Mitigating PLC Interference to Broadcast Radio
CONTENTS
23.2 Test Scenario and Measurement Setup
23.3 Measurement Results and Analysis
23.3.1 Spectrum Analyser-Based Measurements on Power Lines
23.3.2 Field Trial for EMI Issues of PLC
23.4 Adaptive Detection for Cognitive PLC
23.1 Introduction*
In recent years, Smart Grid has received widespread attention from both academia and industrial communities. As the basis for its implementation, multiple communication technologies will be employed. However, power line communication (PLC) can present a more extensive and pervasive solution [2]. Up to now, narrow-band (NB) (3–500 kHz) and broad-band (BB) (2–100 MHz) PLC have been developed progressively [3,4]. Especially in the last half decade, multi-carrier NB PLC with a relatively high data rate ranging from 10 to 500 kbps has emerged and aroused immense industry interest as it is considered to be suitable for a part of the Smart Grid applications.
Although multi-carrier NB PLC is sufficient for automatic meter reading (AMR), future Smart Grid services that have rigorous real-time demand may not be supported if the communication bandwidth is restricted below 500 kHz. Moreover, due to high-noise power, low access impedance and marked time variation, a power line channel at low frequencies is characterised as a rather hostile medium for data transmission. As is well known, high-frequency power line channels have much lower noise level; on the other hand, their attenuation gradually increases with frequency (particularly for underground cables). Even so, both low- and high-frequency channels are still indispensable for PLC links within the access domain. Beyond that, note that the medium-wave (MW) band (500 kHz–1.6 MHz) may not only achieve an attractive trade-off between noise and attenuation, but also expand the channel capacity of NB PLC and enhance the coverage of PLC access networks. Therefore, it may be deemed as another potential opportunity for PLC transmission.
As a preliminary example, to show the potential of the MW frequencies for PLC access systems, a simple comparison between the MW band and low-frequency channels below 500 kHz has been made by investigating noise and attenuation characteristics simultaneously, where the corresponding channel measurements have been performed in typical low-voltage (LV) power line access networks in China. For convenience of comparison, the concept of link quality index (LQI), LQI(f,t), has also been introduced and defined as described in Chapter 3:
(23.1) |
where both Noise(f,t) and Loss(f,t) are obtained at the PLC receiver. Thus, LQI(f,t) can be interpreted as the transmitted signal level required to achieve an equivalent 0 dB signal-to-noise ratio (SNR) at the receiver for a specified resolution bandwidth, at a certain frequency f and time t. Evidently, the smaller the LQI(f,t), the lower the transmitted signal level needed to obtain a required SNR at the receiver, which may indicate that the channel is better. Based on the measurement results, Figures 23.1 and 23.2 show a group of LQI curves with a 10 kHz resolution bandwidth for the 30–500 kHz and 500 kHz–1.6 MHz band, respectively, where the units for noise and attenuation are unified as dBμV and dB in this case, and noise is measured by root mean square detection using a spectrum analyser. It can be seen from the statistical results that the MW band has much lower LQI values, which shows the potential of this new frequency range for PLC transmission.
However, note that when the 500 kHz–1.6 MHz band is employed, possible interferences to MW broadcast radio stations may be caused. This is because the unshielded LV power lines are not designed for data transmission at such high frequencies, and their electromagnetic radiation may disturb the surrounding radio services. Table 23.1 shows the radio frequency division regulation of China for the 505 kHz–1.6065 MHz band [5], which confirms broadcast as the main application of this frequency range.
In fact, many existing literature have paid close attention to the radiation effects of PLC. For example, a thorough investigation of the electromagnetic compatibility issues associated with a PLC system was presented based on laboratory experiments [6]. In China, the State Grid has specified the PLC radiation limit in the AMR specifications [7].
FIGURE 23.1
A group of LQI curves with 10 kHz resolution bandwidth for the 30–500 kHz band.
FIGURE 23.2
A group of LQI curves with 10 kHz resolution bandwidth for the 500 kHz–1.6 MHz band.
