WO2014024097A1

WO2014024097A1 - Method, device and system for determining a channel state in a wireless network

Info

Publication number: WO2014024097A1
Application number: PCT/IB2013/056298
Authority: WO
Inventors: Michael Petrus Franciscus VERSCHOOR; Daniel Martin GÖRGEN; Tim Corneel Wilhelmus Schenk
Original assignee: Koninklijke Philips N.V.
Priority date: 2012-08-07
Filing date: 2013-07-31
Publication date: 2014-02-13

Abstract

A method, device and system for determining a channel state in a wireless network are provided. A method for determining a channel state in a wireless network comprises the steps of monitoring (S10) a channel of the wireless network, performing (S20) at least one randomly timed measurement within a predetermined time interval, and determining (S30) a channel state based on the at least one randomly timed measurement.

Description

Method, device and system for determining a channel state in a wireless network

FIELD OF THE INVENTION

The invention relates to a method, a device and a system for determining a channel state in a wireless network.

BACKGROUND OF THE INVENTION

In wireless communication technology, transceivers that are located sufficiently close together and that emit radio waves at (approximately) the same frequency and with a sufficient power level are known to interfere with one another, thereby reducing a reliability and throughput of the data that is communicated via the radio waves.

In the context of distributed wireless systems, the systems that utilize wireless communication technology to exchange data between the constituting devices, one can distinguish between two types of interference: interference caused by radio sources that are part of the same distributed system, and interference caused by radio sources that are not part of the system. These types of interference are referred to as internal interference and external interference, respectively.

Current state-of-the-art wireless communication systems define a number of mechanisms to regulate the access to a shared RF (radio frequency) medium in order to avoid significant levels of data loss due to interference. The defined mechanisms have in common that they either assume a certain level of (prior) coordination between the devices that wish to access the shared medium, for instance allowing a device to only access the medium during specific time slots (TDMA) or using specific carrier bands (FDMA), or that each device checks whether the medium is actually free before it sends data into the medium (CSMA). As a result, these mechanisms do not cope well with devices that do not adhere to the medium access policy. Such devices may access the medium at the very moment that a nearby device that is part of the distributed system sends data into the medium, based on the assumption that it has the exclusive right to access the medium or following a check that the medium is actually free. The resulting collision of both radio signals can result in a loss of data, hampering system performance and reliability. This specific type of external interference is referred to as a non-compliant external interference.

The performance of a distributed wireless systems in terms of its reliability, responsiveness and data throughput can be critically affected by this external interference. For instance, for a number of the currently marketed wireless lighting systems, the radio spectrum usage in the direct environment of a system is scanned prior to the installation or commissioning. To perform the scan, dedicated tools or a special commissioning system are used. Based on this information, a suitable RF channel is selected for future use by the system.

This approach works reasonably well for small scale installations that are only confronted with static external interference. Large scale management systems or outdoor deployments are much more likely to suffer significantly from dynamic interference in terms of location, power level and time. Examples are outdoor telemanagement and light-on- demand systems. This is specifically relevant for wireless systems used for lighting control, as they typically use unlicensed spectra (for instance the 2.4 GHz, 868 MHz and 900 MHz bands) for communication. In such bands a system does share a medium with other transceivers, and the bands are typically only regulated in terms of the maximally permitted transmit power level.

A number of processes are known in the prior art to detect and quantify the interference experienced by a distributed wireless system. However, these approaches are either unable to distinguish between internal and external interference, or they rely on some form of synchronization between the devices that constitute the system to enable an energy (inference) scan over the relevant frequencies during coordinated periods of radio silence of the system. The disadvantage of synchronized solutions is two-fold. First, the

synchronization process itself will consume bandwidth, thus reducing the bandwidth that is available to the system for data communication. Secondly, the system as a whole will be unresponsive during the synchronized measurement, as all communication is delayed for the duration of the energy measurement.

WO 2010/122488 Al describes a device for detecting a channel state. The detecting device first monitors a channel with a predetermined duration so as to receive a plurality of signals transmitted on the channel, and then determines whether part of the plurality of signals possesses periodicity. If part of the plurality of signals possesses periodicity, then the detecting device determines that the channel is interfered, so as to avoid potential interference.

SUMMARY OF THE INVENTION In view of the above disadvantages and problems associated with the prior art, it is an object of the present invention to provide a method, a device and a system for reliably determining a channel state in a wireless network independently of an operation state of the wireless network.

