Electromagnetic Compatibility Problems in LTE Mobile Networks 800 MHz Frequency Band

This paper examines the LTE UE (Long-Term Evolution User Equipment) in 800 MHz frequency band (the so-called digital dividend) interference to Short Range Devices (SRD) in the band of 863 MHz–870 MHz. The fundamental objective of this research is to evaluate the probability of interference and influence of LTE UE emission mask in this regards. The study was performed in two methods: theoretical – simulations with SEAMCAT software and practical – measurements to verify gathered results in simulations. The results of this paper characterize the required technical conditions for LTE mobile networks deployment in 800 MHz band. DOI: http://dx.doi.org/10.5755/j01.eee.21.6.13770


I. INTRODUCTION
The World Radiocommunications Conference in 2007 designated on an essential base the so-called digital dividend frequency band (790 MHz-862 MHz) to land mobile networks in Region 1 countries from 2015.The administrations from Region 1 could exploit this frequency band for land mobile networks before 2015, but only with the appliance to the Radio Regulations provisions [1].
Region 1 countries in 2009 decided to establish a decision which would guide the involved countries on technical and regulatory framework in the band 790 MHz-862 MHz in order to meet the growing shortage of radio frequency in mobile networks field [1], [2].
Mobile operators have the resources to expand their networks, but it becomes increasingly difficult to avoid interference between adjacent radio stations due to ever growing density of wireless apparatus [3], [4].
This study was conducted to help clarify the situation in the newly forming LTE cellular network called as digital dividend.The LTE frequency band (790 MHz-862 MHz) is quite close to the Short Range Devices band (863 MHz-870 MHz) [5].SRD can be called such facilities as: cordless headphone, intruder alarm, radio microphone, smart meter, telemetry, medical device, RFID (Radio-Frequency Identification) and etc.This paper investigates a possible situation in which LTE device is in the same room with the SRD unit.So the distance between LTE and SRD could be some meters.
Without this type of analysis LTE mobile networks could First of all studies were carried out using SEAMCAT software [7].SEAMCAT enables statistical simulation of different radio network scenarios in order to perform adjacent or co-channel electromagnetic compatibility cases.Then all simulations repeated practically using LTE signal generator and real SRD devices.

II. LTE 800 MHZ BAND SPECIFICATION
The official CEPT document -Report 31 terminates that the prefigured frequency plan for the frequency band 790 MHz-862 MHz should be applied on the FDD (Frequency Division Duplex) mode.The frequency arrangement is stated in [8]: 791 MHz-821 MHz band as downlink, 832 MHz-862 MHz band as uplink, 821 MHz-832 MHz band as duplex gap and 790 MHz-791 MHz band as guard band from DVB-T.
The LTE mobile networks in 800 MHz frequency band can have influence to DVB-T services in the lower block of the LTE frequency band and with SRD in the upper side of the LTE frequency band.The frequency separation is relatively short, so there is possibility to identify electromagnetic compatibility problems.MHz channel (852 MHz-862 MHz -most upper channel), the maximum available resource blocks (leads to maximum emission power) -50, bitrate -20 Mbit/s.Table I shows the OOB (Out-of-Band) requirements in unwanted domain corresponding to 3GPP TS 36.101recommendation (ETSI reference number TS 136 101) [9].It would need to follow the principle that the resolution bandwidth (RBW) of the measuring device should be similar to the measurement bandwidth (MBW).In order maintain an acceptable level of measurement sensitivity and efficiency, the RBW should be narrower than the MBW.

III. SRD MAIN INFORMATION
The 863 MHz-870 MHz frequency band are available for SRD applications within sub-bands as already introduced in ERC Recommendation 70-03 [5].The Short Range Devices technical parameters are described in the list below according to Annexes of ERC Rec.70-03:  Annex 1: Non-specific Short Range Devices;  Annex 7: Alarms;  Annex 10: Wireless microphones;  Annex 13: Wireless Audio Applications.
The following Table II shows the parameters which were used in the SEAMCAT simulations.SRD were considered as a victims receivers.In measurements only Non-specific SRD were used to verify results obtained in SEAMCAT simulations.