Radio Frequency Division Regulation of China for the 505 kHz–1.6065 MHz Band
Frequency Range |
Application |
505–526.5 kHz |
Maritime mobile service, aeronautical radio navigation |
526.5–535 kHz |
Broadcast, aeronautical radio navigation |
535 kHz–1.6065 MHz |
Broadcast |
With the aim of tackling the coexistence problem described earlier, cognitive PLC has been adopted as a potential detection and mitigation solution, which introduces the notching schemes to protect the valid radio services by not emitting at their frequencies. This idea was first proposed by Schwager [8] and similar research could also be found in references Weling [9] and Praho et al. [10]. Especially spectrum analyser- and SNR-based detection were presented for cognitive PLC as dynamic notching approaches [11,12]. Furthermore, ETSI TS 102 578 and the recently approved CENELEC FprEN50561-1 (see Chapters 6 and 22) have also specified the detection and notching criteria to achieve a harmonious coexistence between in-home PLC and short-wave (SW) radio broadcast [13,14].
Note that the scope of most of the research detailed earlier is limited to in-home PLC rather than outdoor power line access networks for Smart Grid applications, where the latter indicates a different coexistence scenario. On the one hand, the degree of impact caused by the outdoor PLC access system on indoor radio listeners has not been thoroughly investigated yet. On the other hand, cognitive PLC can be also utilised to protect the PLC access system from external radio interferences. Specifically, in this chapter, cognitive PLC is employed to explore the potential of the MW band for PLC transmission, which has not been studied in detail in the past.
The remainder of this chapter is organised as follows: Section 23.2 describes the test scenario and the measurement setup in detail. Section 23.3 details the analysis of the measurement results achieved in two typical LV power line access networks in China. Then the adaptive detection (AD) method as a cognitive PLC approach is proposed in Section 23.4. Finally, Section 23.5 concludes this chapter.
23.2 Test Scenario and Measurement Setup
Measurements were performed in two typical LV power line access networks in China.
1. Test site 1 is a village located in a suburb in the northeast part of Yiwu city (Zhejiang province) and represents the rural and overhead line environment.
Part of the LV power line access network at test site 1 is shown in Figure 23.3. According to the network topology, two representative measurement points (MPs) were chosen to investigate the ingress caused by the MW broadcast radio stations on power lines, where MP1 represents the LV outlet end of the transformer in the distribution room and MP2 is a single-phase meter panel, which is installed on the outer wall of a two-storey house.
2. Test site 2 is a residential area located in the eastern part of Handan city (Hebei province) and represents the urban and underground cable environment.
FIGURE 23.3
Part of the LV power line access network at test site 1.
FIGURE 23.4
Part of the LV power line access network at test site 2.
Figure 23.4 shows the corresponding LV power grid topology. It can be seen that the underground cables link the transformer with the house access point (HAP) of building A and building B. Cable lengths are approximately 50 and 350 m, respectively. Building A has three units where each has six floors and a meter panel including 12 single-phase meters, whereas building B has only one unit with eight floors and a meter panel that has 16 single-phase meters. The typical distance between the meter panel and the HAP is approximately 5–20 m. Under this test scenario, four MPs are selected for measurement, where MP1 is the LV outlet end of the transformer in the distribution room, MP2 is the HAP of building A, MP3 represents the meter panel of the second unit of building A and MP4 represents the HAP of building B.
The measurement equipment include a passive coupler, an external attenuator, a spectrum analyser and a laptop. In order to verify the measurement results achieved from power lines, a MW radio receiver and a loop antenna are also required. Figure 23.5 shows the measurement setup and dedicated software have been installed on the laptop to collect and process data. Since the spectrum analyser and the laptop are powered by their own batteries, possible interferences introduced by the measurement equipment can be avoided.
The insertion loss of the passive coupler is shown in Figure 23.6, which indicates that the MW band signals ranging from 500 kHz to 1.6 MHz can be passed without significant loss. The unified parameters set for the spectrum analyser during the test are summarised in Table 23.2.
At both test sites, the following procedures were performed sequentially at each MP:
1. The MW radio receiver was used to record broadcast frequencies with audible sound from 500 kHz to 1.6 MHz.
2. For confirmation, the MW radio signals were measured based on the loop antenna.
3. The noises between phase and neutral on the LV power lines were captured by the spectrum analyser, and the results were compared with those obtained in steps (1) and (2) to verify whether the MW radio stations caused any narrow-band interferences on the power lines.
