LTE Technology: Antenna, RF Front-Ends, and Channel ModelingView this Special Issue
The Communication Solution for LTE System under Radar Interference Circumstance
The interferences of radar and communication system coexistent environment are analyzed in this paper, and an attractive detection mechanism of the radar video signal and proportional fair scheduling algorithm based on channel sensing are proposed to realize the communication by using radar radiant intermissions. And the simulation is designed based on the proposed model to test this plan. The simulated results show that the packet scheduling algorithm is an excellent candidate for the communication solution for LTE system under radar radiant intermissions.
The Third Generation Partnership Project (3GPP) has launched Long Term Evolution (LTE) of next generation mobile communications on the frequency bands of 1710–1885 MHz, 2300–2400 MHz, and 2500–2690 MHz based on the number 223 resolution of the World Radiocommunication Conference in 2007, as discussed elsewhere [1, 2]. The development of 3GPP LTE has witnessed explosive growth driven by the wireless communication system. In addition, version R9 of LTE has been released in March 2010, which forms executive standard of LTE. At present, the Chinese Mobile Communication Corporation (CMCC) has developed the Time Division Long Term Evolution (TD-LTE) system, which has been tested and utilized in many cities across the whole country.
However, the shortage of spectrum resources becomes an increasingly serious problem for LTE system, especially in high-frequency bands, that is, 2500–2690 MHz. The LTE system with working frequency band of 2500–2690 MHz has interference problems with high power radars in the adjacent frequency band, such as the S-waveband weather radars and air traffic-control radars. When the frequency space between radar and LTE system is not large enough, the radar signal power is so high that the band-pass filter could not filter out the radar dominant frequency signal to an acceptable degree, in the radar pulse signal duration. And the residuals signal with high amplitude will still interfere with the communication signals. Meanwhile, the radar out-of-band spurious signal falls into filter’s pass-band, and it will not be attenuated by the filter, thus causing interference. When the radar works at the same frequency band of LTE system, the radar dominant frequency signals could easily burn through the receiver.
The interferences mentioned above will affect the performance of LTE system and restrict its promotion, as well as the further utilization of these new assigned frequency bands. Therefore, this paper applies the principle of cognitive radio (CR) to study the interference between radar and LTE system, as discussed elsewhere [3–5], and it comes up with a precise detection mechanism of radar video signal to realize spectrum sensing, as well as a proportional fair scheduling algorithm based on channel sensing for spectrum allocation. In the past papers, researchers usually suggested a protection distance between radar and communication system to ensure a “clean” electromagnetism reserve. In this paper, when LTE system is not so far away from radar, LTE system can establish communication in radar radiant intermittent periods, so they can work compatibly with slightly lower data rate and approximate block error rate.
2. Materials and Methods
2.1.1. Basic Characteristics of Target System
When radar and LTE cells are located as in Figure 1, the main beam and near zone minor lobe of radar will interfere with the receivers of LTE system in varying degree. There are two radar intermissions that can be utilized: the first one is the pulse intermission when radar beam passes over the cell B area and the second one is scanning intermission when radar beam scans out of cell B area.
By requiring from meteorological department, main technical parameters of weather radar can be shown in Table 1, while the parameters of LTE base stations (eNodeBs) and user equipment (UE) can be listed in Tables 2, 3, by looking up related literatures and actual measurements, as discussed in . These parameters can be used to compute the strength of signal, in order to judge interference condition, or calculate intermissions for communication, or be used for simulation.
2.1.2. Interference Analysis on Radar and LTE Coexistent Environment
According to Friis transmission formula, the receiving power can be calculated by the following formula, which reflects interference signal strength, and it helps in judging whether a certain system is interfered or not:For evaluating whether a LTE system, such as eNodeBs or UE, is interfered with weather radar, it needs to calculate the radar antenna power density in eNodeBs at first, that is, , where subscript represents radar system, is the antenna power, is antenna gain, means antenna minor lobe, and represents distance between radar and LTE system. Then, the signal power received by the LTE system can be computed by , where subscript means LTE system, is the antenna gain, is the antenna minor lobe, and λ means radar wavelength. In addition, it needs considering of different transmission losses , where and are antenna feeder losses, represents maximum multipath signal of radar, and means polarization mismatching loss of the coexistent system. Finally, it needs computing the different spurious suppressions , where is radar spurious suppression at the frequency of LTE system and is the out-of-band and filter suppression of LTE system. In this way, the power of radar interference signal received by LTE system can be calculated, and the interference conditions can be defined by comparing with receiver sensitivity.