This object is solved by the features of the independent claims.

The invention is based on the idea to determine a channel state by using at least one measurement that is performed randomly timed within a properly chosen time window (predetermined time interval). For instance, in case of a distributed wireless system, the channel state may relate to a level of external interference. Accordingly, the measurement may be a measurement of an interference level experienced within the channel at a time and/or location of the measurement. However, such a measurement generally includes a combined level of internal and external interference. By randomly timing the at least one measurement within a properly chosen (predetermined) time interval (window), the probability that a specific measurement is affected, i.e. corrupted, by internal interference is relatively low compared to the probability that it is affected by external interference.

Accordingly, it is possible to detect external interference during normal operation of the distributed wireless system and without the need of any form of synchronization or the guaranteed absence of internal interference.

According to one aspect of the present invention, there is a method provided for determining a channel state in a wireless network, comprising the steps of monitoring a channel of the wireless network, performing at least one randomly timed measurement within a predetermined time interval, and determining a channel state based on the at least one randomly timed measurement.

Accordingly, by performing, within the predetermined time interval, randomly distributed measurements, it may be reliably determined if a channel is affected by external interference. Preferably, a starting time of said time interval is determined when at least one predetermined condition is fulfilled during the monitoring. For instance, the predetermined condition may be fulfilled when no transmission of data packets on the channel is present or a transmission of data packets on the channel is below a predetermined threshold. Hence, a channel state related to external interference may be reliably determined, since the randomly timed measurements are mostly affected by said external interference. In particular, the channel state can be determined during normal operation of the distributed wireless system and without the need of any form of synchronization between individual nodes of the system or with the guaranteed absence of internal interference, thus overcoming the disadvantages of the prior art described above.

Preferably, the measurement is a measurement of at least one of an interference level, a signal or signal level, a power level, a raw received baseband signal, and a power level in a certain frequency band. In a preferred embodiment, at least two of these measurements may be combined in order to increase the reliability of the determined channel state. In particular, a signal level and an interference level may be used to calculate a signal- to-noise ratio, wherein said signal-to-noise-ratio may then be used to determine the channel state. It is an advantage that for these measurements receivers/transceivers included in a conventional node of the wireless network can be used. Hence, the method can be employed in existing networks, without the need for any hardware modifications.

Preferably, multiple measurements taken within at least one predetermined time interval are collected and combined for determining the channel state. In one preferred embodiment, this combination refers to statistics of the measurements. In another preferred embodiment, during a measurement window (time interval), a number of measurements (e.g. energy detections, "ED") may be planned ("plannedEDcounter"). Due to a busy radio (TX or RX), a number of EDs may be skipped ("discardedEDcounter"), in order to enhance the probability that the measurement is only affected by external interference.

"discardedEDcounter" is an indication of the business of the channel/medium (traffic density). Thus, it is an indication of the amount of EDs, which were corrupted by concurrent own (network-internal) traffic/interference. From the performed ED samples,

, , , , _^ ,. , ,,,, _ durationi ED) + durationi preamble) corruptedEDcounter = discardedEDcounter*

average _ duration(RX _ busy) are removed as they are assumed to be corrupted/affected. These can be for instance the corrupted ED counter samples with the highest measured energy value. The remaining ("plannedEDcounter - discardedEDcounter - corruptedEDcounter") samples may then be statistically interpreted. According, the reliability of the determined channel state may be significantly improved.

Preferably, a measurement is discarded when a data packet transmission is detected during the measurement. In analogy to the afore-said, when a data packet transmission is detected during the measurement, the probability that the measurement is affected by internal interference is enhanced. Accordingly, by discarding such measurements, a reliability of the determined channel state, i.e. the probability that the measurement is affected by external interference only, is increased.

Preferably, the measurement and/or additional (successive) measurements are performed after a data packet has been received and/or a data packet has been sent. Again, by performing measurements after reception or sending of a data packet, the probability that the measurement is affected by internal interference is reduced. Accordingly, it is more likely that the measurement is affected by external interference only and hence, the reliability of the determined channel state is improved.