FIGURE 23.5
Spectrum analyser-based measurement setup.
FIGURE 23.6
Insertion loss of the passive coupler for the MW band.
Parameter Setup for the Spectrum Analyser
Resolution Bandwidth [kHz] |
Video Bandwidth [kHz] |
Trace Mode |
Detector |
10 |
10 |
Average 10 times |
Max peak |
23.3 Measurement Results and Analysis
This section includes two parts: In the first part, the measurement results obtained at each test site are shown and analysed. To investigate the MW radio interference to power lines more thoroughly, detailed information about the MW stations around these two test sites have also been provided for reference. Moreover, in the second part, the electromagnetic interference (EMI) issue of PLC is assessed by a simple field trial at MP2 of test site 1.
23.3.1 Spectrum Analyser-Based Measurements on Power Lines
The measurement results obtained at MP1 of test site 1 are shown in Figures 23.7 and 23.8, respectively, where the former corresponds to the results achieved according to test procedures (1) and (2), and the latter focuses on the noise measurement on the power lines. During the test, the MW radio receiver was utilised, and the circles labelled in Figure 23.7 indicate that some broadcast radio stations with audible sound could be detected at those frequencies. It can be seen that 10 MW radio stations were found by the receiver, the frequencies of which coincide very well with those of the signal peaks measured by the loop antenna. In this section, the unit of the vertical axis for the loop antenna measurement results is unified as dBm by using the free space impedance 377 Ω, the antenna factor and the input impedance of the measurement receiver (50 Ω). Since the orientation of the loop antenna also significantly influences the measurement results, the antenna has been rotated in the three-dimensional space to determine the arrival direction of radio signals beforehand. Note that the first two MW radio stations below 600 kHz present the stronger power level above -60 dBm, whereas the signals of the others are relatively weak. In fact, by referring to the local frequency regulation, there are no MW broadcasting stations in Yiwu city. Detailed information about the MW radio stations near Yiwu city are summarised in Table 23.3. It shows part of the 10 audible frequencies determined in the measurement process, whereas the others may correspond to the radio stations located in the neighbouring province, whose detailed information has been omitted here.
FIGURE 23.7
Loop antenna measurement results with 10 kHz resolution bandwidth for the MW band at MP1 of test site 1.
FIGURE 23.8
Power line measurement results with 10 kHz resolution bandwidth for the MW band at MP1 of test site 1.
Detailed Information about the MW Radio Stations near Yiwu City
However, as shown in Figure 23.8, the MW radio stations do not cause any clear ingress on the noise floor of the power lines. Possible explanation is that the power line noise level at MP1 is quite high for the MW band, where the maximum recorded noise power reaches -45 dBm (-85 dBm/Hz). Therefore, it submerges the weak radio signals. Another potential reason is that the electrical grid possibly does not provide loop characteristics according to the access network topology. Due to this antenna effect, the weak MW radio signals may be difficult to be detected on the power lines.
Figures 23.9 and 23.10 further show investigation of the measurements performed at MP2 of test site 1, which represents a single-phase meter panel of a residential house. As mentioned in Section 23.2, MP2 is closer to the power consumer than MP1 and the outdoor overhead line directly links with the meter panel. Note that the difference of the power line noise floor between these two MPs may lead to different radio station detection results. However, similar to the results shown in Figure 23.8, the MW radio stations are still undetectable on the power lines.
FIGURE 23.9
Loop antenna measurement results with 10 kHz resolution bandwidth for the MW band at MP2 of test site 1.
FIGURE 23.10
Power line measurement results with 10 kHz resolution bandwidth for the MW band at MP2 of test site 1.
In order to study the MW radio interferences to the power lines more carefully, another group of measurement results obtained at each MP of test site 2 is shown. The crosses labelled in Figure 23.11 represent the MW radio frequencies according to the local frequency regulation, and detailed information about the MW radio stations in Handan city is presented in Table 23.4. Compared with the power level measured by the loop antenna in Figures 23.7 and 23.9, it can be seen that the signals corresponding to the local MW broadcasts are much stronger in this test site. This is because the radio stations are not located far from the test site. Therefore, the power line narrow-band noises caused by the MW radio interferences become more visible as well. On that basis, it shows that MW radio signals may be detected on power lines in some cases. However, there are also some peaks detected on the power lines, but they do not represent actual MW stations, which may result in false detection.
