industrial use. - Here is some help.
WLAN networks are known to offer many new possibilities for the realization of industrial applications; however, the requirements for the quality of transmission go far beyond those of the office and private sector. Important indicators in this context are
■ the packet loss rate,
■ the latency (transmission delay for the delivery of a message),
■ the data throughput,
■ the roaming interruption and
■ the range of an access point.
Figure 1: The distance between signal and noise - the signal-to-noise ratio - is a key factor in determining the other quality characteristics of a WLAN network.
© BeldenA quality indicator that can be read directly in better WLAN devices is the signal-to-noise ratio. This ratio indicates how much stronger the useful signal of the transmission is compared to the ambient noise. With a high signal-to-noise ratio (SNR), it is possible to communicate clearly at high speed. If the SNR is low, the signal can no longer be easily differentiated from the loud noise. In this case, the WLAN devices only understand each other if they transmit their data very slowly and clearly. Figure 1 shows the SNR graphically: Up to an SNR of 6 dB, communication with the lowest data rate is generally possible in a WLAN network (for example, with a noise level of -86 dBm, a signal of -80 dBm is sufficient). High-quality WLAN devices allow the current SNR to be displayed via the web interface or via SNMP (Simple Network Management Protocol).
The following application scenarios serve to illustrate the influence of the various quality characteristics of a WLAN in practice:
Monitoring and control
Many systems today allow monitoring and control via the network. The use of mobile devices via wireless connections makes it easier for users to operate and maintain machines and systems. Interactive services require a high bandwidth here: for the transmission of video data, for example, 5 Mbit are required for each HD video stream and further reserves for the parallel use of the WLAN system by several users. Moderate response times (e.g. <100 ms) are also required for efficient interactive operation.
Applications such as augmented reality push the requirements even higher, depending on the system. Larger bandwidths (>10 Mbit/s per AR system - often in both directions) and short latencies (<50 ms) are important for a meaningful application. Good network availability at all points of production is also a must. The core requirements for a future-proof Wi-Fi supply in a production facility are therefore: high throughput, low latency and good area coverage.
Connection of system components
Due to the fast transmission of wireless signals and the long range of directional WLAN connections, the transmission standard is ideally suited as a technology for connecting individual spatially distributed system parts. Even networks several kilometers away can be connected via appropriate directional radio links. The range of a wireless system is therefore often of decisive importance. The latency and throughput of a connection are also important for smooth operation. In addition, depending on the frequency band used in outdoor applications, aspects of radar detection must also be taken into account. For applications that are sensitive to the loss of individual packets, redundancy technologies should also be considered. As many applications often share the point-to-point link, higher throughputs (>100 Mbit/s) are usually required.
The use cases mentioned as examples clearly show that a balance between the various performance features of a network is always important. But which technical aspects of a WLAN network actually influence the relevant parameters?
To answer this question, we will first briefly look at the basic technical principles of WLAN communication. Sufficient illumination of the area to be covered is a basic prerequisite for the optimizations discussed below.
In contrast to other radio technologies, WLAN networks use a CSMA/CA access method (Carrier Sense, Multiple Access, Collision Aviodance). This involves first 'listening' to the frequencies used to ensure that there is no competing transmission. If the radio medium is perceived as free, a WLAN device - a so-called station - can start transmitting. If the medium is busy, it must wait until the medium becomes free. In order to recognize faulty packets or packets that have successfully arrived at the receiver (so-called frames) at the transmitter, the receiver sends out an acknowledgement of receipt after successful receipt.
Media access procedures and coexistence
If this acknowledgement is not received, the sender can deduce that there is a reception error. If the retransmission fails several times in succession, the sending station aborts the transmission and transmits the next packet - the packet that has not arrived must then be regarded as lost/undeliverable.