A. Interference Scenario I
The following Fig. 1 shows the test plan used for simulations with SEAMCAT.

B. Results of Simulations
Table III shows the results of simulations as separation distance is up to 10 m.The obtained results show the noticeable level of probability of interference.The systems are electromagnetically compatible when the probability of interference is less than 5 % [7].Note that the interference criterion was used C/(I + N) = 8 dB or 17 dB for Annex 10 (as derived from the EN 300 220 [11], EN 301 357 [12] standards).

V. PERFORMED MEASUREMENTS
All measurements were conducted with certified equipment to replicate the interference scenarios simulated in theoretical analysis.

A. Interference Scenario
Semtech International AG SX1231 transceiver development kit was applied for the measurements to adopt real SRD.The main parameters of Semtech SX1231 device [13] are shown in Table IV.First of all the additional measurement was carried out in order to check whether the officially stated SRD sensitivity level corresponds to the actual running system parameters.The main information from datasheet is presented in Table V.
During the control measurement SRD Tx radiated power was set at -18 dBm, bandwidth -100 kHz, bitrate -57.5 kbit/s.SRD Tx and Rx were then gradually removed from each other, monitoring incoming packages.During the experiment we observed that the connection was lost when the receive signal level dropped to an average of -97 dBm.
Based on the established value of SRD sensitivity for given bandwidth, for the subsequent measurements of interference in the 2nd part of this study, the victim SRD receiver was set to operate at receive level 3 dB above the sensitivity level.

B. LTE User Equipment Interferer
Generation of LTE UE signal for this study was accomplished with the multi-purpose SDR development platform Ettus Research USRP2 device with WBX 50 MHz-2200 MHz Rx/Tx daughterboard.This device is connected to the PC by Ethernet cable, so that the parameters of the generated signal can be adjusted on the computer.Main RF parameters of Ettus WBX 50 MHz-2200 MHz Rx/Tx daughterboard.With this device we simulated LTE UE operation, by setting the respective power level of interfering signal falling within the victim SRD channel to correspond to the LTE UE OOB emissions mask.

VI. INTERFERENCE SCENARIO II
The measurement set up used two SRD devices (see Fig. 2): transmitter and receiver.The SRD equipment was placed and configured so that victim receiver operated with useful received signal level of -94 dBm, corresponding to 3 dB margin over sensitivity level.Then the LTE UE interferer was switched on and brought gradually closer to victim SRD Rx, while constantly monitoring transmission of packets on victim SRD link.The stability of SRD link was monitored by performing cyclical count of proportion of packets received and decoded at the SRD Rx.
During the measurement, LTE UE interfering power did not exceed the OOB limits set in 3GPP TS 136 101.For this, the proxy LTU UE interferer was configured to radiate the equivalent 1 MHz channel that spans the SRD Rx channel and is of power level directly corresponding to the 3GPP TS 136 101 LTE UE OOB mask limit for given frequency separation between the interferer and victim.The following Fig. 3 shows the LTE UE spectrum emission mask overlaid on the SRD device channel.

VII. INTERFERENCE SCENARIO II RESULTS
During the first measurement the proxy LTE UE interferer was brought gradually toward the SRD Rx, while monitoring the packet flow on victim link and trying to establish possible interruption.The following Fig. 4 shows the results obtained.The test revealed that when the separation distance between the LTE User Equipment as interferer and SRD Rx as victim is less than ~ 2.5 m, the interference severs the link between the SRD Tx and SRD Rx.
In the second measurement we calculated the percentage of packets being received at distances greater than 2.5 m.During each measurement we sent a batch of 1000 packets and monitored how many of them were successfully received and decoded in SRD Rx.Semtech SX1231 transceiver may use two modes for coding data packages (FEC) -Manchester and Whitening.
The analysis of SRD parameters indicates that usually

Figure 3
Figure 3 shows the LTE emission mask based on TS 136 101 (100 kHz RBW) with SRD frequency 868.1 MHz.It demonstrates that LTE UE power shall not exceed -23 dBm/100 kHz.The signal was developed to simulate specifically this part of LTE UE out of band emission.

TABLE III .
INTERFERENCE PROBABILITY.

TABLE V .
SRD SENSITIVITY LEVELS FROM DATASHEET.

TABLE VI .
MAIN LTE SIGNAL GENERATOR PARAMETERS.WBX