FIGURE 23.11
Measurement results with 10 kHz resolution bandwidth for the MW band at MP1 of test site 2.
Detailed Information about the MW Radio Stations in Handan City
Figures 23.12 and 23.13 show additional measurement results obtained at MP2 and MP3 of test site 2, respectively. Taking MP2 as an example, due to variation of the radio signal strength, the power level measured by the loop antenna, which corresponds to the 846 kHz radio station, is nearly 20 dB lower compared with the corresponding curves in Figure 23.11. This trend is more obvious at MP3 shown in Figure 23.13. Since the meter panels are installed inside the second unit of building A, the outer wall of the building will prevent the radio signals from being detected by the antenna, which results in even lower power levels. When considering noise measurement on the power lines, part of the local MW radio frequencies can be identified. However, in some cases, certain radio stations may remain undetected on the power lines. Note that the radio frequencies not detected at MP2 and MP3 can be observed at MP1 and vice versa. Therefore, it may become an incentive for implementing a more reliable detection by merging the detection results obtained at different points of the access network.
FIGURE 23.12
Measurement results with 10 kHz resolution bandwidth for the MW band at MP2 of test site 2.
FIGURE 23.13
Measurement results with 10 kHz resolution bandwidth for the MW band at MP3 of test site 2.
FIGURE 23.14
A series of power values measured by the loop antenna with 10 kHz resolution bandwidth for the local MW radio stations at MP4 of test site 2.
As is well known, SW broadcast radio stations may have a very dynamic signal level due to reflections in the ionosphere. Since MW radio signals mainly propagate through the ground wave, the fading of their signal strengths in the time domain may be less obvious compared with SW stations. To verify this characteristic, a series of measurement results obtained at MP4 of test site 2 is shown in Figures 23.14 and 23.15. The measurements were performed every 10 min during the time interval 12:40–14:30 on 27 December 2012, and 12 sets of data were obtained for the loop antenna and the LV power line, respectively. It can be seen in Figure 23.14 that the curves corresponding to the power levels of the six local MW radio stations listed in Table 23.4 are almost flat, except the 1008 kHz frequencies. These results show that the power levels of the MW stations measured by the loop antenna are relatively stable. However, according to the power line measurement results in Figure 23.15, part of the local MW radio stations can be clearly seen across the 2 h, whereas the others may only appear for a certain period of time. It visually illustrates that the time-varying power line noise floor of one measurement location may lead to different detection results for the MW stations.
23.3.2 Field Trial for EMI Issues of PLC
As a preliminary example, this section evaluates the EMI issues of PLC through a simple field trial at MP2 of test site 1. The measurement setup is shown in Figure 23.16, where a signal generator is utilised to inject the swept-frequency signals ranging from 500 kHz to 1.6 MHz into the LV power line and the loop antenna is employed to capture the electromagnetic radiations. During the test, the resolution bandwidth of the spectrum analyser was set at 30 kHz in the Max Hold trace mode and using the Max Peak detector. Through placing the antenna at different distances in a vertical direction from the single-phase meter panel, a group of measurement results corresponding to distinct distances could be recorded.
FIGURE 23.15
A series of power line measurement results with 10 kHz resolution bandwidth for the MW band at MP4 of test site 2.
FIGURE 23.16
Measurement setup for PLC EMI issue assessment.
The measurement results are shown in Figure 23.17. As the radiation source, the top curve shows the power level of the injected signal measured on the power line. In this example, it can be seen that the level of the injected signal, which is between 0 and 7 dBm/30 kHz, is rather strong. The rest curves in Figure 23.17 show the power levels measured with antenna as a function of its distance to the power line. It suggests that the shorter the distance from the antenna to the power line is, the stronger the radiation is. Note that when the distance is 3 m, the radiated signal strength quickly decreases to below -50 dBm according to the measurement result; however, it still submerges most parts of the radio background noise measured by the antenna when the signal generator is closed. Although both electric wire pairs in Figure 23.16, the red–black one as well as the red–blue one, do not look very symmetrical, where high radiation may be expected, this field trial still indicates that potential EMI issues may occur near the power line, which renders certain detection and mitigation solutions indispensable to avoid harmful interferences.