The procedure described has an impact on the expected parameters such as transmission delay and packet loss. In largely interference-free situations, packet loss rates in the region of 0.1% can easily be achieved with WLAN connections. In disrupted situations, error rates in the range of several percent cannot be ruled out. In addition, disturbed transmissions also have a direct effect on the transmission time, the so-called latency, due to the multiple transmission of the packets. Furthermore, in disrupted situations, the medium is often already occupied by other transmitting stations, meaning that a wait is required before transmission can even begin. An important question when optimizing a WLAN network is therefore how systematic interference with transmissions (for example due to an overloaded channel) can be avoided from the outset or how successful transmission can be made possible even in disrupted situations - for example by using a different frequency range.
Use of frequencies and channels
The first fundamental decision with an influence on network quality is the choice of frequency range to be used for radio transmission. For wireless transmissions, for example, frequency bands in the 2.4 GHz and 5 GHz range are available, which in turn are divided into channels (Fig. 2). A channel represents a 20 MHz wide section of the frequency spectrum. The choice of frequency band has a significant impact on the range and frequency of interference from neighboring networks. The choice of frequency band also determines which coexistence mechanisms are to be used for shared use with other users of the frequency range.
The frequencies of the 2.4 GHz frequency band are traditionally used by a large number of different radio systems. On the one hand, this leads to a higher load on the radio channels; on the other hand, this fact has also resulted in strict regulatory requirements in Europe for compliance with coexistence mechanisms. Since the ETSI EN 300 328 V1.8.1 standard came into force, WLAN systems must comply with stricter requirements when accessing the transmission channel. One aim of these stricter requirements is to improve the coexistence of the various wireless communication systems used in the 2.4 GHz band.
For this reason, the receiving electronics must be able to adapt to the environment in order to detect other active transmitting systems in the current channel. If the transmission of a coexisting system is detected, own radio transmissions must be delayed until the medium is released again. To ensure that a transmitter does not occupy a channel for too long and thereby gain an advantage in the bandwidth it occupies, the following additional requirement is defined in the ETSI EN 300 328 V1.8.1 standard: Depending on the access mechanism used, an upper time must be observed when transmitting radio signals. In contrast to other wireless technologies, these requirements can be met relatively easily with WLAN connections.
Depending on the choice of frequency, coexistence mechanisms for shared frequency use with radar systems are also necessary in Europe. Channels 36, 40, 44 and 48 (5.15 - 5.25 GHz) are not affected by this. However, these channels are only approved for use in buildings. Other channels - for example the channels from 52 to 64 (5.25 - 5.35 GHz) and 100 to 140 (5.47 - 5.725 GHz) can also be used in outdoor areas. The regulatory authority in each country can release parts of the frequency band between 5.725 GHz and 5.875 GHz for WLAN use. In Germany, channels 155 to 171 (5.765 GHz - 5.865 GHz) of this frequency band are reserved for permanently installed broadband systems and therefore cannot be used by every operator.
When using outdoor channels in the 5 GHz range, the devices must recognize radar installations and, if necessary, withdraw from the frequencies used. Firstly, before a channel is used, it must be scanned for radar patterns for one minute, and secondly, if radar patterns are detected during operation, the channel must be released immediately. This means that the entire WLAN network must automatically switch to another channel. This usually leads to a brief interruption of all communication in the network.
The frequency bands also differ in terms of the permitted transmission strength. In general, it can be said that the frequency bands in the 5 GHz range allow a higher transmission power than the 2.4 GHz frequencies. However, it is worth comparing the data sheets and the legal regulations in order to fully exploit the potential of the legal framework.
Modulation schemes, MCS and rate adaptation
In WLAN systems, the signal quality between access points and clients can change frequently. As the signal-to-noise ratio cannot be accurately predicted from the outset, WLAN networks offer various adaptation mechanisms to achieve the highest transmission performance for a given situation. As a general rule, a higher transmission rate always requires a better signal-to-noise ratio, and lower transmission rates can also be achieved with a poorer signal-to-noise ratio.