For instance, when the LTE system is working at 2670 MHz, weather radar works at 2720 MHz, and Tables 4, 5 list other related parameters. The interference power received by LTE system can be obtained in decibel bywhich is higher than the receiver sensitivity of eNodeB with value of −117 dBm; it can foresee that the eNodeB is interfered by the main beam of weather radar within 300 meters away.
When considering other effects including near or far zone minor lobe, multipath echo signal, and different distances, it can be concluded that when the weather radar is within 30 to 2500 meters away from the LTE base station, the LTE system is interfered by radar, while interferences from LTE system to radar can be ignored.
The Communication Solution in the Coexistent Systems. After analyzing interference condition, the radar pulse intermissions need to be found for communication; therefore, a precise detection mechanism of radar video signal is designed to realize the spectrum sensing, and a channel-sensing-based LTE packet scheduling algorithm is proposed to approach spectrum access.
2.1.3. Sensing of Radar Radiant Environment
The technical parameters of weather radar remain stationary and simple, which also can be detected previously, and it is required for a cognitive system (LTE system) to detect the authorized users (weather radars) precisely in real time. Therefore, this paper proposes an attractive detection mechanism of the radar video signal based on the matched filter in frequency domain, which can be divided into three parts as shown in Figure 2.
(i) RF Pilot Frequency Matching Channel. The first part is the RF pilot frequency matching channel. This channel is imported with the high-frequency signal of the weather radar, and the signal passes through the band-pass filter, RF amplifier, mixer, and analog to digital conversion in turn, to be transformed to digital IF signal.
(ii) Matching Detection in Frequency Domain. The second part is the matching detection in frequency domain. The matching detection part is made up of computation modules, such as FFT, square, and threshold comparison.
This part uses spectral transform and computation to detect the radar signal and obtain the frequency and energy of the radar signal, as well as the raw detection of pulse repetition period. Because the energy of spectrum with central frequency and bandwidth iswhere is the FFT length, represents digital radar signal, means sample rate, and and are the upper and lower bounds of frequency points. After comparing with (3) and detection threshold , which is detected in the engineering environment, the proposed mechanism can ascertain whether there is authorized signal received in cognitive system receiver by formula (4) and detect the frequency and energy of the authorized signal and make detection on pulse repetition period
(iii) Precise Video Signal Detection. The final part is the precise video signal detection. In this part, the processed radar signal passes through envelope detection, amplifier, threshold detection, and computation modules in turn. As long as the matching detection part detects radar signal, the mechanism can immediately obtain the measurement of pulse start time , pulse end time , pulse width , and pulse amplitude . These parameters are used to ascertain which LTE scheduling period is interfered.
And then, the mechanism can acquire the repetition period of pulse signal by and the antenna scanning period , which is the intermission between the two maximum values of pulse amplitudes measured by statistics. Finally, the scanning intermission is got as .
This sensing mechanism starts sensing when the LTE eNodeB begins working. After the sensing mechanism has detected and calculated radar pulse start time , the scheduler of eNodeB system will use the detected parameters for scheduling. Although radar signal changes within a period length of several microseconds (μs), the radar radiant sensing scheme should operate precisely to catch changes, which may lead to high time consumption. However, the weather radar does not change its working parameters frequently. Therefore, when sensing scheme has detected and known all parameters mentioned above, the eNodeB system can use these “fixed” parameters for a long period, until the sensing scheme finds the changes in parameters. In this long period, the sensing scheme focuses more on validating interfered periods for eNodeB’s scheduler comparing with the simple precise detection. Hence, the scheduler uses predetected parameters to compute which transmission time interval (TTI) is interfered and skips those TTIs, while sensing scheme detects and tells the latest pulse start time to the eNodeB system, to check whether it skips interference in right TTIs. In this way, tiny time consumption is permitted, and sensing scheme can tell pulse start time to the eNodeB system immediately for validation, which is before completing all other parameters to find out changes and do regularly precise detection at set intervals, for example, half an hour.
2.1.4. LTE Packet Scheduling Algorithm for Spectrum Access
Orthogonal frequency division multiple access (OFDMA) system is the downlink multiple access technology of LTE system, and it uses packet scheduling mechanism to allocate subchannels, which can be seen as resource blocks (RBs), to different users. There are three common packet scheduling algorithms, as discussed elsewhere , and proportional fair (PF) scheduling algorithm is chosen in this work, because it makes a compromise between system throughput and user fairness.