FIGURE 23.17
Measurement results with 30 kHz resolution bandwidth for electromagnetic radiation of PLC at MP2 of test site 1.
23.4 Adaptive Detection for Cognitive PLC
From the measurement results, it shows that MW radio stations can be detected in the power line access networks in some situations. Although the power levels of some detectable MW stations are quite weak, cognitive capability may still be essential for such PLC access systems to prevent themselves from suffering external radio interferences. Note that ETSI TS 102 578 has specified two thresholds with the 9 kHz resolution bandwidth for SW radio detection on in-home power lines, namely, an absolute (-95 dBm) and a relative (14 dB) threshold above the noise floor, where the noise floor of each SW radio band is defined as the median power value of the adjacent frequency blocks [13]. However, measurement results in this chapter show its inapplicability for the MW band in some cases because of two reasons:
1. The power line noise level of the MW band is much higher than that of the SW band. In many situations, the noise powers across the whole MW band are all above -95 dBm with the resolution bandwidth as required, which makes the absolute threshold specified in ETSI TS 102 578 inapplicable.
2. The measurement results show that the narrow-band interferences due to MW stations may not reach 14 dB above the noise floor for most of the time, which will lead to high miss detection rate if this detection criterion is utilised. To be more intuitive, the power line noises recorded at both test sites are utilised for a statistical analysis of the MW radio interference level above the noise floor. For consistency, the noise floor is determined similarly to ETSI TS 102 578 by employing the whole MW band for median calculation. Figure 23.18 shows the corresponding statistical histogram, where the indistinguishable weak narrow-band noises possibly caused by the MW stations have not been included in the statistics. It can be seen that for nearly 80% cases the observable radio interference levels are less than 14 dB above the noise floor. This trend can be seen more clearly in Figure 23.19 by utilising the cumulative distribution function.
Based on the previous analysis, a different detection method for identifying MW stations on power lines should be devised. Note that the method to determine the noise floor by employing the median value is quite simple, however, it may not reflect the real noise floor of the power lines in some cases because the median of noise powers is closely related to the bandwidth used for calculation and the number of radio stations appearing within this sub-band. If the narrow-band interferences caused by the radio signals occupy only less than half of the concerned sub-band and not the other part as power line background noise, the median value can be deemed as the proper noise floor of this sub-band. Otherwise, median value is not a good choice.
FIGURE 23.18
Histogram of the MW radio interference level above the noise floor.
FIGURE 23.19
Distribution function for the MW radio interference level above the noise floor.
In this section, AD has been proposed as a detection approach for cognitive PLC. The main idea of this method is to first determine the appropriate noise floor of power lines by employing the iterative average, and then adding an empirical relative threshold to it to yield the final detection threshold. Moreover, the detected radio frequencies should be notched from the PLC transmission spectrum.
Let F be regarded as a string of numbers that correspond to the power values of different frequencies. Since the power line noise floor over the whole MW band may fluctuate with the frequency, the set F is first divided into N subgroups according to the sub-band division, denoted by F1, F2, …, FN, with N is chosen such that the variation of the noise floor in each subgroup is small. For example, in the case of an orthogonal frequency division multiplexing system, which is employed by most BB PLC solutions, this sub-band division can be naturally realised by dividing the signals according to their sub-carrier frequencies. For the next procedure, the proposed AD algorithm performs the following iterative average steps for each sub-band:
1. Computes the initial threshold εn (n = 1, 2, …, N) as the mean power within sub-band Fn (n = 1, 2, …, N).
2. The power value of each sample in sub-band Fn (n = 1, 2, …, N) is compared with the threshold εn (n = 1, 2, …, N). If it is smaller than the threshold, the index of this sample should be included in set H0.