Preferably, a data packet is detected based on a known signal sequence preceding a data packet and/or based on a system/protocol specific header field. Accordingly, it is ensured that only system-internal data packets are taken into account (e.g. for deciding if a measurement should be discarded). In other words, it can be reliably concluded that the detected data packet contributes to internal interference rather than external interference, and that the measurement during which said data packet is detected should be discarded.

Preferably, a measurement period of the at least one randomly timed measurement has a same or shorter duration than the known sequence preceding a data packet. Accordingly, by keeping the measurement period/time short, the probability that the measurement is affected by internal interference is reduced. In particular, as will be explained later in this specification, the probability that a measurement is affected by external interference but not by internal interference may depend on a duration of the known sequence preceding the data packet. In other words, said probability increases when a duration of the measurement decreases.

Preferably, a measurement period of the at least one randomly timed measurement is shorter than a time needed to transmit a data packet of a representative length. For instance, in IEEE 802.15.4-2006 using O-QPSK modulation with 100 Kbps, the representative length may be a packet duration of 3.2 ms to 10 ms, a duration of energy detections (e.g. 320 μ8), or the duration of a preamble of a data packet (e.g. 400 μ8).

Preferably, at least one performance metric is calculated from the at least one measurement. Possible performance metrics are:

- the percentage of measurement values above (or below) a certain predetermined threshold,

the longest time period during which successive measurements (e.g. ED samples) were above (or below) a predefined threshold, the number of slots in a measurement window, during which all successive measurements (ED samples) were below a predefined threshold, and the slot was longer than a predefined time period (for instance a time required to sent one data packet), which is 10 ms in IEEE 802.15.4-2006 with 100 Kbps, and

an estimation of the duty cycle percentage and/or pattern of the interference either within the measurement window or across multiple measurement windows.

According to a further aspect of the present invention, a device for

determining a channel state in a wireless network is provided. The device preferably comprises a monitoring unit adapted to monitor a channel of the wireless network, and a control unit adapted to perform at least one randomly timed measurement within a predetermined time interval and determine a channel state based on the at least one randomly timed measurement.

Preferably, the device is adapted to perform the above-described method.

Preferably, the device is associated with at least one node of the wireless network. The device may be integrated in or operatively coupled (attached) to a node of the wireless network. For instance, in case of a lighting system, the device may be integrated in or attached to a light source (e.g. a street light). Accordingly, each device may reliably determine a channel state for its associated node, e.g. if external interference is currently present at the location of the node or not.

Preferably, the device is further adapted to identify sources of external interferences based on the measurements. For instance, the device may analyze a temporal behavior and/or (energy) levels of the external inference and compare the analysis result with known interference patterns of known external interferers (e.g. stationary transformers).

Further, the device may send the measurements and/or analysis results to a central controller, which may then create a map of stationary interferers that affect the network, as described below.

Preferably, the control unit is further adapted to increase a frequency of the randomly timed measurements when

an amount of received acknowledgements is below a predetermined threshold, and/or

an amount of received packets is below a predetermined threshold, and/or an amount of received packets rejected by CRC (cyclic redundancy check) is above a predetermined threshold, and/or

a number of observed neighbors of the node is below a predetermined threshold, and/or

- a request from central controller for one of the neighbors is received, and/or

an amount of carrier sensing failure operation is above a

predetermined threshold.

Accordingly, when the above conditions occur, there is a certain probability that there is an increased level of external interference, since system performance is degraded. Accordingly, by increasing the frequency of the randomly timed measurements, the channel state may be determined more quickly. Thus, the device may react quickly to a sudden/dynamic change of the channel state (e.g. due to a changed external interference level), e.g. by changing communication parameters based on the current determined channel state. Further, increasing the frequency increases the probability that the measurement is affected by external interference only but not by internal interference.

Preferably, the control unit is further adapted to change at least one communication and/or network parameter of the device and/or the network based on the determined channel state. Accordingly, if external interference of a high level is observed, the control unit may adapt communication parameters, for instance transmission power, in order to counter-act the influence of the external interference. Hence, system performance of the wireless network can be adapted according to the occurrence of external interference.

Preferably, the device additionally measures and stores the power levels during normal receptions of valid (system-internal) data packets. These measurements may then be used in determining the channel state, e.g. as a reference.