(i) Proportional Fair Scheduling Algorithm. PF scheduling algorithm means every user can obtain equal scheduling opportunities, as discussed elsewhere [8, 9]. In each scheduling period, each user will be assigned a corresponding priority by user’s instantaneous speed divided by average speed , and each subchannel will choose the user with highest priority to transmit data. Hence, if there are users who require for transmitting data at scheduling period , the chosen user will be After one scheduling period, the average rates of users will be updated as where is a time window parameter.
(ii) Proportional Fair Scheduling Algorithm Based on Channel Sensing. Because working frequency of the LTE system is near radar’s operating frequency band, when the radar pulse signal occurs in a certain scheduling period, that is, transmission time interval (TTI), all subchannels in the system will be interfered with radar radiant pulse signal. At this time, all channel qualities are not good enough for users to communicate. Hence, this project adopts channel sensing factor to proportional fair scheduling algorithm and makes the channel qualities of all TTIs interfered with pulse signals zero. And then, the scheduler will not assign that channel to that user. Because all channels are interfered, the scheduler will stop working for the TTI and recalculate users’ priority for scheduling in the next TTI. Then PF algorithm will be revised as where where is the start time of the scheduling period when radar pulse signal occurs for the first time and is the start time of the first detected pulse signal when radar antenna scanning into sector ; then is used to find when radar pulse occurs in the TTI. is the pulse repetition period, represents the pulse number detected by eNodeB within one antenna scan cycle, is the antenna scan cycle, while is the antenna scan angle over sector , and is doing decimals to round down numbers. Therefore, represents pulse repetition time, and is used to compute the TTI number when radar pulses occur. Besides, represents the antenna scan times, and is computing the time of different antenna scan cycles. In addition, the unit of , is microsecond (μs), the unit of is second, and the duration of one TTI is 1 ms, which is also the length of scheduling period.
3. Results and Discussion
3.1. System Level Simulation Platform of Radar and LTE Coexistent System
For the system level simulation platform, it mainly concentrates on problems of scheduling and interference management. The platform is set up by Matlab with ten modules, which can be seen from Figure 3. The main body of the simulation platform includes the simulation process control module, the configuration module, the radar pulse signal interference module, the network layout module, and the display interface module, while the transmission environment module, the antenna module, and the user module are the submodules of the network layout module. In addition, the link measurement and self-adaptive strategy module provides channel conditions for the resources scheduling module, which belongs to the user module. The simulation mainly concentrates on three modules: the first one is the radar pulse signal interference module to produce interference, the second one is the user module for detecting interference, and the third one is the resources scheduling module to allocate RBs.
Data rate and block error ratio (BLER) are used as evaluation metrics for this simulation platform. Data rate is the transmitted data flow per second, while BLER is the ratio of the number of received error blocks and that of total received blocks.
And the simulation parameters configuration can be seen in Table 6, where some parameters are chosen as discussed elsewhere [10–13]. The users are dropped randomly and immovably in the entire LTE cell, while the traffic model is Full-Buffer, thus simplifying the simulation model.
3.2. Simulation Results and Analysis
3.2.1. Channel Allocation
When 10 pieces of UE are assigned to each sector randomly, then RBs allocation can be seen from Figure 4. When the number of RBs assigned to each user decreases to 0, there are special blue broken lines occupying all 25 RBs, which occur in every three or four TTIs. It is a virtual UE piece 0, which represents the scheduler of eNodeB. The RBs assigned to UE piece 0 represent that these RBs are interfered by radar pulse signal at that TTI, and they are not assigned to any real user due to their poor channel quality but reserved by the scheduler itself.
(a) Radar pulse signal starts at TTI 23
(b) RBs allocation in TTI 80 to TTI 100
(c) Radar pulse signal ends at TTI 159
When the first pulse signal starts at 23.211 ms, that is, TTI 23, if the scan period of radar antenna is 1 s, the scan angle of the radar antenna over the cell area is 5 degrees; then all pulse start times and their corresponding TTIs can be calculated by formula (9), and the results are in accordance with the time when UE piece 0 occurs as shown in Figure 4. Hence, this simulation platform can certainly keep away from the interfered TTIs and then communicate in other TTIs by using precisely detected parameters of radar pulse signal when radar pulse signals occur.
3.2.2. Data Rate
By allocating a different number of users in each sector, such as 5, 10, 15, 20, 25, 30 pieces of UE, this platform can obtain the data rate of the entire cell with or without radar interference, as shown in Figure 5.