3. Computes λn (n = 1, 2, …, N) as the mean power of the samples, which corresponds to set H0.
4. If the following termination condition is met, the algorithm stops and λn (n = 1, 2, …, N) represents the final noise floor of this sub-band. Otherwise, it updates εn (n = 1, 2, …, N) with λn (n = 1, 2, …, N) and returns to step (2) for iteration:
(23.2) |
In the previously given equation, τ is a proper stop criterion. Note that in the ideal case, if the noise floor is strictly flat, will tend to be zero with the iterations obviously, namely τ → 0 can be deemed as a proper stop indicator. However, for real power line background noises, τ can be chosen as an empirical value determined by the measurement results. The earlier iterative average process is shown by the flow chart in Figure 23.20.
After determining the noise floor, a relative threshold has to be added to it. Based on the measurement results, 7 dB is chosen as the proper relative threshold for the MW band. Therefore, the final detection threshold can be achieved.
To validate the effectiveness of the proposed algorithm, the measurement results at MP1, MP2 and MP3 of test site 2 are utilised for assessment and the power line noises ranging from 500 kHz to 1.6 MHz are investigated to detect the radio stations. For simplicity, this whole MW band can be divided into four sub-bands by employing the division criteria as mentioned, namely, 0.5–0.75 MHz, 0.75–1 MHz, 1–1.25 MHz and 1.25–1.6 MHz. For each sub-band, the noise floor can be determined individually based on the iterative average process detailed earlier and the 7 dB relative threshold is then added to it.
The detection results for MP1 of test site 2 are shown in Figure 23.21 as an example, where the detection actually equals to performing a binary hypothesis test based on the thresholds determined by the presented algorithm. It can be seen that the local MW radio stations are all successfully identified except one that corresponds to 1098 kHz. There are also three falsely detected stations, which do not represent real MW radio stations. In summary, the detection results of the proposed AD algorithm for each MP of test site 2 are shown in Table 23.5. Note that ETSI TS 102 578 has further specified certain time requirements to minimise misdetection rate, namely, the criteria of -95 dBm and 14 dB above the noise floor should both be met for more than 30% of the time in any 10 s interval [13]. Since the research work has just begun, similar procedures have not been included in the proposed AD algorithm, which is of great interest and essentiality to be investigated in the future.
FIGURE 23.20
Flow chart of the iterative average process in the proposed AD algorithm.
FIGURE 23.21
Detection results for the MW radio stations at MP1 of test site 2 (10 kHz resolution bandwidth).
Detection Results of the Proposed AD Algorithm
Detected MW Radio Stations |
Falsely Detected MW Radio Stations |
|
MP1 |
5 |
3 |
MP2 |
2 |
2 |
MP3 |
5 |
3 |
Furthermore, as discussed in Section 23.3, detection of MW radio stations at one place is not reliable enough. Therefore, cooperative detection in cognitive radio can be introduced by merging the detection results at transformer (MP1), HAP (MP2) and meter panels (MP3) to deliver better performance where the AND, MAJORITY or OR rule may be applied:
1. AND rule: An MW radio station is considered to be present if and only if each MP detects it.
2. MAJORITY rule: An MW radio station is considered to be present if over half of the MPs detect it.
3. OR rule: An MW radio station is considered to be present as long as one MP detects it.
In Table 23.6, the cooperative detection results of AD based on different fusion rules are shown. In this case, it can be seen that although the number of falsely detected ones reaches 7, the OR rule still achieves best performance because all local MW radio stations are correctly detected. For these falsely detected frequencies, they may also be notched by PLC access systems to make communication more reliable. From this point of view, the OR rule may be considered as the preferred candidate for the fusion method of cooperative detection.
Cooperative Detection Results of AD Based on Different Fusion Rules
Detected MW Radio Stations |
Falsely Detected MW Radio Stations |
|
AND rule |
1 |
0 |
MAJORITY rule |
5 |
1 |
OR rule |
6 |
7 |
Finally, as far as design aspects are concerned, since the logical topology of the power line access network often shows master–slave characteristics, the proposed radio detection capability for cognitive PLC may be implemented in the concentrator of the network. Therefore, dynamic notching for protecting valid radio services and keeping PLC away from narrow-band interferences can be controlled by the central node more efficiently.