According to a further aspect of the present invention, a system for determining a channel state in wireless network is provided, wherein the wireless network comprises a plurality of nodes, and the system comprises at least one of the above-described devices. The at least one device may transmit the monitoring results and/or the determined channel state and/or determined performance metrics and/or derivatives thereof to a central controller of the wireless network. The central controller may use the measurements and/or the determined channel state and/or determined performance metrics and/or derivatives of the at least one device (and preferably of a plurality of devices) to determine, which areas of the wireless network are affected by external interference. Accordingly, the system may create a map of external interference affecting the wireless network and may provide this map, for instance on a display, to a user/administrator. Further, the central controller may instruct at least some of the nodes to change communication parameters based on the received measurements and/or the determined channel state and/or determined performance metrics and/or derivatives and/or based on the determined map.

Preferably, the system is a lighting system. As mentioned above, the device may be associated with at least one node/lamp of the lighting system. This is in particular advantageous, since lighting systems are generally affected by a huge number of external disturbers, both static and dynamic. However, the present invention is not limited to lighting systems but may also be applied to a variety of other networks.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings:

Fig. 1 shows frame structure, transceiver states, and frame TX duration when using 100Kbps O-QPSK modulation as specified in IEEE 802.15.4-2006.

Fig. 2 shows an example of a wireless mesh network.

Fig. 3 shows a flow diagram of a method according to an exemplary embodiment of the present invention.

Fig. 4 shows a device according to an exemplary embodiment of the present invention.

DETAILED DESCRIPTION As a first illustrative example, an embodiment of the present invention is based on the key notion that although an energy level, which is measured by a transceiver within the bandwidth of a specific RF channel, is a measure for the combined level of internal and external interference experienced within said channel at that time and location, the probability that a specific energy measurement is actually affected (corrupted) by internal interference is relatively low compared to the probability that it is affected by external interference when some basic constraints are used to time the energy measurements.

In the following it is referred to figure 1, which shows an exemplary frame structure, transceiver states, and frame TX duration when using 100Kbps O-QPSK modulation as specified in IEEE 802.15.4-2006. A first step in ensuring energy measurements are 'rightly' timed may be to disallow or delay measurements while the transceiver is in a transmission state (TX or TX_BUSY) or while it is receiving a frame (RX_BUSY). In general, the transceiver cannot even perform an energy measurement while being in a transmission state, while the energy measurement will be certainly corrupted by the reception of a frame.

When the energy measurements are randomly timed during the remaining "listening" (RX) state, the chance that an individual measurement is corrupted is determined by the following two probabilities:

the probability that the energy measurement starts just after a neighbouring device (i.e., a device within the reception range of the transceiver) started the transmission of a frame, but before the transceiver status changed to the RX_BUSY state, and the probability that a neighbouring device starts a transmission during the energy measurement.

The first probability is directly related to the delay between the start of a transmission by a neighbour and the resulting change of the transceiver status from RX to RX_BUSY. This delay is approximately equal to the duration of the frame preamble including any synchronization headers, i.e. equal to the duration of the specific symbol sequence which signals the start of a new frame.

The second probability is directly related to the duration of the energy measurement.

As a result, the overall probability that an energy measurement that is randomly timed during a given interval in the RX state is corrupted by interference from a specific neighbouring device is approximately:

p + m

100 - i where p is the duration of the preamble and m the duration of the energy measurement, both expressed as percentages of the interval duration, and t is the percentage of time the specific neighbour spent on transmission during that interval.

Assuming that the neighbours of a device time their transmissions independently of each other except for the fact that they cannot overlap (i.e., ignoring the timing effects of back-off mechanisms), the probability that an energy measurement that is randomly timed during a given RX interval is corrupted by internal interference is approximately: p + m

loo -y^N _ti where p is the duration of the preamble and m the duration of the energy measurement, both expressed as percentages of the interval duration, and t is the percentage of time that neighbour i, with 1 · i · N, spends on transmission during the interval.

The probability that an energy measurement is corrupted by internal interference thus strongly depends on the duration of the preamble and the duration and number of the energy measurement(s) that are performed within a certain time interval. It is important to notice that the duration of the preamble is in general short compared to the duration of the transmission or reception of a complete frame - a difference of one (or more) order(s) of magnitude is not uncommon. As an example, in IEEE 802.15.4-2006 standard as shown in figure 1, the preamble including the synchronization header lasts 10 symbol periods, which equals 400μ8 when using 100Kbps O-QPSK modulation. The transmission of a frame with a maximum data payload of 127 bytes takes -l lms. "MHRS" designates MAC headers.