When 5 users are allocated in each sector, the average data rate of interfered cell is 10.7865 Mbps, while that of cell without interference is 11.023 Mbps. Hence, the attenuation of average data rate is 2.1455%. The attenuation trend increases with the increase of UE pieces number and is stable to 7%, which is within the acceptable range, as shown in Figure 6. Hence, the simulation results show that the proposed communication scheme for the radar and LTE coexistent system is effective.
In addition, the above data is mainly obtained when radar antenna beam scans over cell B area, as shown in Figure 3, which occupies only a small part of the whole radar antenna scan period. After considering the periods when radar antenna scans over the other directions, the data rate of cell B is not interfered. Therefore, when comparing a normal LTE system (which is not adjacent to a radar), with a LTE system that is affected by radar and that used the presented solution, the data rate of the first and the second case should be the same. And then, the data rate attenuation ratio should decrease a lot in the whole radar antenna scan period.
By allocating a different number of users in each sector, such as 5 to 30 pieces of UE, the BLER variations can be seen from Figure 7.
When 5 pieces of UE are allocated in each sector, BLER of interfered cell is 0.0467, while that of cell without interference is 0.0482. Hence, BLER gain is 3.3195%. The trend of average BLER gain decreases with increase of UE pieces number, and finally it is stable within 1% range comparing with BLER value of cell without interference.
In general, the BLER of interfered cell is floating slightly around that of cell without interference. Hence, by sensing of radar pulse signal, this communication scheme can make the LTE system keep away from TTIs when radar pulse signals occur. And then, the LTE system can communicate at other TTIs, which will decrease radar influences on BLER.
This paper applies the principle of cognitive radio to interdisciplinary knowledge of radar and LTE system. And then, a communication scheme is proposed, which utilizes radar radiant intermittent periods to realize communication in radar and communication coexistent system. The proposed scheme has been analyzed and the analysis proves that it can keep away from interference of radar pulse signal and communicate in radar pulse intermission according to precisely detected signal parameters. Simulation results show that the data rate attenuation is stable to 7%, while BLER gain is stable within 1%, which proves the effectiveness of the communication scheme.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
In this research, my knowledge of radar and LTE system has improved greatly, and I have learnt many useful learning methods and thinking perspectives to solve problems. All thanks are due to our tutors, Yunjie Li and Zhenglei Huang, and other friends who helped us a lot and encouraged us when we had problems.
S. Sesia, I. Toufik, and M. Baker, LTE—The UMTS Long Term Evolution: From Theory to Practice, Including Release 10 for LTE-Advanced, John Wiley & Sons, New York, NY, USA, 2nd edition, 2011.
“IEEE standard for safety levels with respect to human exposure to radio frequency electromagnetic fields, 3 kHz to 300 GHz,” IEEE Standard C95.1, 2005.View at: Google Scholar
J. Mitola, Cognitive radio: an integrated agent architecture for software defined radios [Ph.D dissertation], Royal Institute of Technology (KTH), Stockholm, Sweden, 2000.
3GPP TS 36.814 V9.0.0. Evolved Universal Terrestrial Radio Access (E-UTRA); Further advancements for E-UTRA Physical layers aspects, Dec. 2009.
C. Suh, S. Park, and Y. Cho, “Efficient algorithm for proportional fairness scheduling in multicast OFDM systems,” in Proceedings of the IEEE 61st Vehicular Technology Conference (VTC '05), vol. 3, pp. 1880–1884, Stockholm, Sweden, June 2005.View at: Google Scholar
H. Kim, K. Kim, Y. Han, and S. Yun, “A proportional fair scheduling for multicarrier transmission systems,” in Proceedings of the IEEE 60th Vehicular Technology Conference (VTC-Fall '04), vol. 1, pp. 409–413, September 2004.View at: Google Scholar
3GPP TR 36.942 V8.1.0, “Technical Specification Group Radio Access Network, Evolved Universal Terrestrial Radio Access (E-UTRA), Radio Frequency (RF) system scenarios,” September 2009.View at: Google Scholar
N. Arsalane, M. Mouhamadou, C. Decroze, D. Carsenat, M. A. Garcia-Fernandez, and T. Monedière, “3GPP channel model emulation with analysis of MIMO-LTE performances in reverberation chamber,” International Journal of Antennas and Propagation, vol. 2012, Article ID 239420, 8 pages, 2012.View at: Publisher Site | Google Scholar