In this chapter, the coexistence issue between MW broadcast radio stations and PLC access systems that use the frequency band covering 500 kHz–1.6 MHz has been studied. First, spectrum analyser-based measurement results for the MW band are shown, where both overhead line and underground cable environment as typical power line access networks in China have been covered. It shows that the presence of MW broadcast radios in the LV access power line network cannot be easily detected as the SW broadcast radios. One reason can be that the radiation effect of power line network as a potential antenna in the MW band is not as strong as that in the SW band. The high noise floor in the MW band is another factor. Nevertheless, this needs further in-depth investigation. Statistical analysis shows that the MW radio interference level above the power line noise floor is usually less than 14 dB. Therefore, AD for identifying MW stations is presented as a detection approach for cognitive PLC, where an iterative average process is employed to determine the noise floor and a relative threshold above the noise floor smaller (e.g. 7 dB for the examples in this chapter) than 14 dB as specified in ETSI TS 102 578 for the SW radios is proposed to get the absolute detection threshold. Moreover, cooperative detection has also been discussed by merging the detection results to get performance promotion. Since the research work is ongoing, power line noise measurement and analysis on the MW band need further investigation and the interference level between PLC access systems and indoor radio listeners also calls for assessment.
This work was funded by the research project ‘New Generation Smart PLC Key Technology Research’ of the State Grid Corporation of China (SGCC).
1. Lu Y. and Liu W. 2013. Spectrum analyzer based measurement and detection of MW/SW broadcast radios on power lines for cognitive PLC. Proceedings of IEEE Seventeenth International Symposium on Power Line Communications and Its Applications (ISPLC), 2013, pp. 103–108, 24–27 March 2013.
2. Galli S., Scaglione A. and Wang Z. F. 2011. For the grid and through the grid: The role of power line communications in the smart grid. Proceedings of the IEEE 99(6): 998–1027.
3. Latchman H. and Yonge L. 2003. Power line local area networking (Guest Editorial). IEEE Communications Magazine 41(4): 32–33.
4. Pavlidou N., Vinck A. H., Yazdani J. et al. 2003. Power line communications: State of the art and future trends. IEEE Communications Magazine 41(4): 34–40.
5. Ministry of Industry and Information Technology of China. 2010. Radio frequency division regulation of China.
6. Pagani P., Razafferson R., Zeddam A. et al. 2010. Electromagnetic compatibility for power line communications. Proceedings of IEEE 21st International Symposium on Personal Indoor and Mobile Radio Communications (PIMRC), Istanbul, Turkey, pp. 2799–2804.
7. Q/GDW 374.3. 2009. Power user electric energy data acquire system technic specification, part 3: Communication unit. State Grid Corporation of China.
8. Schwager A. 2010. Powerline communications: Significant technologies to become ready for integration. PhD dissertation, University of Duisburg-Essen, Essen, Germany.
9. Weling N. 2011. Expedient permanent PSD reduction table as mitigation method to protect radio services. Proceedings of IEEE 15th International Symposium on Power Line Communications and Its Applications (ISPLC), Udine, Italy, pp. 305–310.
10. Praho B., Tlich M., Pagani P. et al. 2010. Cognitive detection method of radio frequencies on power line networks. Proceedings of IEEE 14th International Symposium on Power Line Communications and Its Applications (ISPLC), Rio de Janeiro, Brazil, pp. 225–230.
11. Weling N. 2011. Feasibility study on detecting short wave radio stations on the powerlines for dynamic PSD reduction as method for cognitive PLC. Proceedings of IEEE 15th International Symposium on Power Line Communications and Its Applications (ISPLC), Udine, Italy, pp. 311–316.
12. Weling N. 2012. SNR-based detection of broadcast radio stations on powerlines as mitigation method toward a cognitive PLC solution. Proceedings of IEEE 16th International Symposium on Power Line Communications and Its Applications (ISPLC), Beijing, China, pp. 52–59.
13. ETSI TS 102 578 v1.2.1. 2008. Powerline telecommunications (PLT); coexistence between PLT modems and short wave radio broadcasting services.
14. CENELEC FprEN 50561-1. 2012. Power line communication apparatus used in low-voltage installations – Radio disturbance characteristics – Limits and methods of measurement – Part 1: Apparatus for in-home use.
* Chapter in parts based on [1], Copyright © 2013 IEEE, with permission.