The duration of an energy measurement depends on the wireless communication technology and the modulation method that is used, but short measurements are often sufficiently accurate to detect interference, while the sensitivity can be easily improved by aggregating the results of multiple, independently timed measurements. In the IEEE 802.15.4-2006 standard, an energy measurement lasts 8 symbol periods or 320μ8 when using 100Kbps O-QPSK modulation.

As for the sensitivity to external interference: the probability that a randomly timed energy measurement is affected by external interference depends solely on the frequency and duration of the external interference. The probability that an energy measurement that is randomly timed during a given interval is affected by external interference is approximately equal to the percentage of time the external interference occurred during that interval divided by 100.

As a result, the probability that an energy measurement that is randomly timed during a given RX interval detects interference (i.e., is affected by external interference but not corrupted by internal interference) can be approximated by: i o⁾ 100 -ΓΥ where p is the duration of the preamble and m is the duration of the energy measurement, both expressed as percentages of the interval duration, t is the percentage of time neighbour i, with 1 · i · N, spends on transmission, and x is the percentage of time external interference occurs during the given interval.

Given the comparatively short duration of the preamble and the energy measurement, the probability that external interference is actually detected is reasonable (and can be improved by increasing the frequency of the energy measurements if needed) whenever x is sufficiently large compared to · t; with 1 · i · N. This is in general not a very limiting condition, as external interference that occurs significantly less than the summed internal interference can be regarded as rather irrelevant, while the amount of transmissions between neighbouring devices in a distributed wireless system will in practice be limited by sensible system design and occasionally by legal limitations.

Preferred applications of the present invention are outdoor lighting system (e.g. for streets, parking and public areas), indoor lighting systems for general area lighting (e.g. for malls, arenas, parking, stations, tunnels etc.) or sensor network.

In figure 2, a typical network topology is shown. A plurality of nodes 10 (N) is connected to each other by wireless communication paths 40. Some of the nodes 10 function as data collector nodes 50 (N/DC), which receive data packets from the other nodes 10 via single hop or multi-hop transmissions and transmit them to a control center or central controller 60 and vice versa. The wireless communication paths 40 between the nodes 10 and the data collector nodes 50 may be constituted by radio frequency transmissions, while the connections 70 between data collecting nodes 50 and the control center 60 may use the Internet, mobile communication networks, radio systems, Ethernet, DSL, cable or other wide or wireless data transmission systems.

Compared to other so-called ad-hoc mesh network, a telemanagement system for an outdoor lighting control network is stationary, i.e. the nodes 10 do not move.

Consequently, external interference will be mainly due to a changing environment, e.g. due to traffic, and due to stationary devices, such as transformers.

Figure 3 shows a flowchart of method according to an exemplary embodiment of the present invention. In step SI a waiting time is randomly determined. When the waiting time expires, a transceiver state is checked in step S2. If the transceiver state is TX,

TX_BUSY, or RX_BUSY (see also figure 1), the method returns to step SI. However, if the transceiver state is an RX state, the method proceeds with step S5. In step S5 the transceiver is blocked from entering the RX_BUSY and TX states. After that, in step S6 an energy measurement is performed. The measured energy value may be stored and/or an average energy value may be calculated or updated. Here, the method may use at least one energy value of the same measurement window, which at least one value has been previously measured (and was stored), to calculate the average energy value. A measurement window may comprises a relevant set of measurements and/or time windows. Afterwards it is checked, whether the measured energy value and/or the average energy value of the measurement window are significantly different from respective (average) energy value that can statistically be expected as a result of system-internal (own) interference. If it is determined in step S8 that the measured energy value and/or at least one further previously measured and stored energy value and/or the average energy value do not differ significantly from the respective expected values (i.e. are within a predetermined range), the method returns to step 1. However, if the measured energy value and/or at least one further previously measured and stored energy value and/or the average energy value differ significantly from the respective expected values (i.e. are without a predetermined range), an alarm may be given. For instance, an operator/user may be informed/altered using visual and/or acoustic means. Then, the method returns again to step 1.