Since the introduction of IEEE 802.11n, various techniques for increasing the transmission speed have been grouped into modulation and coding scheme (MCS) classes and numbered consecutively. For example, MCS 0 stands for the slowest and most robust transmission rate, while MCS 23 stands for the fastest data rate that can be achieved with three antennas and MCS 31 for the fastest data rate with four antennas. The resulting maximum gross transmission rates are, for example, 15 Mbit/s for MCS 0 and 600 Mbit/s for MCS 31.
An important component of these data rates is the number of spatial streams. Multiple input-multiple output technology (MiMo) makes it possible to transmit several signals on the same frequency simultaneously with several antennas without interference. For example, an access point with just one stream and one antenna (e.g. with IEEE 802.11n) can transmit a maximum of 150 Mbit/s, while an access point with four streams and four antennas can increase throughput to 600 Mbit/s.
In industrial plants, conditions often remain constant with regard to the required transmission speeds, so that a dynamic adaptation of the transmission rate to the mechanism that always has the highest transmission rate (and is most susceptible to interference) may have a negative effect. Adapting the data rate to the best possible transmission speed can in turn lead to the selection of a less robust transmission method and thus to higher packet losses.
In particular, if there is frequent switching between different data rates, as can be observed in mobile scenarios, for example, this leads to a fluctuating packet loss rate. With high-quality access points, it is therefore possible to set the maximum transmission rate - i.e. the highest MCS that should still be used. If only a data throughput of a few Mbit/s is required, a maximum transmission rate of MCS 10 or MCS 17, for example, can be set. This means that throughputs of up to 45 Mbit/s can still be achieved in practice, while the required signal quality is significantly lower than the signal quality of more sophisticated modulation schemes. Such a specification makes the behavior of a WLAN system much more controllable and has a positive effect on the packet loss rate and the variance in the transmission delay (jitter) - especially if it contains mobile parts. Switching between the modulation and coding schemes thus opens up scope for optimizing a WLAN system beyond the predominant optimization in the office environment, which always aims for the highest data rate.
Part 2 of the article series deals with the selection of the right antennas and further possibilities for increasing performance in industrial WLAN networks.
Authors:
Prof. Dr. Tobias Heer works on future technologies at Hirschmann Automation and Control;
Dr. Bernhard Wiegel works in the Embedded Software Development department at Hirschmann Automation and Control.
You might also be interested in
The future of CANopen
Similar to the automotive industry, industrial automation also requires more and more bus bandwidth. In addition, the cloud is becoming increasingly important in terms of communication. How does the CAN protocol, which has long been established in...
read more...Unauthorized access fended off
Plant modernizations often require new safety precautions. This was also the case for a manufacturer of cleaning agents and disinfectants, who had to safeguard the material flow in the area of a roller shutter door, among other things.
read more...Compact controls and drives a must
Smart factories promise operators a high degree of flexibility - for example through the use of driverless transport systems or mobile robot solutions. The automation of these subsystems in turn requires extremely compact control and drive concepts.
read more...Simply integrate safety
Machine and system manufacturers are increasingly turning to mapping safety functions via the communication bus. How can integration on the component side be achieved with reasonable effort?
read more...Flexible buffering via software
The complete automation of manual or semi-automatic processes is a major challenge with very short cycle times. When it comes to workpiece carrier transport, the machine manufacturer Goldfuß relies on a system that allows the creation of flexible...
read more...Individual IoT systems
How big? How much power? How robust? Which housing? The requirements for IoT systems are extremely varied. By designing individual embedded applications, devices can be created that are perfectly tailored to the application.
read more...A digital image
The digital twin accompanies machines and systems throughout their life - from the initial idea to ongoing operation and a look into the future. The aim is to avoid errors, optimize systems and prevent failures.
read more...Stumbling blocks with direct current
The question "direct current or alternating current?" is becoming increasingly important - not least due to the increasing use of renewable energy sources that generate direct current. When switching and protecting with direct current, however,...
read more...Traceability optimized
Directly marked 2D data matrix codes ensure the traceability of engine and transmission parts at KIA Motors. Read rates of 99 % are the target.
read more...