Figure 4 shows a device 10 according to an exemplary embodiment of the present invention. Device 10 may include a monitoring unit 11, a control unit 12 and a transceiver 13. Monitoring unit 11 may be used for detection of data traffic on the channel, e.g. data packets as described above. Further, transceiver 13 may be used by monitoring unit 11 to monitor the channel of the wireless network. Furthermore, transceiver 13 may be controlled by control unit 12 and used to perform the at least one measurement (i.e., transceiver 13 serves as a measuring unit). Alternatively, device 10 may include any other suitable means for performing said measurement. In a preferred embodiment, transceiver 13 may be included in monitoring unit 11. It is noted, that, for the present invention, it is sufficient to provide a simple receiver instead of transceiver 13.

Distributed wireless systems or wireless communication technologies that need to detect or measure external interference without interruption of the normal operation by means of coordinated periods of radio silence will benefit from this invention.

When a system detects or measures external interference using coordinated measurements and periods of silence, the normal operation is temporarily suspended leaving the individual system parts unresponsive to commands that are normally conveyed (by the user or other system components) via the communication medium. The duration of the suspension depends on the size of the distributed wireless system (in terms of the number of system nodes) and the level of synchronicity that can be achieved with the communication technology. As an example, a typical lighting system, which may consist of several hundred up till a few thousand nodes (i.e., lamp poles), would require coordinated periods of radio silence of several seconds to guarantee that an energy measurement is not corrupted by coinciding internal interference. Using the invention, measurements can be performed in ~0.7ms, albeit with a small chance of corruption by system-internal communication.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. For example, it is possible to operate the invention in an embodiment wherein some

measurements are carried with a random timing and other measurements with a regular timing. Then, some decision as illustrated in the variants above can be taken to weigh down (or up) or exclude corrupted measurement.

Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. Method for determining a channel state in a wireless network, comprising the steps of:

monitoring (S10) a channel of the wireless network;

performing (S20) at least one randomly timed measurement within a predetermined time interval; and

determining (S30) a channel state based on the at least one randomly timed measurement.

2. Method according to claim 1, wherein the measurement is a measurement of at least one of

an interference level;

a signal;

a power level;

a raw received baseband signal; and

a power level in a certain frequency band.

3. Method according to claim 1 or 2, wherein multiple measurements over at least one predetermined time interval are collected and combined for determining the channel state.

4. Method according to one of the preceding claims, wherein a measurement is discarded when a data packet transmission is detected during the measurement.

5. Method according to one of the preceding claims, wherein additional measurements are performed after

a data packet has been received; and/or

a data packet has been sent.

6. Method according to claim 4 or 5, wherein a data packet is detected based on a known signal sequence preceding the data packet and/or based on a system/protocol specific header field.

7. Method according to claim 6, wherein a measurement period of the at least one randomly timed measurement has a same or shorter duration than the known signal sequence preceding the data packet.

8. Method according to one of the preceding claims, wherein a measurement period of the at least one randomly timed measurement is shorter than a time needed to transmit a data packet of a representative length.

9. Method according to one of the preceding claims, wherein at least one performance metric is calculated from the at least one measurement.

10. Device (10) for determining a channel state in a wireless network, comprising:

a monitoring unit (11) adapted to monitor a channel of the wireless network;

a measuring unit (13); and

- a control unit (12) adapted to

perform at least one randomly timed measurement within a predetermined time interval by using the measuring unit (13); and

determine a channel state based on the at least one randomly timed measurement.

11. Device (10) according to claim 10, wherein the device (10) is associated with at least one node (10, 50) of the wireless network.

12. Device (10) according to claim 11, wherein the control unit (12) is further adapted to increase a frequency of the randomly timed measurements when

an amount of received acknowledgments is below a predetermined threshold, and/or an amount of received packets is below a predetermined threshold, and/or an amount of received packets rejected by CRC (cyclic redundancy check) is above a predetermined threshold, and/or

- a number of observed neighbors of the node (10, 50) is below a predetermined threshold, and/or

a request from a central controller (60) or one of the neighbors is received, and/or an amount of carrier sensing failure operation is above a predetermined threshold.

13. Device according to one of claims 10-12, wherein the control unit (12) is further adapted to change at least one communication and/or network parameter of the device (10) and/or the network based on the determined channel state.

14. System for determining a channel state in a wireless network, wherein the wireless network comprises a plurality of nodes (10, 50), and the system comprises at least one device (10) according to one of claims 10-13.

System according to claim 14, wherein the system is a lighting system.